About Electronic Publication of this Book

Contents

List of Figures

List of Tables

Preface

This book discusses user mobility in a wide area network (WAN) environment. In this discussion, a mobile data device is one which can receive WAN services from essentially any location without requiring any special actions by the user of the device. User mobility is described in the context of the Cellular Digital Packet Data (CDPD) standard, developed by ourselves and others on behalf of the North American cellular industry.

Two trends provide a backdrop for this subject matter. The first of these is the rapid growth of the Internet¹ . Both in terms of numbers of users and traffic, this growth has been nothing short of phenomenal. What was once strictly the domain of highly technically-literate people has now become headline news. Censorship of the World-Wide Web is now discussed by politicians and URLs are commonly displayed in advertisements.

The popularity of the Internet reflects a change in the media of choice for people wishing to communicate. Email allows the thoughtfulness of a letter while providing the potential immediacy of the telephone. Complex ideas can be conveyed in an organized manner, then further developed by the receiving party. Several rounds of a discussion can take place in a matter of minutes, quickly resolving issues that might be difficult to present orally. The CDPD specification was itself rapidly developed by remote parties largely via Email discussion.

The second trend is that of mobile communications. The cellular industry is experiencing explosive growth, with over 32 million subscribers in North America at year-end 1995. The paging industry has also experienced rapid growth; the advent of new two-way messaging services is likely to extend that growth in the face of competition from low cost (to the subscriber) mobile cellular handsets.

The next step in this evolution of communications is that of mobile data communications. Mobile data is expected to grow from 200 thousand subscribers in 1990 and 1.1 million in mid1995, to 5.2 million subscribers in 2000.² Several technology developments aimed at mobile data communications are in various stages of progress or completion.

The Mobile IP Task Force of the IETF has been addressing the requirements for mobility in data communications. Their charter is to define the protocols necessary for a correspondent to send and receive data anywhere. The media to be used is unspecified. Presumably one will in the future be able to find an Internet ”socket” in the wall of a hotel room as readily as they currently find electrical outlets. However, many ”real world” considerations such as usage accounting remain unaddressed by this group.

The Ram and ARDIS mobile data services, supported by RadioMail, provide gateway connections between proprietary radio technology and the Internet or other wide-area networks. However, the need to port applications to nonstandard proprietary mobile devices and APIs limits the generality and user adoption of these service offerings.

Rather than using gateways between proprietary radio technology and the Internet, CDPD defines an open standard which allows mobile devices to be as directly accessible as any other IP host. Standard APIs allow the immediate use of current data applications such as Email on mobile devices. We have done this many times.

This book is intended to complement the CDPD specification but not replace it. Our emphasis is on the data networking aspects of CDPD and its solution to the mobility problem. CDPD is a data network which happens to have an RF-based data link resembling Ethernet (but at a lower speed). The fixed end of the radio link, called the Mobile Data Base Station or MDBS, is little more than a LAN hub from a data networking perspective.

This book is clearly focused on CDPD as the preeminent wide area mobile data solution. We don’t apologize for our bias–we were highly involved in the creation of CDPD. However, no system or technology lasts forever; one of our design goal was that CDPD be readily amenable to evolution. CDPD is much more than an airlink–it is an architecture that supports host mobility over a wide area.

The chapters which follow describe the CDPD solution to the challenge of mobility in wide-area networks. A discussion of mobility (of which wirelessness is a special case) is followed by a summary of cellular technology, an overview of CDPD, a description of CDPD architecture and how it supports mobility, a description of security and other support services provided by CDPD (and needed by any public mobile data network!), a survey of other (noncellular) mobile systems and finally, a discussion of future directions in mobility in the wide-area environment. For readers unfamiliar with data networking concepts, a primer on this subject precedes the first chapter.

The target audience for this book is any individual interested in mobile data communications or, more specifically, the rationale behind the design of CDPD. The discussion of technical issues avoids the jargon and abstractions necessary and typical in technical specifications. Because we are not radio engineers, we focus on the system and networking aspects of CDPD rather than the radio technologies, which are better described elsewhere. Our goal is to explain mobility and CDPD in plain English. Please let us know whether we succeeded at mark.taylor@airdata.com, wwaung@direct.ca and mohsen@neda.com.

Bellevue, Washington

June 11, 1996

Preliminaries

This chapter introduces some of the standard data networking terminology and concepts used throughout this book. Those readers already familiar with data networking technology could begin with Chapter 1, which is the real first chapter, with no loss of continuity.

Familiarity with the concepts presented in this chapter is important to understanding the issues of mobility and is assumed in the chapters that follow. Topics discussed in this overview include the communications channel, protocols, connection-oriented and connectionless protocols, the OSI reference model and it layers, protocol data units and networking entities.

This chapter is presented as a survey and is no substitute for the real thing–it is necessarily brief. Many fine texts, such as [STAL93], [PERL92] and [TANN95], are devoted to teaching data networking and cover this subject much more rigorously. Of course, true expertise comes only with study of actual standards documents.

Basic Data Communication Model

Communication is the conveyance of a message from one entity, called the source ortransmitter, to another, called the destination or receiver, via a channel³ of some sort. A simple example of such a communication system is conversation; people commonly exchange verbal messages, with the channel consisting of waves of compressed air molecules at frequencies which are audible to the human ear.⁴ This is depicted in Figure 1.

The conveyance of a message could be followed by a reciprocal response message from the original destination (now a source) to the original source (now a destination) to complete one cycle in a dialogue between corresponding entities. Depending on the application or need for the information exchange, either atomic one-way transactions or a two-way dialogue could be appropriate.

The only way that a message source can be certain that the destination properly received the message is by some kind of acknowledgment response from the destination. Conversing people might say ”I understand” or nod their head in response to a statement made by their peer. This acknowledged form of dialogue is the basis of reliable communications–somehow the source must get feedback that the destination correctly received the message.⁵

Variations on a Theme

The conveyance of a message could be direct between the corresponding entities or it could be indirect, with one or more intermediaries participating in the message transport. The presence or absence of an intermediary depends on the definition of the source and destination entities and the channel used to communicate; data communication between entities at one level might be considered to be direct and at another level to be indirect.

Considering the directness or indirectness of the communicating entities simply depends on the relevance of any intermediaries to the discussion. In Figure 2, the translator is important but should not really be a factor in the communication between the source and destination. Perhaps in a twist on the old parents’ saying, a good translator is heard but not seen.

Communication can be from a source to a single destination, known as point-to-point or unicast, or to multiple destinations, known as point-to-multipoint or multicast. A special case of multicast is the conveyance of a message from a source to every possible destination, which is referred to as broadcast; the broadcast can be local or global in scope.

The primary difference between multicast and broadcast is that multicast communication is targeted at specific destinations, regardless of location, while broadcast communication is targeted at all possible destinations within the range (location) of the source. Multicast and broadcast communications are typically one-way ”best efforts” modes of communication which are unacknowledged.⁶

Communication can also be described in terms of the relative timeframes of the corresponding entities. Depending on the definitions of source, destination and channel, the communication could be asynchronous, synchronous or isochronous.

In asynchronous communication, there is a minimal assumed timing relationship between the source and destination. In such a typically byte-oriented system, each character or byte is transmitted and received individually as a message. Asynchronous protocols were predominant in the early days of data communications because of limited processing capability and low quality transmission infrastructure.

In synchronous communication, the relative bit timing of the source and destination is similar, allowing transmission and reception of relatively large groups of bits in a single message; the source and destination must be ”in sync.” This bit-oriented mode of communication can be much more efficient than asynchronous communications, but places requirements on the source (processing), channel (quality) and destination (more processing). Synchronous data communications are predominant today.

Isochronous communication is the extreme case of synchronous communication–source and destination are ”in sync” in the absolute sense of real time, allowing continual transmission of bits. An everyday example of isochronous communication is a telephone conversation; if such a conversation occurs across a large distance (such as trans-Atlantic), the delays introduced can be disconcerting because the isochronicity which people are accustomed to has been negatively impacted. Effective isochronous communication depends on both transmission delays which are inconsequential to the corresponding entities and a consistent high quality transmission.

The Communications Channel

A communication channel can be simplex, in which only one party can transmit, full-duplex, in which both correspondents can transmit and receive simultaneously, or half-duplex, in which the correspondents alternate between transmitting and receiving states (such as conversing adults). Even though the channel might be capable of supporting full-duplex communication, if the corresponding entities are not capable of transmitting and receiving simultaneously, the communications system will be half-duplex (as in the example of the conversing adults).

Communication between two entities can be considered either in-band or out-of-band, depending on context. In-band communication is communication which occurs via the primary channel between the communicating entities. Out-of-band communication occurs via an alternative channel, which is not considered to be the primary channel between the entities.

Which channel is primary and which is an alternate depends on context and the existence of an alternative channel. In the case of a conversation between two people, the primary channel could consist of verbal communication while the alternate channel consists of visual body language. Of course, if emotions rise, these two channels might reverse roles, with body language becoming the primary channel!

Channel Characteristics

A communications channel may be described in terms of its characteristic properties. These channel characteristics includebandwidth (how much information can be conveyed across the channel in a unit of time, commonly expressed in bits per second or bps⁷ ), quality (how reliably can the information be correctly conveyed across the channel, commonly in terms of bit error rate or BER⁸ ) and whether the channel is dedicated (to a single source) or shared (by multiple sources).

Obviously a higher bandwidth is usually a good thing in a channel because it allows more information to be conveyed per unit of time. High bandwidths mean that more users can share the channel, depending on their means of accessing it. High bandwidths also allow more demanding applications (such as graphics) to be supported for each user of the channel.

The capability of a channel to be shared depends of course on the medium used. A shared channel could be likened to a school classroom, where multiple students might attempt to simultaneously catch the teacher’s attention by raising their hand; the teacher must then arbitrate between these conflicting requests, allowing only one student to speak at a time.

Reliability of communication is obviously important. A low quality channel is prone to distorting the messages it conveys; a high quality channel preserves the integrity of the messages it conveys. Depending on the quality of the channel in use between communicating entities, the probability of the destination correctly receiving the message from the source might be either very high or very low. If the message is received incorrectly it needs to be retransmitted.

If the probability of receiving a message correctly across a channel is too low, the system (source, channel, message, destination) must include mechanisms which overcome the errors introduced by the low quality channel. Otherwise no useful communication is possible over that channel. These mechanisms are embodied in the communication protocols employed by the corresponding entities.

The effective bandwidth describes what an application experiences and depends on the quality of service (QOS) provided by the channel. For example, modems scale back their transmission speed based largely on their perception of channel quality in order to optimally use the transmission medium.

In general, shared and reliable channels are more resource efficient than those which enjoy neither of these characteristics. Shared channels enjoy greater efficiency than dedicated ones because most data communication is bursty in nature, with long idle periods punctuated by brief message transmissions. Reliable channels are more efficient than unreliable ones because retransmissions are not required as often (because there are fewer transmission-induced errors).

Communication Protocols

Protocols specify the rules for communicating over a channel, much as one person politely waiting for another to finish before they speak. Protocols coupled with channel characteristics determine the net efficiency of communications over the channel.

Protocols can improve the effective channel quality. An example is an ARQ (automatic repeat request) protocol in which a source automatically retransmits a message if it fails to receive an acknowledgment from the destination within some predefined time period following the original transmission of the message. The destination knows whether to acknowledge the message based on some error detection capability, which is typically based on redundant information added to the message, such as a parity code or cyclic redundancy check (CRC).

Figure 3 depicts a message (”Pick-up at 1:30 PM.”) being transmitted from a source to a destination via ”packets” which contain four characters at a time. Additional redundant information in each packet allows the destination to know whether or not it has received that packet correctly. Once the destination is satisfied that it has correctly received a packet, it sends an acknowledgment (”ack”) message to the original packet source. When the source receives the acknowledgment, it may transmit the next packet in sequence. In an ARQ arrangement, failure to receive an acknowledgment within a specific time period causes the source to retransmit the packet which was not acknowledged. Only a single transmitted packet remains outstanding (i.e., unacknowledged) at a time.

Error detection and recovery mechanisms can be much more sophisticated than this simple ARQ scheme. One way to enhance the effective channel performance is to allow multiple packets to be outstanding at a time. Individual packets are assigned a sequence number reflecting their order in a sequence of packets flowing from a specific source to a specific destination, which allows them to be separately acknowledged or retransmitted in the event of a failure.

This type of windowing scheme is commonly used when significant delays are involved in the end-to-end data transmission or when the channel has a relatively high quality. When an individual packet is not received correctly, the destination could request retransmission of either the individual bad packet, called selective packet rejection, or that packet plus all succeeding packets. Which of these modes is employed depends on the nature of the communication and the medium used.

Figure 4 depicts such a windowing scheme applied to the packet-based transmission example from above, but with each packet individually numbered in sequence. In this example, up to three packets may be outstanding at a time and the destination must notify the host of the next expected packet number, implicitly acknowledging all preceding packets. Unless a time-out occurs at the source, it will continue to transmit packets until the window-size of three outstanding unacknowledged packets is reached. The destination periodically acknowledges all received packets.

If sufficient redundant information is added to a message, it could enable the receiver of the message to not only detect an error but also correct it. Although this requires some additional processing by the destination, it could obviate the need for retransmission. This error correction capability is generally desirable in channels which are expensive, prone to distortion, or suffer from long latency in the dialogue cycle.

Connection-Oriented and Connectionless Protocols

Protocols can be either connection-oriented or connectionless in nature. In connection-oriented protocols, corresponding entities maintain state information about the dialogue they are engaged in. This connection state information supports error, sequence and flow control between the corresponding entities. The windowing scheme presented earlier is an example of a connection-oriented protocol.

Error control refers to a combination of error detection (and correction) and acknowledgment sufficient to compensate for any unreliability inherent to the channel. Sequence control refers to the ability for each entity to reconstruct a received series of messages in the proper order in which they were intended to be received; this is essential to being able to transmit large files across dynamically-routed mesh networks. Flow control refers to the ability for both parties in a dialogue to avoid overrunning their peer with too many messages.

Connection-oriented protocols operate in three phases. The first phase is the connection setup phase, during which the corresponding entities establish the connection and negotiate the parameters defining the connection. The second phase is the data transfer phase, during which the corresponding entities exchange messages under the auspices of the connection. Finally, the connection release phase is when the correspondents ”tear down” the connection because it is no longer needed.

An everyday example of a connection-oriented protocol is a telephone call. The call originator must first ”dial” the destination phone number. The telephony infrastructure must setup the end-to-end circuit, then ”power ring” the call terminator. From this point on, the connection is in place until one of the parties hangs up. Once the called party answers the phone, another level of connection (between people) must be established before real messages can be exchanged.

Connectionless protocols differ markedly from connection-oriented protocols in that they do not provide the capability for error, sequence and flow control. Nor do they have any connection state maintenance requirement. Each message is considered to be independent of all others in a connectionless protocol. Whether or not a given message is received correctly and when has no bearing on other messages; somehow the destination must sort things out and make sense of it all. Connectionless protocols are always in the data transfer phase, with no explicit setup or release phases as in connection-oriented protocols.

The OSI Reference Model

The Open Systems Interconnect (OSI) reference model is commonly used to describe in an abstract manner the functions involved in data communication. This model, originally conceived in the International Organization for Standardization (ISO), defines data communications functions in terms of layers.

In the OSI reference model, each layer is responsible for certain basic functions, such as getting data from one device to another or from one application on a computer to another. The functions at each layer both depend and build on the functions–called services– provided by the layers below it. Communication between peer entities at a given layer is done via one or more protocols; this communication is invoked via the interface with the layer below.

The OSI reference model is depicted in Table 1. Successful communication between two applications depends on successful functions at all seven layers. In terms of implementation, it is possible for some layers to be trivial; in the end what is required depends on the needs of the applications (and people) engaged in communication.

We must emphasize that the definition of a layered data communication architecture is only an abstraction. The intent of this definition is to unambiguously describe the functions involved in data communication in a way which allows different systems to be compared. The OSI reference model definition is intended to neither imply nor constrain the implementation of any communication system.

Although various companies and standards bodies have created different layered communications models, the OSI reference model remains the universally-accepted common denominator for abstract definition. Other models define the layer functions somewhat differently and often have fewer than seven layers. In some cases constituent protocols were specified before the abstract models defining the end-to-end communication.

We will now review the functions of the OSI layers and some of the primary protocols at each layer.⁹

Layer 1 - The Physical Layer

The physical layer functions include all physical aspects of communicating between two directly-connected physical entities. Typically these physical properties include electromechanical characteristics of the medium or link between the communicating physical entities such as connectors, voltages, transmission frequencies, etc. This layer summarizes the physics which underlie the communication path.

The essential service provided by the physical layer consists of an unstructured bit stream, which can be used by higher layers to provide the basis for higher layer communication services. An example of a physical layer is the ink on paper used by this book to convey information. Another example is the radio frequencies used in a wireless communications system.

Layer 2 - The Data Link Layer

The data link layer accepts the unstructured bit stream provided by the physical layer and provides reliable transfer of data between two directly-connected Layer 2 entities. ”Directly-connected” means that the Layer 2 entities’ communication path does not require another Layer 2 entity. However, this does not imply a dedicated path; in the case of Ethernet, many Layer 2 entities can be sharing a common (physical) medium such as a coaxial cable or a 10BASE-T hub.

Layer 2 functionality is limited in scope–delivery of messages over a local area. It could be likened to an intra-office correspondence between co-workers; there is a need for reliability but addressing is relatively simple. Local area networks (LANs) operate at Layer 2.

The data link layer is itself conceptually subdivided into two sublayers–medium access control and logical link control–which more specifically define the primary aspects of data link layer functionality. However, this conceptual partitioning by the IEEE 802 committee is somewhat arbitrary and subject to debate.

The MAC Sublayer

The medium access control (MAC) sublayer is closely associated with the physical layer and defines the means by which the physical channel (medium) may be accessed. It coordinates the attempts to seize a shared channel by multiple MAC entities, much as a school teacher must arbitrate between pupils’ conflicting desires to speak. The MAC layer commonly provides a limited form of error control, especially for any header information which defines the MAC-level destination and higher-layer access mechanism.

Ethernet (IEEE 802.3) is a prime example of a shared medium with a defined MAC sublayer functionality. The shared medium in Ethernet has traditionally consisted of a coaxial cable into which multiple entities were ”tapped,” as depicted in Figure 5. Although this topology still applies conceptually, a hub and spoke medium is now typically used, in which the earlier coaxial cable has been physically collapsed into a hub device.

As a contention medium, Ethernet defines how devices sense a channel for its availability, wait when it is busy, seize the channel when it becomes available and back-off for a random length of time following a collision with another simultaneously transmitting device. On a shared channel, such as Ethernet, only a single entity can transmit at a time or messages will be garbled.

Not all shared channels involve contention. A prime example of a contentionless shared medium is token ring (IEEE 802.5), in which control of the channel is rotated between the devices sharing the channel in a deterministic round-robin manner. Conceptually, control of the channel is given to the entity currently possessing a ”token.” If the device has nothing to transmit, it passes the token to the next device attached to the topological ”ring,” depicted in Figure 6.

IEEE-defined MAC sublayer addresses are six bytes long and permanently assigned to each device, typically called a network interface card orNIC. The IEEE administers the assignment of these addresses in blocks to manufacturers to assure the global uniqueness that the MAC sublayer protocols rely on for ”plug Ôn play” network setup. Each manufacturer must assure individual device identifier uniqueness within their assigned block.

The LLC Sublayer

The logical link control (LLC) sublayer is responsible for reliable transfer of messages–called frames or, more formally, link protocol data units (LPDUs)–between two directly-connected Layer 2 entities. Functions needed to support this reliable transfer include framing (indicating where a Layer 2 message begins and ends), sequence control, error control and flow control.

The degree to which sequence, error and flow control are provided by the LLC sublayer is determined by whether the link protocol is connection-oriented or connectionless. A connectionless link protocol provides little if any support for these functions. A connection-oriented link might use a windowing technique for these functions, in which frames are individually numbered and acknowledged by their sequence number, with only a few such frames outstanding at any time.

The connection-oriented functions of sequencing, error and flow control provide a foundation for services provided by higher layers. As mentioned earlier, not all layer or sublayer functions are explicitly designed or implemented in any given system. Provision of these functions depends on the services required by higher layers.

If the connection-oriented functions of the LLC sublayer are not implemented, they must be performed by higher layers for reliable end-to-end communication. If these functions are provided by several layers, they might be somewhat redundant and add unnecessary overhead (inefficiency) to the system. In the worst case, redundant provision of these functions at multiple layers could serve cross purposes and actually degrade overall system performance.

An example of a connectionless LLC protocol is frame relay (T1.617, 618), which defines point-to-point links with switches connecting individual links in a mesh topology. In a frame relay network, endpoints are connected by a series of links and switches. Because frame relay is defined in terms of the links between frame relay access devices (FRADs) and switches, and between switches themselves, it is an LLC protocol.

Connectionless Layer 2 protocols are best suited for high quality transmission media. With high quality transmission media, errors are rarely introduced in the transmission between network layer entities and discovery of and recovery from errors is most efficiently handled by the communicating hosts. In this case, it is better to move the packets quickly across the traversed subnetworks from source to destination rather than checking for errors at Layer 2.

Frame relay is derived from the X.25 (ISO 8208) protocol which spans Layers 2 and 3. X.25 is a connection-oriented packet-switching technology which defines how neighboring packet switches exchange data with one another in a reliable manner from end-to-end. Frame relay simply removes the connection-oriented functions of error and sequence control; however, congestion control functions are provided in frame relay, to prevent the total traffic seen at any point in the network from overwhelming it.

Connection-oriented Layer 2 protocols are best suited for low quality transmission media where it is more efficient and cost-effective to discover and recover from errors as they occur on each hop than to rely on the communicating hosts to perform error recovery functions. With ever-increasing quality of transmission facilities and decreasing costs of computation capability at hosts, the need for connection-oriented network layer protocols is diminishing. However, X.25 remains popular outside of North America, where it has been tariffed at levels which encourage its use.

End-to-end communications may be via shared or dedicated facilities or circuits. Shared facilities involve the use of packet switching technology to carry messages from end-to-end; messages are subdivided as necessary into packets, which share physical and logical channels with packets from various sources to various destinations. Packet switching is almost universally used in data communications because it is more efficient for the bursty nature of data traffic.

On the other hand, some applications require dedicated facilities from end-to-end because they are isochronous (e.g., voice) or bandwidth-intensive (e.g., large file transfer). This mode of end-to-end circuit dedication is called circuit switched communication. Because the facilities are dedicated to a single user, this tends to be much more expensive than the packet switched mode of communication. But some applications need it–it is an economic trade-off.

Dedicated circuits are a rather extreme form of connection-oriented protocol, requiring the same setup and tear-down phases prior to and following communication. If the circuit setup and tear-down is statically arranged (i.e., out-of-band), it is referred to as a permanent virtual circuit or PVC. If the circuit is dynamically setup and torn-down in-band, it is referred to as a switched virtual circuit or SVC.

Layer 3 - The Network Layer

The network layer defines the functions necessary to support data communication between indirectly-connected entities. It provides the capability of forwarding messages from one Layer 3 entity to another until the final destination is reached.

The network layer introduces another layer of abstraction to the data communications model. It moves messages–called packets or, more formally, network protocol data units (NPDUs)–between communicating Layer 3 entities–called end systems, nodes or hosts. Network layer functions include route determination or routing and forwarding of packets to their final destinations.

In order to forward a packet to its destination host, routing information must be provided to theintermediate systems (ISs) or routers responsible for forwarding packets to their respective destinations. This routing information includes the address of the destination, which is contained in each packet. The next hop to be traversed by the packet is determined primarily by this destination address. We will talk more about addressing and routing in Chapter 1.

This packet forwarding and routing is accomplished independent of both the media and transmission types used at any step along the way. The unimportance of local topology to the network layer is demonstrated by the common use of ”cloud diagrams” to depict networks, as in Figure 7. Since the network layer is concerned with getting packets across many local networks, called subnetworks, its title would be more accurate if it were the ”Internetwork Layer.”

The network layer functionality is global in scope–delivery of messages over a wide area. It could be likened to the postal system, in which correspondence is passed from location to location until it eventually reaches the destination address on the envelope.¹⁰ The network layer is the domain of wide area networks (WANs).

In order for routers to know how (i.e., on which link) to forward packets, they must have some knowledge of network topology. This knowledge may be complete or partial, and is dynamically created and maintained via routing protocols, used by routers to share their knowledge of network topology with each other. Routing is essentially the reduction of global internetwork topology to local ”hop-by-hop” routing decisions made independently by each router.

As with Layer 2, Layer 3 protocols may be connection-oriented or connectionless. A connection-oriented Layer 3 protocol, such as X.25 (ISO 8208), operates more statically. The basic idea is that an end-to-end route (X.25 virtual connection) is established from the originating data terminal equipment (DTE) to data communications equipment (DCE), from DCE to DCE through the network, then from the last DCE to the terminating DTE; this is the call setup. Packets are then transmitted via this prearranged route, with all packets following the same path through the network. Finally the route is torn down (release) and packets cease flowing.

X.25 operation is like a phone call because it is a phone call. X.25 Layer 3 operation assumes that a reliable connection-oriented service is provided by Layer 2 (also defined by the X.25 standard), although it does provide flow control via sequence numbers.

Connectionless Layer 3 protocols, such as the ever popular internet protocol (IP)(RFC¹¹ 791 and 792) and its ISO counterpart connectionless network protocol (CLNP) (ISO 8473), route packets dynamically. There is no prearranged path which is followed by subsequent packets flowing from one host to another. Instead each packet is individually routed through a routing mesh; there is no reason to believe that sequential packets flowing between hosts will follow the same path. So sequence errors may be introduced at Layer 3, which must be corrected by a higher layer entity.

Connectionless data packets are commonly referred to as datagrams and the service provided by connectionless Layer 3 protocols is referred to as datagram service. Stateless datagram service is simpler for Layer 3 entities than connection-oriented network layer services. Because there is no state information to maintain, dynamic routing protocols can be used. If a router fails during the dialogue between two communicating hosts, neighboring routers will discover this via the routing protocols and find alternate routes which bypass the failed router.

There seems to be a fair amount of ambiguity between the network layer and the LLC sublayer. Both can provide connection-oriented or connectionless services to higher layers. To a large extent, if Layer 3 is explicitly implemented, there is no need for an LLC sublayer. The primary difference is in scope–LLC addresses and protocols are oriented toward a more local environment whereas network layer addresses and protocols are global in scope.

Excellent references to routing and forwarding of data packets can be found in [PERL92] and [STEN95].

Layer 4 - The Transport Layer

The transport layer is concerned with getting Layer 4 messages–called segments or, more formally, transport protocol data units (TPDUs) –from source to destination in a reliable manner. The perspective of Layer 4 is of end-to-end communications rather than the hop-by-hop perspective of Layer 3. Layer 4 assumes that packets can be moved from network entity to network entity, eventually getting to the final destination host. How this is accomplished is of no concern to Layer 4 functionality.

Like other layers, transport layer protocols can be either connection-oriented or connectionless, depending on the services required by higher layers. A common implementation of Layers 3 and 4 involves a connection-oriented transport layer protocol running over a connectionless network layer protocol, such as the ubiquitous TCP/IP protocol suite. In this instance, the communicating hosts maintain state information on communications with each other to determine when and what to send. This state information defines the connection between the communicating Layer 4 entities.

The general idea here is that two communicating hosts need not be concerned with the topology of the internetwork which lies between them. They only need to know the state of their pairwise communication. If part of the intervening internetwork ”cloud” suffers a failure, the Layer 3 entities (routers) will deal with it and recover dynamically. Aside from potential retransmission of any lost segments, the hosts’ Layer 4 entries do not have to be at all concerned with routing and recovery activities at Layer 3.

In the IP protocol suite, the primary connectionless Layer 4 protocol is the User Datagram Protocol (UDP)(RFC 768), which is carried by IP; the primary connection-oriented protocol is the Transmission Control Protocol (TCP)(RFC 793). The ISO world defines five classes of transport layer protocol, beginning with Class 0 (TP-0) for connectionless operation and range up to Class 4 (TP-4)(ISO 8073) for connection-oriented operation.

Layer 5 - The Session Layer

The session layer provides a control structure for communication between applications on hosts. The communication at layer 5 is called a session, which defines the relative timing of communications between the hosts’ applications. Synchronization of communicating applications comes into play when coordinated timing of corresponding events at the endpoints is imperative, such as in financial transactions.

Remember, layers define communication functions, not implementations. It is unlikely that a session layer would be explicitly implemented as a stand-alone program, although its functions would be implemented somewhere. Session layer functions depend on the reliability of communications between the endpoints, and session layer functions must therefore be implemented above Layer 4.

Layer 6 - The Presentation Layer

The presentation layer performs any necessary data transformations or formatting required by the end applications. Functions provided by the presentation layer include data compression, file formatting and encryption. Common data formatting is important because it allows the same application file to be accessed by the application running on different computer platforms. This book is itself the product of an application running on different platforms, with common files being modified via these different platforms.

Abstract Syntax Notation (ASN.1) is commonly used to specify data values in a way which allows processors to communicate independent of their varying native integer sizes, bit orderings (big or little endian), character sets, etc. ASN.1 is a transfer syntax, a presentation layer formatting, which appears frequently in the CDPD specification for unambiguous definition of network management, accounting, limited size messaging and other functions.

An example of ASN.1 encoding from an accounting Traffic Matrix Segment in the CDPD specification is the following:

TrafficType ::= INTEGER {

registration (0),
deregistration (1),
ip(2),
clnp(3)
}

Layer 7 - The Application Layer

The application layer provides the services which directly support an application running on a host. These services are directly accessible by an application via common well-known application program interfaces (APIs), which can actually occur at many layers. Examples of layer 7 services include FTP (file transfer protocol), Telnet and SNMP (simple network management protocol). Most network management activities are based on the services provided by layer 7 application entities, which in turn rely on lower layer services to be able to perform their functions.

Protocols, Primitives and Services

In general, a Layer (N) entity provides services to higher Layer (N+1) entities and relies on the services provided by the Layer (N-1) and below entities supporting it. Layer (N) services consist primarily of transferring messages from one Layer (N+1) entity to another; both the source and destination Layer (N+1) entities rely on their underlying Layer (N) entities to accomplish this task.

A Layer (N) entity requests services of a local Layer (N-1) entity via primitives directed at a Layer (N-1) service access point (SAP). If the primitives are explicitly implemented, they can be thought of as function calls.

Protocols refer to relationships and messages between peer entities at a given layer. A Layer (N) entity communicates with another Layer (N) entity via a protocol. A protocol message is actually invoked by means of a service request primitive–(N)-PRIMITIVE_NAME.request–to its underlying Layer (N-1) entity, where ”PRIMITIVE_NAME” is the name of the operation being invoked by the primitive. The peer Layer (N) entity receives the Layer (N) protocol message via a service indication primitive–(N)-PRIMITIVE_NAME.indication–from its underlying Layer (N-1) entity.

In a connection-oriented protocol, the peer Layer (N) entity responds via a response primitive–(N)-PRIMITIVE_NAME.response, to its underlying Layer (N-1) entity. The original Layer (N) entity receives the Layer (N) protocol response from its underlying Layer (N-1) entity via a (N)-PRIMITIVE_NAME.confirm primitive. This primitive flow supporting the Layer (N) protocol is displayed in Figure 8.

Protocol and Service Data Units

Protocol data units or PDUs are the messages passed between entities at a given layer. Layer 2 PDUs are called LPDUs or frames; Layer 3 PDUs are called NPDUs or packets; Layer 4 PDUs are called TPDUs or segments.

In general, a PDU–regardless of the protocol layer–consists of header¹² and data fields. The header field contains the information necessary to get the PDU to the peer entity and typically includes the source and destination addresses appropriate for that layer as well as error sequence and flow control information. The data field contains the information carried by the Layer (N) protocol in support of Layer (N+1); it is formally referred to as the Layer (N) service data unit or SDU.

Conceptually, when a Layer (N+1) PDU is passed via primitive to Layer (N) as a Layer (N) SDU, a Layer (N) header is prepended to create a Layer (N) PDU. Sometimes part of the ”header” is actually appended at the end, usually for error correction purposes. The Layer (N) PDU is then passed via primitive to Layer (N-1) as a Layer (N-1) SDU where a Layer (N-1) header is added. This process continues as data units are passed down the OSI reference model ”stack.” This is depicted in Figure 9.

Similarly, when a Layer (N-1) SDU is passed up to Layer (N), the Layer (N-1) header is removed from the Layer (N-1) PDU. Likewise, the Layer (N) PDU header is stripped to provide the Layer (N) SDU for Layer (N+1). This process continues as data units are passed up from layer to layer in the OSI reference model. Eventually, as shown in Figure 9, the original data element has been recreated at the application layer of the destination.

Mobile Data Communications Entities

At each layer of the OSI reference model there are protocol entities communicating with each other. They are the sources and destinations of PDUs at that layer. Because this book is about mobility in WANs, the entities of greatest interest are Layer 3 entities, commonly called hosts, nodes or end systems. Layer 3 PDUs, commonly called packets, are exchanged between hosts via Layer 3 entities commonly called routers.

Adding the capability of mobility to a wide area data network creates a need for defining additional entities, as depicted in Figure 10.

A mobile host or mobile is a host which can receive network services regardless of its location. The extent to which this host enjoys transparent location-independence is a key concern. Different systems use different terminology for the mobile; CDPD adopts ISO terminology by calling it a Mobile End System (M-ES). The mobile is an occasionally-connected entity, which means it may or may not be connected at any given moment to a subnet somewhere in the mobile WAN.

The second role necessary in any communications is the opposite side of the correspondence. In this book we refer to this as the correspondent host or correspondent. The correspondent is the location of the opposite side of a mobile’s application association; it could be the ultimate source of data destined for the mobile or another entity such as a store-and-forward device. The correspondent could itself be mobile or fixed in location, but this is generally not material to our analysis. CDPD refers to the correspondent as the Fixed End System (F-ES) when it is fixed in location.

In circuit-switched systems, there will be a maximum of one correspondent per mobile host. However, in the packet-switched systems of greatest interest to us, there can always be multiple correspondents per mobile host.

Associated with the mobile communications is the assisting entity or assistant. The assistant is an enabler of mobility. It could be a network store-and-forward device or mobility-supporting intermediate system (router). Most likely, it consists of multiple entities in a mobile network infrastructure which collectively support host mobility. In CDPD this role is largely filled by a combination of the Mobile Serving Function and the Mobile Home Function; the Mobile IP Task Force calls this combination the foreign agent and the mobile router or home agent.

As we shall see, the essential problem of mobility is getting data to the mobile. Thus, the mobile will generally be the destination host in any mobile system scenario of interest. Consistent with this, we will adopt a system-centric viewpoint by defining the flow of data traffic to a mobile as moving in the forward direction (i.e., mobile network to mobile host). Likewise, the flow of data traffic from a mobile is said to move in the reverse direction (i.e., mobile host to mobile network). This terminology is consistent with the cellular industry, which is the origin of CDPD, and is displayed in Figure 11.

Summary

This chapter has presented a brief survey of data networking terminology and concepts used throughout the remainder of this book. We strongly encourage readers who are new to this terminology to peruse the many fine texts on this subject, some of which have been referenced in this chapter. Given the terminology presented in this chapter, we will now begin discussing mobility.

Chapter 1
Introduction to Mobility

This chapter introduces some of the key concepts of mobility, setting the stage for later detailed discussion of mobile WAN systems. Although the emphasis of this book is CDPD, this chapter presents a more generalized and abstract view of mobility.

Previous books on the subject of mobile data communications have covered wirelessness¹ more than mobility. Our intent is to begin from a data networking perspective, independent of media-specific considerations, then consider the effects of media on networking technology. Indeed, one of the primary conclusions of this book is that mobility is most naturally handled in the network layer (Layer 3) and so any discussion of mobility should begin there.

1.1 What is Mobility?

Mobility means different things to different people. Some people are quite happy being able to get around town. Others view the world in terms of time distance–four hours from Chicago to San Francisco by airline, perhaps. Obviously, range of motion is an important aspect of mobility.

Another factor in mobility is ease of access. What might be considered mobile in one context is quite immobile in another. Certainly the pioneers crossing the North American continent in ox-drawn wagons covered the same distance as the airliner from Chicago to San Francisco. But we would today hardly consider these pioneers to have had much mobility.

A more pertinent example of mobility is the ever decreasing size of cellular telephones. What was once considered a ”mobile phone” had to be transported in a vehicle. A major step forward was the ”transportable” phone, which freed the user from their vehicle but weighed in at about twenty pounds, still huge by today’s standards. With the advent of ”brick” phones in the mid-1980’s came the era of ”portable” phones. This continuing decrease in size and weight of handsets has greatly increased the mobility of cellular subscribers.

In this book on mobile data communications we define mobility asthe ability to send and receive communications anytime anywhere. Mobility means that both source and destination devices, applications and people are free of the constraints imposed by physical location. Access to an Ethernet port, for example, need not limit one’s ability to send and receive data in a mobile WAN environment any more than access to a landline phone currently limits one’s ability to place a voice call in an area covered by cellular service.

1.2 Basic Approaches to Mobility

The issues of mobility in a WAN environment can be discussed in the context of a modern day ”road warrior,” whom we shall call Gary² . Gary is never totally alone in his travels–as in real life, he cannot survive on the road for long without active support from his administrative assistant, whom we shall call Pat³ . If you had to communicate with Gary, who (as always) is on the road, how could you accomplish this?

1.2.1 Approach 1: Application Awareness

One way to communicate with Gary is to contact him at his hotel or wherever he is. If he had provided you with a copy of his itinerary and none of his plans changed (certainly an ideal case!) you could call or send facsimiles to Gary at the location specified in the itinerary.

Of course, Gary would have had to anticipate your need to contact him or he would likely have not provided you a copy of his itinerary. In the absence of clairvoyance, Gary or Pat would have to provide copies of his itinerary to all people who might ever have a need to contact Gary while he is traveling–all of the time! Even people he might not really want to hear from...

Obviously there are serious logistical problems with this approach to mobile communications. On top of the logistics and overhead of broadcasting Gary’s itinerary to the world, there are issues of privacy and the undesirability of always being accessible. It is just possible that Gary would prefer to not be interrupted while in the midst of a serious contract negotiation for an unrelated issue.

An equivalent situation in mobile data communications would be a mobile host having to directly notify each of its applications’ peers about its current location or address, as depicted in Figure 1.1. Any peer application not receiving this location update would be unable to communicate with or engage in a session with the mobile host. This would be like trying to call someone at their office phone number when they are really at home, at a different number (address).

The primary advantage of this Application Awareness approach to mobility is that the network infrastructure does not itself need to be involved in or even aware of the mobility.⁴ Mobility management and network access are entirely within the scope and control of end users and applications. The mobile host and application must provide location and routing information to each of the application’s peers. Each peer then cooperates by connecting with the mobile application according to the new routing rules.

The primary disadvantage of this scheme is that by eliminating network involvement, end users and applications must perform routing tasks typically handled by other entities. This approach requires custom (i.e., nonstandard) applications to furnish mobility awareness and support. These applications on both the mobile and peer sides would then differ from those which operate in an immobile world. However, success of a mobile data technology depends to a large extent on the ease of application attachment to the mobile network.

This disadvantage is exacerbated in that end applications would have to collect and maintain evolving routing information for hosts which might be mobile. This is highly inefficient and goes against the grain of a functionally-layered network architecture. Application layer entities should not be concerned with network layer issues such as routing and addressing.

A variation of the Application Awareness approach was traditionally used in the cellular phone environment for ”roaming” subscribers. Prior to ”roaming” into another service provider’s territory, a cellular subscriber would have to prearrange for service in that area with their home service provider.⁵ In this case, the end user was involved in what is essentially a routing issue to be able to receive calls on their cellular handset.

Perhaps for these reasons, there are no good implementations of the Application Awareness approach to mobile data communications. This approach is only feasible in vertical applications running in closed networks in which protocol layers are not functionally separated.

1.2.2 Approach 2: Directory Lookup

Another way to contact Gary the road warrior is via Pat, his administrative assistant. If Gary always notifies Pat about his location (by itinerary and verbal updates), you could first call Pat’s office to get current phone or facsimile numbers for Gary. Armed with this information, you could then contact Gary directly.

This approach is more flexible than only using a previously-furnished itinerary, which is subject to change. However, if Gary has been less than diligent in his location updates, Pat may not really know where Gary is. Furthermore, anyone wishing to contact Gary must first go through Pat, which means they must know how to reach Pat. And Pat must always be available.

Although this approach is logistically feasible, Pat may not always know whom to give Gary’s location information to. It is possible that Gary uses Pat as much for call screening as for support of his mobility. But at least you would know to always contact Gary by first calling Pat.

This Directory Lookup approach is similar to the use of the Domain Name System (DNS) to determine addresses used for routing in the conventional⁶ Internet. In the World Wide Web, sites are typically accessed via Universal Resource Locator (URL) identifiers and not via Internet addresses. These human-recognizable names are used by Web browser software to interrogate one or more DNS servers, which respond with an Internet address reflecting the Web site’s network location. The IP address returned by the DNS query is used to route subsequent Web page accesses (i.e., data packets).

The DNS system provides this ubiquitous name-to-address translation function via a highly distributed and replicated database of translation information. The degree of replication of this database provides dependable and rapid worldwide address translation. However, this replication also makes the synchronization of updates throughout the database somewhat slow. This is the price of success.

The DNS mechanism works well because Web sites typically do not change network access points (i.e., addresses) very often. The relatively static domain name to IP address translation information can be safely propagated and cached without fear of rapid obsolescence. Unfortunately, this is not the case for mobile hosts, whose locations change far too quickly for efficient DNS tracking.

Unlike current DNS queries, mobile host location queries would have to be done (almost) every time peer applications wanted to correspond with the mobile host, as depicted in Figure 1.2. Although some caching of location information could be done–as in DNS–this information is extremely time sensitive. If the destination host is in motion, the location information could quickly become obsolete. Thus, the overhead of determining a route (address) for a mobile would have to be incurred prior to sending every packet to the mobile.

To avoid this overhead with the Directory Lookup approach to mobility, the mobile host cannot move to a different network attachment point, while one of its applications is engaged in a session with a remote peer. Doing so would break the connection, forcing the session and application to end prematurely in a highly disruptive manner.

To prevent this disruption in a Directory Lookup scheme, the mobile host must restrict its movement whenever connections are active. While this might be acceptable for mobile host-originated application associations, it is clearly deficient for application associations originated by application peers of the mobile.⁷

The Directory Lookup approach enjoys the advantage that any peer application wishing to contact the mobile host need only query a central repository of location information. Likewise, the mobile host need only notify the central directory service of its location. There is no need to contact each and every one of the potential peer applications which might have a need to communicate with that mobile host.

However, the Directory Lookup approach to mobility still involves application participation. Applications must understand mobility enough to maintain associations between the mobile host and the directory lookup procedures. If the mobility directory is implemented in a standardized manner a la DNS, the lookup can be a natural part of any application. However, if it is not standardized, applications would need to be modified specifically for mobile hosts.

For all of these reasons, CDPD does not use the Directory Lookup approach to mobility. The CDPD network is required to support mobile devices which change their point of access frequently. Within a metropolitan area, a moving host may traverse multiple points of network attachment in a few minutes. This type of movement requires extremely frequent mobile directory updates.

Furthermore, as we have seen, the Directory Lookup approach to mobility is ineffective for any session lasting longer than a few seconds involving a moving host. Since an application connecting with a mobile host cannot rely on routing information obtained from the mobility directory more than a few seconds ago, it must reacquire routing information for every new association. This is a change to the current DNS mechanism as well as inefficient from a system perspective. For all of these reasons, this approach is inappropriate for a mobile data network such as CDPD.

1.2.3 Approach 3: Mailbox Service

A third way to contact Gary the road warrior is to use a messaging service of some kind, such as voice mail. With this approach, all of Gary’s associates would call the number on his business card to leave messages for him. At a later time, Gary could retrieve his messages from the answering service and return calls.

Unfortunately this approach does not allow any live discussions between Gary and his peers. Unless Gary happens to check his messages frequently or happens to return calls at just the right moment, he will simply end up leaving a message himself. Although many business relationships can proceed via exchange of voice mail messages, the unavailability of live communication will eventually become a problem. This approach is intentionally used by many people to screen unwanted phone calls.

This Mailbox Service approach to mobility is similar to Email repository services such as RadioMailTM. This approach allows the users of mobile hosts to retrieve their Email whenever they are in an area of mobile connectivity. However, this approach prohibits interactive activities such as on-line database queries.

This approach benefits from peer applications no longer having to involve themselves in the mobility tracking process. Each peer application wishing to communicate with the mobile host simply sends information to the designated message repository for the mobile user. In this way the peer application is no different than any standard electronic messaging application.

In order to use the Mailbox Server approach to mobility, a peer application must send a message to the mailbox and await a response, as depicted in Figure 1.3. The mobile host will only become aware of the message by interrogating the mailbox or by the mailbox broadcasting notifications to the mobile host, wherever it is. As we shall see, this dilemma has been addressed by CDPD with the Status Notification Service (SNS) support for applications like messaging.

The Mailbox Service approach does not provide real-time communications and is thus inappropriate for interactive applications. It also requires mobile network-specific protocols between the mobile host and the network infrastructure. Both the network and application are impacted by and must be modified for mobility. In that sense it could be considered to be the worst of both worlds.

1.2.4 Approach 4: Administrative Redirection

Another approach to contacting Gary the road warrior is for your calls and facsimiles to be forwarded to his current location. You don’t even have to know that this is happening. Packages and mail can also be redirected to someone on the move like Gary–Pat simply repackages them for overnight delivery to Gary’s anticipated future location.

This approach greatly simplifies communicating with Gary. All you need to know is his ”anchor” location–his home office, where a combination of modern telephony and Pat support Gary’s mobility. As long as Gary notifies Pat of his anticipated next day’s whereabouts and keeps his phone forwarded (or, better yet, uses a mobile phone), no-one else needs to be aware of Gary’s absence from the office.⁸

This Administrative Redirection approach to mobility can be applied in a data networking context by associating the mobile host with a ”home” location. This network address and its corresponding domain name, URL, etc., would be the communication coordinates used by applications corresponding with the mobile host. All of these coordinates could be advertised in the conventional ways–via routing information protocols, DNS servers, etc.

Whenever corresponding entities wish to communicate with a mobile host, they would send packets to the home address associated with the mobile host, as depicted in Figure 1.4. These data packets would be redirected by a mobility-aware home server or agent, much as Pat would repackage and forward written correspondence to Gary. In this way, peer applications (and people) could remain blissfully unaware of the mobility and location of the host (person) they are corresponding with.

The primary advantage of the Administrative Redirection approach to mobility is that it supports direct interactive communication between a mobile host and peer applications in a transparent manner. The data connection or association is achieved without the peer applications being aware of mobility and routing management issues. This greatly eases application attachment, which in turn encourages mobile network acceptance and adoption.

Obviously this approach requires some work on the part of the network to maintain transparency of mobility at the application level. The home mobility server must always know where the mobile host is, then redirect data packets to that location. Standard techniques available for this redirection include readdressing the original packets or encapsulation of the packets in new packets with updated destination addresses. In any case, new protocols and procedures must be established for the network infrastructure to support transparent mobility.

The Administrative Redirection approach to mobility has been adopted by both the CDPD and the Mobile IP Task Force specifications, as well as Novell in its proprietary Mobile IPX protocols. This approach can be implemented via one of several mobility management schemes, described in a subsequent section.

1.3 Aspects of Mobile Communications

There are two aspects to mobile communications: mobile network access and mobility management, which are depicted in Figure 1.5. In this book we will address both of these aspects in the context of mobile data communications, as well as other issues which, although less centric to mobility, are important to any ”real world” business.

1.3.1 Mobile Network Access

The first aspect of mobility is accessing or connecting to the communications network. This network access is media-specific and might involve using a pay phone, plugging a 10BASE-T jack into a wall socket or powering-up a wireless device within an appropriate coverage area. We discuss network access for CDPD systems in Chapter 5.

Because mobile network access is a big issue, previous books on mobile data communications have tended to focus on it. But ”getting connected” is only half of the battle.

1.3.2 Mobility Management

The second aspect of mobile communications is the ability to efficiently send and receive communications from wherever you are, once you have accessed the network. This is called mobility management or location management. Somehow the rest of the world must be able to reach you wherever you are. This is the essential challenge of mobile communications and the manner in which this challenge is addressed has broad implications on mobile data applications.

1.4 The Essential Challenge of Mobility Management

If we ignore (for now) the physical issues involved in data transmission and network access, it is clear that sending messages from a mobile entity is not a challenge from a data networking perspective. If a mobile device has data to send, it does so and, neglecting medium-specific issues, the data can be forwarded in the usual way to its destination. [IOAN93] states that ”routing traffic from the mobile [device] is considered a trivial problem; if the destination is an ordinary host in the network, normal routing procedures should be followed.”

However, being able to receive data at a mobile entity is a significant challenge from a data networking perspective. The essential problem of mobility management is efficiently getting data to a mobile entity, which can be anywhere⁹ or nowhere. The mobile is essentially nowhere if it is powered down or out of range of the system; this is an important situation to consider in any mobile system.

1.4.1 Knowing Where the Mobile is

The first part of getting data to a mobile device is knowing where the mobile is. Of the previously-described basic approaches to mobility, only the Mailbox Service approach did not require prior knowledge about the mobile’s location by another entity. So any interactive data communications application will require some assisting entity in the mobile network to know where the mobile is before engaging in the communication.

There are two basic strategies that a data sender can employ to know the location of the intended mobile recipient. The sender (or, alternatively, the system) must either continuously track the location of the mobile device or search for the mobile immediately prior to sending data to it. Which of these mobile device location strategies should be used depends to a large extent on the nature of the intended communication.

There is a tradeoff between ease of tracking and ease of searching in a mobile environment; the efficiencies realized by these techniques are mutually exclusive. One of the primary conclusions of [IOAN93] is that a system can only be optimized for tracking or searching, but not both. A mobile system could support both strategies, but only be optimized for one of them.

If the nature of the intended communication is brief and bursty with potentially multiple sources sending data asynchronously to a mobile host, continuous tracking of the mobile device is more efficient than trying to locate the mobile in real time [IOAN93]. Continuous tracking generally requires the mobile device to actively participate in the tracking process by notifying the system of its initial location and any subsequent changes in its location.

This continuous tracking technique is how systems such as Mobile IP and CDPD operate. It is also similar to conventional routing protocols, such as RIP, OSPF, etc., which allow routers to automatically adapt their routing tables based on changes in network connectivity, although at a much slower rate.

If the nature of the intended communication is relatively lengthy and typically involves a single peer at a time, greater efficiences can be realized by searching for the mobile device immediately prior to sending data to it [IOAN93]. The relative overhead per communication event is small and the system can be greatly simplified by not requiring a continuous tracking mechanism for mobiles.

Searching for a mobile device generally involves some kind of paging operation by the system immediately prior to sending data to it. This is how circuit-switched systems such as cellular voice or circuit-switched data operate. With relatively lengthy sessions involving the mobile host, this is often the best option. The search can be optimized to a degree by beginning where the mobile was last observed.

If the network itself cannot perform this searching function, the corresponding entity must itself locate the mobile, perhaps via a ”mobility agent.” Such an agent would look for the mobile device and report its location to the corresponding agent. Of course, all of this must happen quickly to be of much use.

Regardless of whether a mobile system is optimized for tracking or searching, it is essential to maintain an information base of mobile locations and to be able to quickly propagate that information whenever needed. The approach used will, of course, depend on the tracking vs. searching system design decision. This location information base is analogous to routing table information in conventional networks.

1.4.2 Routing Data to the Mobile

The second part of getting data to a mobile device is routing the data to the mobile host. This consists of determining the route to be followed by the data in its journey to the mobile, then actually forwarding the data along that route. Both of these steps are identical to packet routing and forwarding in conventional connectionless data networks, and depend on readily-available and accurate mobile location information.

1.5 Mobility Management is a Network Layer Function

It should be clear from our discussion thus far that mobility management consists largely of routing data packets (NPDUs) to hosts which change location and network access frequently relative to conventional hosts.¹⁰ By definition, routing is a network layer function and thus mobility management should also be a network layer function. Efficient support for mobility must be designed into the network layer for any mobile WAN.

As observed by [IOAN93]: ”The problem of Mobile Internetworking [can be] posed as follows: how to provide seamless and transparent network connectivity to mobile networked computers (or other communicating devices) as they change location, networking interfaces, or even service providers. We term this work ÔMobile Internetworking’ because it enables mobile entities to communicate within an internetwork, i.e., a network of networks, and not just a local, connected network.”

1.5.1 Network Layer Addresses

Data packets are routed across conventional internetworks via their destination host network layer addresses. The ability to route packets toward their final destination is based on the fact that network layer addresses are typically composed of a network-identifying part and a host-identifying part. The network-identifying part of the address specifies on which network the host may be found. The host-identifying part of the address specifies which host on the network is desired.

Figure 1.6 depicts an IP address, which is four bytes long. The network-identifying part is called the netid and the host-identifying part is called the hostid. Originally the separation between these fields was required to be at one of the three byte boundaries; three corresponding classes of network address space were defined. Now with IP version 4 (IPv4), the boundary between netid and hostid is identified via a bit mask, allowing complete flexibility for IP address space allocation.

Because of the rapid adoption of IP-based technology and the Internet, IP address space had become a precious resource by the early 1990’s. Four bytes sounds like a lot of addresses, but typically host address assignments are relatively sparse based on organizational and network boundaries.¹¹ Partly as a result of the address space limitation and also because of the protocols defined, the IP addresses are typically manually administered and statically assigned to hosts. IPv6 (the ”next generation”) [BRAD96] has extended the size of IP addresses to 16 bytes, largely to alleviate these concerns about address space availability.

1.5.2 Network Topology Changes

The assignments of network layer addresses to hosts is based on the topology (state of connectivity) of the network. Routing information (contained in the routers in the network) can be considered to be a form of distributed database, where a partial view of the network topology information is contained in each router. Each router must be capable of selecting the ”next hop” for each packet based on its ultimate network layer destination address.

Figure 1.7 depicts routing in conventional data networks. A data packet is forwarded from its source located somewhere in the rest of the world to its destination host, Host 1. Table 1.1 depicts the corresponding entries in the routing tables at Routers W, X, Y and Z, which participate in the packet forwarding. The null entry in the ”next hop” field for Router X indicates that Host 1 is local to Router X.

Over time this distributed routing database evolves to reflect changing WAN connectivity. As links and nodes come and go, the routing tables must reflect these changes. This routing table adaptation can either be manually administered (called ”static routes”) or automatic (i.e., based on router protocols such as RIP and OSPF) [PERL92]. Automatic routing table updates support changing network environments due to configuration changes (intentional!) and failures (unintentional!).

Novell’s proprietary IPX protocol uses ten-byte network layer addresses in a creative fashion to support a plug-and-play network configuration capability. The first four bytes of the address represent the network and are called ”network.” The last six bytes are called ”node” and identify the host (”node” in Novell terminology). This field is identical to the device’s permanent six-byte MAC sublayer identifier.

In IPX there is no requirement for network administrators to explicitly configure each node address, only each router’s network value. Plug-and-play capability–certainly a factor in mobility–is supported in that a host only needs to determine its ”network” value by querying a local router when the host is joined to a network. Then the node must notify each of its application peers of its new ”permanent” address which it has self-configured.

This self-configuration by IPX significantly reduces the effort required to move devices from one location to another and prevents node address conflicts. It also eliminates the need for a protocol, such as ARP¹² , to provide the network layer to data link layer address mapping for local routing.

1.5.3 Routing Table Updates

Rapid convergence of router protocols (i.e., adaptation of the routing information database to changing network state) is a primary concern in WANs. Routing protocols must converge more quickly than network topology changes occur or the internetwork operation will break down from the congestion caused by misdirected packets. In fact, one of the biggest drawbacks to the popular RIP protocol is its slow convergence in large-scale internetworks.

Conventional data network routing table updates must be done frequently enough to prevent congestion in the event of a link or router failure or network reconfiguration. [IOAN93] observes that ”if the links go up and down faster than the [routing] protocols can converge, routing may not be possible even though the physical paths exist.”

The need for rapid convergence is amplified in mobile environments, where the movement of hosts creates and destroys links (to those hosts) dynamically and presumably more frequently than failures occur. Routing information needs to be rapidly shared amongst mobility routers to ensure consistency between their routing tables and the represented physical network topology.

[KRIS95] notes that ”conventional routing protocols were not designed for networks where the topological connectivity is subject to frequent, unpredictable change.” Although current routing information protocols support adaptive routing updates to reflect network and host connectivity changes, these protocols are designed for infrequent updates and failures, and thus inadequate to support mobility.

1.6 Mobility Management Schemes

Network layer-based mobility management schemes can be categorized as falling into one of three basic types per [IOAN93]. They will be summarized in the following subsections.

1.6.1 Permanent Address Scheme (PAS)

The first mobility management scheme is called the Permanent Address assignment Scheme [IOAN93]. In PAS the mobile host network layer address remains constant, with the routing system adapting itself to changes in the host’s location.

To support PAS, each router in the mobile internetwork must maintain a host route (routing table entry consisting of host address and next hop for that host’s current location) for each mobile host in the internetwork. As each mobile host moves throughout the internetwork, all routers must update their routing tables accordingly.

Figure 1.8 depicts a mobile host, Mobile 2, changing its point of access to the mobile internetwork from Area X to Area Z. Accommodating this change in PAS network topology entails changes to the routing tables in Router X, Router Y and Router Z; these routing tables are displayed immediately prior to the movement of Mobile 2 in Table 1.2 and immediately following the movement of Mobile 2 in Table 1.3. Because Router W has Router X as the ”next hop” for both Area X and Area Z, its routing table remains unchanged by the movement of Mobile 2.

PAS is problematic because it requires every router in the mobile internetwork to have a current host route entry for every mobile host–which might number in the millions. This simple scheme clearly does not scale well and is thus inappropriate for large mobile networks. Too bad–it’s easy for the mobile hosts and transparent to their correspondents.

1.6.2 Temporary Address Scheme (TAS)

The second basic mobility management scheme is called the Temporary Address assignment Scheme [IOAN93], which is the logical opposite of PAS. In TAS a mobile host adopts a network layer address consistent with its current subnet location. When the host moves, its network layer address changes to reflect its new location.

In TAS, the onus of work supporting mobility is placed on the mobile host and its applications rather than the network infrastructure. TAS is a form of the Application Awareness approach to mobility.

Figure 1.9 depicts a mobile host, Mobile X.1, moving from Area X to Area Z; in the process its temporary address changes from X.1 to Z.2. Because this new address reflects its new location in Area Z, packets addressed to Mobile Z.2 will be forwarded to it via conventional routing. The routing tables for Routers W, X, Y and Z are depicted before the move in Table 1.4 and following the move in Table 1.5. In all routing tables, the entry for host X.1 has been deleted and a new entry for host Z.2 has been created.

Unfortunately, peer applications will not be able to create associations with the mobile host for a while because its network layer address changed. Until peer applications know to use the Z.2 address rather than the X.1 address, the Mobile host will be unable to receive packets.

Obviously, TAS is trivial for routers to support–they don’t need to do anything special.¹³ However, maintenance of accurate DNS name server information is extremely difficult with TAS.¹⁴ Also, mechanisms are required to ensure global network address uniqueness.¹⁵

Changing network addresses impacts TCP and other transport layer protocols which assume permanent¹⁶ network layer addresses. Under TAS, sessions would have to be torn-down whenever a host move required a change of address. This is the same problem created by the Directory Lookup approach to mobility and conflicts with our desire for transparent mobility.

1.6.3 Embedded Network Scheme (ENS)

The third basic scheme for mobility management is called the Embedded Network Scheme [IOAN93]. ENS embeds a virtual network of mobile hosts and support infrastructure (mobility routers and other assistant entities) in the midst of a larger internetwork. These mobility routers serve to provide mobility in the virtual network; other elements in the data network infrastructure, such as routers, can remain ignorant about host mobility.

Figure 1.10 depicts a mobile host, Mobile X.1, moving from Area X to Area Z. Only the routing tables for the ”special” mobility-aware routers–Routers X* and Z*–have been changed to reflect this change in network topology. The routing tables for all routers are depicted in Table 1.6 prior to the mobile’s movement and Table 1.7 following the mobile’s movement.

By decoupling mobility management from the conventional routing mechanisms, ENS provides a more efficient means of routing to mobile hosts and supports constant network addresses. ENS typically involves either data packet encapsulation or the use of secondary temporary addresses by the ”special” routers to provide mobility. ENS embodies the Administrative Redirection approach to mobility in its transparency to the rest of the world.

CDPD services are provided via an ENS-type of mobility management; this mobility is transparent to existing routers and hosts. The Mobile IP definition and Novell’s Mobile IPX are also implementations of the ENS concept. The primary difference between these systems is in the way that the assisting entities are defined.

ENS Variations

In CDPD and Mobile IP, two cooperating assisting entities provide mobility management. One of them–the Mobile Home Function in CDPD, the mobility router in Mobile IP–serves as the public local router for packets destined to the mobile host. The second assistant–the Mobile Serving Function in CDPD, the foreign agent in Mobile IP–receives packets which have been redirected by the first assistant and forward them on to the mobile via the local data link.

Packet encapsulation or tunneling is used by CDPD and Mobile IP to redirect packets from the first to the second assisting entity, as depicted in Figure 1.11. The details of this mobility management scheme is presented in Chapter 4 (for CDPD) and Chapter 10 (for Mobile IP).

Novell’s Mobile IPX differs in that it uses a single assisting entity, called the home router, to forward packets to a mobile node (host). The home router advertises an ”internal” network number which the mobile adopts as its network number field in a so-called ”permanent” address; the node number field for this permanent address is sequentially assigned by the home router to registering mobiles.

Whenever a mobile relocates, it determines its location-dependent (physical) IPX address in the usual way (by prepending a local router’s network number to its hardwired MAC identifier, as we discussed in Section 1.5.1). The mobile then notifies its home router of its new location. All the home router has to do is maintain a table of permanent address to physical address mappings.

Whenever an IPX packet destined for the mobile’s permanent address arrives at the home router (based on standard IPX packet routing), the home router replaces that destination address (the mobile’s permanent address) with the mobile’s current physical address and redirects the packet onward as depicted in Figure 1.12. The mobile then internally replaces the substituted physical address with its permanent address to avoid confusion by its higher layers and applications. This destination address substitution may be compared with the packet encapsulation technique used by CDPD and Mobile IP.

1.7 Steps in the Mobility Management Process

Three steps are involved in accessing and using mobile data networks. These steps are necessary for mobility management and are analogous to the phases of a connection-oriented protocol. The details of each step depend largely on whether the system uses a tracking or searching mobility management strategy as described in Section 1.4.1.

1.7.1 Registration

At some point, the mobile host must announce itself to the system (”here I am!”) before it can receive services. This announcement or registration serves several purposes, including establishment of a SubNet Point of Attachment or SNPA (i.e., the local address where the mobile can be accessed), obtaining network connectivity via the local mobility-supporting router, authentication of its identity, authorization for use, encryption key exchange, link layer parameter negotiation, etc. The mobile registers to a single mobility area at a time.

If the system employs a tracking-based mobility management strategy, the mobile will register once and thereafter remain virtually connected to the system, even if it neither sends nor receives data. If the system employs a search-based mobility management strategy, the mobile will register each time it wants to transmit data, or in response to a system page (because another entity wants to send it data).

Registration includes but is not limited to mobile network access. After performing the actions necessary to access the mobile network, the mobile host must then initiate the mobility management process.

1.7.2 Usage

Once the mobile host has registered to the system, it may use the resources of the system to access the rest of the world. At this point, the mobile host must behave and serve its applications much the same as a conventional host. Mobile network usage could last a long time or for only the duration of a brief data communication. During this interval, the mobile host is accessed (i.e., data packets are sent to it) via the mobility router to which it is registered.

1.7.3 De-registration

Following its use of the system, the mobile host leaves its subnet point of attachment. The de-registration serves the purpose of freeing-up network resources (such as identifiers, memory blocks, routing table entries, etc.) that would otherwise be allocated to a quiescent mobile. A mobile host might deregister because it no longer requires mobile access or because it has moved out of the mobility area to which it was registered.

1.8 A Simple Taxonomy of Mobility

Mobility can exist in many forms. To assist in analyzing different mobile systems, we offer the following simple taxonomy of mobility. In this taxonomy, levels of mobility depend on the degree of location-independence and in-motion capability provided by the mobile system.

We should point out, before beginning our mobile taxonomy, that mobility could be considered to be just one of the dimensions of a mobile data system. As pictured in Figure 1.13 the medium in use and the protocols supported are additional dimensions to consider. Depending on the situation at hand, different mobility solutions are more or less appropriate.

1.8.1 Type 0 Mobility: Stationarity

Type 0 Mobility describes systems which do not support mobility. Network entities are limited to essentially fixed locations for communication with others. This is the level of mobility provided by conventional data networking technology.

In conventional networks, the network layer address indicates the topological location of the host. The netid part of the address identifies the subnet on which a host can be found and is used by routers to forward a packet to its destination. Once the packet reaches the ”last router” which connects the destination subnet to the rest of the world, the hostid part of the address is used to identify the destination host on that subnet¹⁷ . In general, the network address (netid + hostid) defines the host’s location relative to well-known places in the internetwork.

Typically, human intervention is required to configure and administer network addresses of hosts in Type 0 systems. In these systems, moving a host from one location (subnet) to another amounts to creating a new entity (i.e., network address). [RICH95] states that ”the static addresses of traditional network architectures bind a computer to a specific LAN or subnet. Current versions of IP, for example, assume that an IP node has a fixed point of connection to its network.”

Type 0 systems are location-dependent and are thus not mobile. Examples of Type 0 systems include not only traditional networks, but also wireless LANs and campus networks such as Metricom Ricochet. The limitations to mobility in such systems are often due to the technology not being scalable to the magnitude required for ubiquitous mobile operations. In any case, the mobility is limited to the subnet or immediate (LAN) medium containing the host.

1.8.2 Type 1 Mobility: Location Independence

Type 1 Mobility describes systems in which hosts enjoy mobility regardless of location. We make no distinction between systems which could be deployed everywhere and those which are deployed everywhere. Rather than discuss the current state of deployment, which is itself a moving target, we assume that a system capable of global deployment is globally deployed; this makes it location-independent. Examples of Type 1 systems include Mobile IP, RAM and Ardis.

As mentioned earlier, transparency is a key aspect of mobile systems. If a peer application needs to know about the mobility of a host, then the benefit of that mobility is limited. In particular, systems requiring effort to port applications to them, while supporting mobility, are limited by the lack of mobile transparency.

Typically these systems require gateways to ”spoof” protocols on one side of their networks and use proprietary protocols to convey data over radio channels. The horizontal market acceptance of these systems has been limited due to the effort required to port applications to nonstandard APIs and mobile devices.

In many cases, Type 1 systems have been designed from the bottom up, with an initial design focus on the Physical Layer, which is typically RF-based. Early mobile systems were developed from an RF perspective and customized to minimize radio traffic, using techniques such as canned messages and proprietary protocols at all layers. This effort to optimize the Physical Layer resulted in systems with physical (RF) protocols visible to applications. These systems are poor examples of protocol layering.

A very recent mobile technology development is that of Mobile IP, which is media-independent. Regardless of the underlying physical and MAC layers, Mobile IP provides the capability for two mobile hosts to directly communicate regardless of location. CDPD’s mobility management paradigm is based on that of Mobile IP.

1.8.3 Type 2 Mobility: Transience

Type 2 Mobility describes systems in which in-motion operation is supported in a transparent and location-independent fashion. Type 2 mobility is the ultimate form of anywhere anytime communications capability. An example of such a system is CDPD, which was originally conceived of as a cellular overlay. In this system, the efficiency of mobility management (and radio resource management) is optimized to support active moving hosts which are capable of automobile speeds.

1.9 Range of Mobility

In order to describe mobility in WAN environments, we need a terminology describing range of motion. Mobility consists of receiving service anywhere; movement to another region should result in receiving service in the new region. Although the application running on the mobile host should be oblivious to the regional boundaries (transparency of service), the mobile host itself will likely need to be aware of the boundary crossing and actively participate in it.

The region types we define are a hybrid of conventional data network and cellular terminology; this reflects our need to define mobility in both (network) topological and geographic terms. Since WANs are hierarchical in nature, so are these regional definitions. This is depicted in Figure 1.14.

1.9.1 Channel

A channel is the means by which a mobile host receives service; it is the subnet used by the mobile to access the network. The channel is both a physical media and a logical location (i.e., network access point) of the mobile host. In many cases, such as CDPD, the channel is shared by multiple mobile hosts.

1.9.2 Cell

A cell is defined by the geographical area covered by a channel. This means that a mobile host could potentially continue to receive service from the channel even while moving about the cell.

A cell can be either large or small, depending on the system and the situation; media-specific physical limitations and capacity constraints determine the size of a cell. In a wireless system, a cell is defined by the RF coverage area of a single channel. Depending on the mobile system, the cell might be identical to the mobility area; in this case, geography and network topology would be essentially identical.

Multiple channels could cover a single cell. This would be akin to multiple Ethernet LANs being available to office workers in a building; which Ethernet is used is based on need for logical interconnectivity (e.g., a group of users who work together a lot) or traffic (load balancing)¹⁸ .

Sometimes, depending on the medium used, cells overlap. In such a situation, a mobile could receive service from any of a number of channels. This could be significant from the standpoint of mobility management if the channels covering the mobile’s geographic location are based in different network topological locations. An example of this is the A-side and B-side systems in the cellular industry.

In some systems, the mobile hosts in a cell can directly communicate with one another in a peer-to-peer fashion. In other systems, the mobile hosts in a cell must communicate with one another via some kind of hub. The requirement for a hub is typically due to the limitations of the physical media in use.

1.9.3 Mobility Area

A region controlled (from a mobility perspective) by a single mobility-supporting router is called a mobility area; this is analogous to a routing area in conventional data networks [PERL92]. The mobility router is responsible for accepting network layer packets destined for mobile hosts in its area of control (its mobility area) and forwarding those packets to their destinations.

From a WAN perspective, the mobility router is directly connected to all of the mobile hosts in its mobility area; this is a network layer topological concept. The mobility router is the last entity to handle the NPDUs (in their Layer 3 form) before the destination host; any intervening entity operates at a strictly lower layer. The mobility router is an assisting entity in any mobility scenario.

A mobility area will generally consist of one or more cells. Each cell is contained in its entirety within a single mobility area.

1.9.4 Administrative Domain

An administrative domain in a mobile network is the same as in a conventional data network: a region under the control of a single authority. This is important from the standpoint of accounting, directory services, security and authorization. The administrative domain is the highest-level region for mobility.

An adminstrative domain is under the jurisdiction of a single authoritative body; this body either accepts or rejects a mobile entity requesting services in the domain. The idea is that someone has to pay for and run a network (or subnetwork); the collection of subnets under control of one organization is an administrative domain. An administrative domain is a logical concept and is sometimes referred to as an autonomous system.

An administrative domain will generally consist of one or more mobility areas. Each mobility area is contained in its entirety within a single administrative domain.

1.10 Mobility is not Wirelessness

Our discussion thus far has introduced many issues of mobility which are independent of the media used to access a network. This is not an original idea–groups such as the Mobile IP Task Force have worked on mobility for some time without any media-specific considerations.

From this it is obvious thatmobility is not equivalent to wirelessness. Mobility is the ability to communicate anytime anywhere; this is a topological capability–always being able to ”connect with” another party. Ideally, this connectivity can be maintained regardless of the location or motion of the mobile entity. This location independence should be available over an area which is physically too large for any single medium, such as an Ethernet cable or an RF channel.

A wireless host is a communications end-point¹⁹ which is physically untethered by a communications link or cable; this is a capability of the physical media in use. Obviously, wirelessness enables greater mobility than is possible with wired media, especially in-motion correspondence.

Mobility management is closely associated in wireless systems with radio resource management. Radio resource management is typically concerned with assuring proper (effective and efficient) use of the RF medium and is part of accessing the mobile network. In cellular-type systems such as CDPD, radio resource management could actually be considered to be a highly granular form of mobility management. However, in this book, we shall be very specific about mobility management as a Layer 3 activity.

However, WAN mobility does not require a wireless medium. We can easily conceive of ubiquitous Internet taps sometime in the future which would support mobility but not require wireless access. A user of such a system could take a portable computer from place to place, connecting via readily-available universal network taps to send and receive data; this would comprise a mobile capability which doesn’t involve any wireless technology.

Conversely, a wireless capability in a host does not imply mobility. There are a number of wireless LANs and campus networks which, although free from the physical constraints of cables, cannot be considered to be WANs because of their limited range of operation. Many telemetry applications such as building security systems or utility meters entail wirelessness, but not mobility.

1.10.1 Wireless Considerations

Wireless systems are limited by constraints such as the availability of radio frequencies, licensing from the Federal Communications Commission (FCC) and other regulatory authorities²⁰ , and the expense associated with whatever technology is used to transport data ”over the air.” Radio Frequency spectrum is depicted in Figure 1.15.

The easiest-to-use (and therefore least expensive) radio spectrum has already been assigned to commercial broadcast applications such as television and radio; much of the previously-allocated spectrum is reserved for government use. Higher radio frequencies are available, but require more complex (more expensive) technology and suffer from greater attenuation (and thus, limited range). As we shall see, cost and coverage are key issues for wireless systems.

Some radio spectrum is freely available for use without requiring FCC licensing. This unregulated spectrum is typified by the 902-928 MHz region, commonly referred to as the ISM (Industrial, Scientific and Medical) band. The ISM band is used for applications such as garage door openers, remote controls, home security systems, microwave ovens, etc. Effective use of this unregulated spectrum for data communications requires the use of jam-proof radio technology such as spread spectrum (which is not an inexpensive technology).

The U.S. government has now reassigned and held public auctions for radio spectrum to be used for two-way Personal Communications Systems (PCS). This newly-available spectrum will encourage further advances in the use of wireless technology for data communications. As a result, we can expect continued growth in (wireless) mobile data applications.

Other factors currently limit the ubiquity of wireless data communications, including power control and signal propagation. Power control is a significant issue because the wireless host can operate only as long as its battery will allow. Although battery technology continues to make great strides, battery life is still a critical design factor for wireless systems; techniques such as sleep mode operation have been designed as key system functions which support wireless hosts by extending their battery life.

Signal propagation is always a factor in wireless systems because an RF signal at a given frequency and power level can be reliably received within a limited range of the transmitter. Increasing power levels tends to extend the range of wireless communication, limited by the battery capabilities of the wireless host. Having smaller RF coverage areas helps reduce wireless host battery requirements, but increases the network deployment costs because of an increased number of base stations providing landline connectivity.

All of these considerations are important to the design and deployment of wireless systems. However, they have little to do with mobility.

1.11 Challenges of Mobility

The capability for mobility introduces a number of challenges to conventional wide area data networking. One measure of a mobile system’s usefulness in the real world is the extent to which that system addresses these challenges.

1.11.1 Geography vs. Network Topology

The first challenge of mobility is the mapping of geographic coordinates to network addresses. Data networking is inherently topological in nature: two hosts are either (directly or indirectly) connected or they are not. Physical location of the hosts is immaterial, except for the physical limitations of the medium in use.

A network address, including a netid and a hostid, indicates the topological connectivity of a host (i.e., the subnet defined by the netid) but not its geographic location. Moving a host to a new geographic location might or might not require a change in network address, depending on whether the topological (network) connectivity has changed.

However, mobility is inherently a geographic capability–being able to be anywhere anytime. Since the essential problem of mobility is being able to receive communications anywhere and the routing of data to a host is via network address (netid), there needs to be a mapping of some sort between the geographic location of the mobile host and the network connectivity. This mapping can be either explicit or implicit, but it must be done to get data packets to a mobile host.

Because the geographical to topological mapping is loose and dependent on the systems technology and architecture, movement in the geographic sense may or may not be reflected in the topological sense. Either the system can provide this geographic to topological location mapping or corresponding entities must do this mapping themselves. This is the heart of mobility management.

1.11.2 Part-time Destinations

Another challenge of mobility is the part-time connectivity of mobile entities. This is an issue for any networked host which is frequently unavailable.

Accessing an occasionally-connected host is an issue because it requires effort above and beyond what is normally required in conventional networks. Data networking paradigms today assume the essentially continuous availability of networked hosts to receive correspondence. If a host is unavailable, current applications must either continuously page for the intended recipient or have the temporarily disconnected host poll each of its potential application servers as soon as it rejoins the network.

In conventional networks and LANs, polling is not a problem because there is generally plenty of bandwidth to accommodate this overhead. However, in mobile WANs, which could support millions of users, accessing part-time destinations (which could be anywhere) is an issue! If each mobile host had to poll each of its application servers (accross a wide area) the result would be a system which does not scale well.

Another aspect of part-time destinations is the fact that connectivity is a network layer concern, while correspondence between entities is typically at the application layer. Thus, there needs to be an efficient mechanism for bridging between applications layer entities’ desire to correspond and network layer connectivity.

What is needed in a mobile WAN environment is the capability for efficient storing and forwarding of data in a manner which is transparent to corresponding application entities. This might involve storing and forwarding of messages (application layer PDUs) rather than packets (network layer PDUs).

For example, if two systems need to exchange data but are never operating at the same time, an intermediary system of some kind is needed to facilitate the data exchange. The intermediary store-and-forward entity would first receive the data (e.g., Email) from the source when both are operating. The intermediary would then forward the data on to the destination system when they are both operating.

1.11.3 Moving Targets

Another challenge of mobility is getting data to a potentially moving target. Given a geographic to topological address mapping capability and an efficient store and forward capability, there remains a need for efficient routing of the data to a host which frequently changes location. Routing tables need to be updated more frequently than a mobile host changes its cell location. This challenge is exacerbated in transient (i.e., Type 2) entities which can be in-motion at the time a correspondent is communicating with it.

Some systems address this moving target challenge by having the current mobile serving function notify its predecessor. This solution can be effective, but suffers from the creation of a potentially long trail of MSFs which forward messages from one to another until eventually reaching the current one. This is not unlike the method used by the postal service²¹ or cellular voice systems.

1.11.4 Application Transparency

Another challenge of mobility is transparency. Neither an application on a mobile host nor the peer with which it is communicating should be directly impacted by location-independence. The application should operate and–to the human using it–it should look and feel the same as it would in a non-mobile context. Transparency requires standard Application Program Interfaces or APIs to provide the normal ”look and feel” to applications and the humans using them; as we shall see, lack of standard APIs has limited the adoption of some otherwise appropriate mobile data systems.

The mobility of the host should also be transparent to the rest of the world with which it is communicating. Any entity wishing to communicate with a mobile entity should be able to do so regardless of the target’s mobility.

The ”hourglass” of protocols of Figure 1.16 depicts one way that transparency is naturally provided in internetworking environments. By having a common Layer 3 protocol–typically IP–many applications can be supported independently of the many subnetwork technologies and media available.

The conventional world of data networking is based on a paradigm of routing data packets based on the network address of the destination (host). This paradigm should be unaffected by the mobility of the destination. Whether the host is ”at home” or ”on the road,” the peer attempting to communicate with it should not have to do anything special beyond what normally would be done to communicate with another host. As we shall see, this has many implications in areas such as directory services, routing, security and accounting.

Transparency of mobility has further implications. Since many applications require connection-oriented transport layer services, Type 2 Mobility data systems must maintain end-to-end transport connections even while the mobile host is in motion. This implies that host network addresses must not change to support mobility, otherwise transport and higher layers are impacted.

Finally, [IOAN93] says that: ”an issue that cannot be handled in lower level protocols, is that of coping with intermittent operation, disconnected operation, or operation during which network characteristics such as bandwidth and latency change significantly. Even if the network layer hides the addressing aspect of mobility from higher levels, applications may want to be Ômobile-smart’, and function differently as the service provided by the network link changes (for example, by increasing the granularity of communication).”

1.11.5 Name-to-Address Mapping

Another challenge of mobility is support for DNS-type directory services. The primary function provided by the Domain Name System (DNS) is the translation between host network layer addresses and more user-friendly names. These services allow applications such as Email to use names such as ”joe@BigCompany.com” as destinations rather than the more human error-inducing addresses such as ”joe@123.45.6.789.” If changing geographic coordinates implies changes to network addresses, clearly DNS services cannot work in their current format. This again seems to imply the need for ”permanent” network addresses of some kind for mobile hosts.

1.11.6 Security

There has been much recent discussion about security in WAN environments. The focus of this discussion is the use of firewalls of various kinds [KAUF95] to prevent unauthorized access to networks from the outside. The assumption is that an attack will come from outside of the local subnet; unfortunately, a high percentage of security ”incidents” involves people inside the subnet (e.g., disgruntled employees).

However, mobility of hosts creates new opportunities for compromised network security because the physical security of a subnet is not longer provided. Each time a mobile host establishes a new subnet point of attachment (SNPA), it must authenticate itself to the subnet it is attaching itself to. ( Authentication is the process of verifying a host’s identity and is essential to detection and prevention of clone devices.)

Typically authentication involves the use of a password of some kind. Protection of this password as well as the identity and data of a mobile from potential ”eavesdroppers” requires the use of encryption techniques. This (encryption and security in general) raises the costs of providing mobile data services in terms of network and processing overhead, royalties (for the security technology), key management, etc. All of this needs to be embedded within a mobile system, not a later add-on.

Security issues in mobile data networks are discussed in Chapter 6.

1.11.7 Scale

Mobility in WANs raises concerns of scalability to higher levels than ever before. In a mobile WAN environment one could expect millions of hosts and routers.

However, current routing update protocols (e.g., RIP) fail with much smaller numbers of hosts. The need to exchange mobile routing information across the network more frequently adds significantly to network overhead. Propagation of routing information takes longer the larger the network grows (e.g., larger routing tables); so does the amount of computation required at each router (to determine the next hop for a packet).

Additional concerns for authentication and accounting for system usage exacerbate the scalability concern. In a mobile WAN, where mobile hosts can suddenly appear anywhere, rapid access to authentication data is essential. However, replication of the data to enable rapid access raises concerns in areas such as distributed database consistency, etc.

1.12 Summary

In this chapter we have provided an overview of mobility in the WAN environment and issues which result from mobility. For the most part, these issues are due to inherent conflicts between conventional WAN technology and mobile systems. Mobility directly challenges many of the underlying assumptions of the conventional WAN world, most notably the relative stationarity of host location. Historically, mobile systems arose from the connection-oriented paradigm of telephony. The following chapter describes cellular systems, the archetype of the mobile world.

Chapter 2
Introduction to Cellular Systems

In this chapter we introduce cellular technology, the most pervasive mobile communications technology. Although cellular systems have historically been voice-oriented, much of the underlying technology provides a model for mobile data systems such as CDPD. As we shall see, the application (voice) has driven the design of cellular systems in a way which is less than optimal for data applications.

This chapter presents topics such as cellular radio transmission, cellular capacity, the North American Advanced Mobile Phone System (AMPS) voice and data services, digital cellular and the emerging Personal Communications Services (PCS). Other cellular-based systems which are not ”common carrier” systems (e.g., two-way paging) are described in Chapter 9. Although it provides an overview of cellular technology and the cellular industry, this chapter is by no means a complete portrayal–several references such as [CALH88], [LEE-89], [LEE-93] and [PAHL95] provide a much more complete presentation of a fascinating technology.

2.1 The Ubiquity of Cellular

Rapid and accelerating growth of the cellular telephone industry over its first 12 years has resulted in extensive coverage of populated areas by cellular services. Cellular is now the dominant two-way mobile communications technology, with more than eighteen thousand cell sites covering over 95% of the U.S. population and serving over 33 million subscribers –a 13% market penetration–at year-end 1995, as depicted in Table 2.1.¹

This extensive coverage provided by cellular voice services would seem to make cellular the ideal medium for providing ubiquitous wireless data services. Unfortunately, the cellular industry’s voice heritage impacts its ability to support data applications. Cellular’s circuit-switched orientation and radio channel characteristics conflict somewhat with the needs of data applications.

Cellular systems have followed the traditional circuit-switched channel model of telephony. In this model, the end-to-end circuit, including the cellular channel, is dedicated to a single user or application before they can transmit on the channel. The channel remains dedicated to the user or application for the duration of the transmission, until it is explicitly released.

A single dedicated channel per user may be suitable for voice applications albeit somewhat inefficient from a channel perspective. However, a dedicated channel per data user is extremely inefficient and thus prohibitively expensive, unless the data application involves bulk transport of large quantities of data² or the application is of a high-performance mission-critical nature³ . In support of the circuit-switched nature of cellular, billing systems have been oriented toward billable units of time on the order of minutes of airtime, rather than quantities which better match the activities of data users.⁴

The characteristics of the radio channel used by cellular systems have also challenged data applications attempting to use the cellular channels. As we shall see, these radio characteristics are further exacerbated by call control messages which are transmitted in-band while the call (data transmission) is in progress.

2.2 Radio Channels

Cellular channels areradio frequency (RF) channels. In an RF-based system, any receiver within range of a transmitter, i.e., within its coverage area, can tune to the frequency in use by the transmitter and, with the proper demodulation and decoding, capture the information transmitted.

However, two or more nearby transmitters simultaneously using a common frequency or channel ⁵ will interfere with one another, unless perfectly synchronized in both timing and content. A receiver within range of both transmitters will most likely receive a garbled message, which is undecipherable.

This potential for interference limits the capacity of any RF-based system. Like any other system employing a shared medium, such as an Ethernet LAN, only a single device can transmit at a time. The greater the number of devices sharing the physical channel, the less often each can transmit. The only difference between an RF-based system and a LAN is the scope of the shared medium; the RF-based system is not as physically bound, thus the opportunity for interference is greater.

This RF capacity constraint increased the expense to users of early mobile phone systems such as Mobile Telephone System (MTS) and Improved Mobile Telephone System (IMTS). These systems typically broadcast all channels from a single antenna location. Within a city covered by one of these systems, the number of simultaneous users was limited by the number of RF channels available to the system; each RF channel could be used by only one transmitter at a time.

2.3 The Cellular Concept

RF bandwidth has always been the primary constraint in wireless systems; there is never too much. Efficiently using this precious resource involves what is called frequency reuse, in which a radio channel is allowed to be simultaneously used by multiple transmitters as long as they are sufficiently separated to avoid interference. The essential idea of cellular radio is to transmit at power levels sufficiently low so as to not interfere with the nearest location at which the same channel is reused.

In this way a physical (RF) channel can be used more than once in a given city. The greater the reuse distance, the lower the probability of interference. Likewise, the lower the power levels used in cells sharing a common channel, the lower the probability of interference.⁶ Thus, a combination of power control and frequency planning is used in cellular systems to prevent interference.

The unit area of RF coverage for cellular is called a cell.⁷ In each cell, a base station transmits from a fixed cell site location, which is often centrally located in the cell, to mobile stations or subscriber units. The base station and mobiles are allowed to use a subset of the RF channels available to the system. These channels cannot be reused in any potentially interfering cells.

Base stations are supported by and interconnected to each other and the public switched telephone network (PSTN) via mobile switching centers (MSCs), as depicted in Figure 2.1 The operation of AMPS systems has historically been based on intelligent MSCs controlling the operations of the base and mobile stations. Cellular mobility management is handled by home location registers (HLRs) and visiting location registers (VLRs), described in Section 2.7.7.

Cellular system capacity or spectrum efficiency can be most easily and inexpensively increased by subdividing cells into smaller cells⁸ or by sectorizing the cells. Sectorization consists of dividing an omnidirectional (360 degree) view from the cell site into non-overlapping slices called sectors, which when combined provide the same coverage but are considered to be separate cells.⁹ This trend has continued with the creation of microcells, which are aimed at increasing capacity in areas of dense user populations¹⁰ . While cells typically range in size from two to twenty kilometers in diameter, microcells range from about a hundred meters to a kilometer in diameter.

The capacity gain provided by cellular systems is offset somewhat by loss of trunking efficiency, which is the queueing efficiency resulting from a large number of customers receiving service from a set of servers rather than proportionally assigning each customer to one of the servers.¹¹ If a disproportionate number of mobile stations are simultaneously located in a single cell, a cellular system might actually end up supporting fewer users than a wide area radio system. Because relatively few of the users who are aggregated in the cell can receive service (due to the fact that only a subset of channels is available in the cell), the cellular system could appear ineffective. If the cell can only support m channels, the (m+1)st simultaneous user could be blocked from receiving service.¹²

So there is a trade-off: an n-cell frequency reuse scheme, in which RF channels can be ”reused” every n cells, provides better channel quality the larger the value of n (due to reduced opportunities for interference).¹³ However, an n-cell frequency reuse scheme allows only 1/n of the total number of channels to be available in each cell, which greatly increases the probability of blocking for a user trying to access the system. Sectorization is actually more reuse efficient in that a smaller number of cells are needed in the reuse pattern, each providing a larger fraction of the total frequency spectrum. Typical values for n are 7 for sectored cells (typically partitioned into three sectors, as in Figure 2.2) or 12 for omnidirectional (non-sectorized) cells.

Frequency planning–the assignment of channels to cells–can be static or dynamic. A static assignment of channels to cells and sectors is referred to as fixed channel allocation or FCA. FCA has historically been used by cellular service providers in their frequency plans and results in each cell having a fixed capacity for serving mobiles. The maximum number of simultaneously-transmitting mobile stations is equal to the number of channels statically assigned to the cell.

A more recent technique for channel assignment is called dynamic channel assignment or DCA. With DCA there is no fixed association of channels to cells. Each of the channels available to a cluster of cells could be used in any cell or sector within the cluster as needed. DCA eliminates the need for up-front frequency planning and provides the ultimate flexibility for capacity. However, DCA requires processing and signalling to coordinate dynamic channel assignments and avoid interference. It is really frequency planning on the fly.

2.4 Cell Handoff

One of the goals of a cellular system is for a user to remain ”in touch” even as they move through the system. When a user moves from the coverage area defining one cell into that of another, the system must provide the capability for that user to remain ”in touch” even while breaking the connection with one base station and establishing another connection with another base station. This operation is called a handoff. Smaller cells means more frequent handoffs, which requires greater system resources to support and coordinate. Handoff is really a localized form of mobility.¹⁴

Cellular handoff is done in one of two ways, as shown in Figure 2.3. Hard handoff is when the airlink connection between the mobile and its initially-serving base station are momentarily severed before reconnecting with a new base station. This is the method traditionally used in existing cellular systems, because it requires the least processing by the network providing service. However, it causes a momentary interruption in reception which is sometimes noticeable to the humans engaged in the call being handed off.

The second handoff mode is called soft handoff, in which two base stations are briefly simultaneously connected via the airlink with a mobile during the handoff. As soon as the mobile’s RF link with the new base station is acceptable, the initially-serving base station disengages from the mobile. Diversity techniques are employed at both ends of the radio link to ensure a smooth handoff, which is largely undetectable to the humans affected.

Handoffs can be further categorized as being either controlled or assisted by the network or the mobile. A network-controlled handoff is referred to as abase-controlled handoff or BCHO.¹⁵ Mobile-controlled handoff or MCHO is less commonly used in voice systems, although it is used in CDPD because of the burst mode of transmission employed by CDPD mobiles. Second generation cellular voice systems take advantage of greater intelligence in mobile stations and time-division techniques to perform mobile-assisted handoff or MAHO, in which the mobile participates but does not control the handoff.

Whichever technique is employed, the handoff process is complex. A decision algorithm is used to determine when the handoff should occur, based on factors such as the received power level and signal quality (bit error rate or supervisory tone). Once predetermined threshold values have been exceeded, indicating that the edge of cell coverage has been reached, another decision must be made–where should the mobile next receive service?

The target cell for the handoff is determined by RF measures designed to minimize interference coupled with capacity considerations such as the need for load balancing, availability of idle channels, etc.¹⁶ All of the decisions for a handoff must be made quickly, because the subscriber could be traveling at highway speeds. This need for rapid handoff decision-making is accentuated by the ever decreasing cell sizes used in urban areas. The requirement for rapid handoffs can only be met with a sufficient level of processing and signalling capability.

The ability of a network to support cell handoffs can be a capacity constraint. Therefore, it is important to avoid unnecessary and undesirable handoffs. The system (network plus mobiles) must distinguish between an actual movement from one cell’s coverage area to another and a mobile simply moving to a fringe area, where the RF reception is poor. Smart algorithms involving timers, power control and hysteresis¹⁷ have been developed to reduce the number of unnecessary handoffs.

2.5 Cellular Channel Quality

One of the physical measures of RF channel quality is the carrier-to-interference or C/I ratio¹⁸ . This ratio is logarithmically proportional to the signal quality enjoyed by the receiver of the signal.¹⁹ The larger the C/I ratio, the better the channel quality.

C/I ratios of 17 dB are ideally used to determine the edge of coverage for a cell. If the measured C/I falls below this level, the mobile should be in the coverage region of another cell and a cell handoff should be performed. The interior of the cell should provide C/I ratios which exceed 17 dB, unless the mobile is located in an RF coverage ”hole.”

There are two kinds of RF interference possible: cochannel and adjacent channel. Either of these forms of interference can occur if the cellular frequency reuse scheme is inadequate (i.e., the ”n” in n-cell reuse is too small for the geography available and power level in use).

Cochannel interference results when two transmitters within range of a common receiver use the same channel (frequency) simultaneously. The receiver will receive a combination of the two signals and will be unable to make any sense of the combined signal.²⁰ In this case, the two channels can be said to have interfered with one another.

Adjacent channel interference occurs when two transmitters within range of a common receiver use adjacent channels (i.e., neighboring frequencies) simultaneously. Because the physical characteristics of the RF channel causes some spill-over of the signal into neighboring frequencies, adjacent channel transmissions could interfere with one another.

Because of the interference caused by the simultaneous transmission on cochannels or adjacent channels within a cell, only one can transmit at a time. The earliest mobile phone systems used the same channel for the network and the mobile devices. This half-duplex mode of operation greatly hindered the efficiency of these early mobile RF systems.

Current cellular systems are full-duplex. This is accomplished by using different physical channels in the forward channel direction (i.e., network to mobile) and the reverse channel direction (i.e., mobile to network). The channels used in each direction are sufficiently separated in the frequency domain (they are separated by 45 MHz in current AMPS systems) so as to prevent interference. Of course, full-duplex operation means that the mobile and the base station must each have two RF transceivers (i.e., one transmitter, one receiver) simultaneously engaged.²¹

Even if no other transmitters are causing interference, a received RF signal can be garbled due to a phenomenon known as multipath orRayleigh fading. This form of self-interference occurs when multiple out-of-phase copies of the same signal destructively interfere with one another due to reflections of the signal off of natural or man-made surfaces.²² In a fading situation, the reflected signal is delayed sufficiently that it is out-of-phase enough to interfere with the direct line-of-sight path. Multipath fading can occur when the mobile is stationary or in motion.

2.6 Power Control

An important part of radio resource management is controlling the power levels used by transmitters. This power control is important because even in the best conditions the received power level is inversely proportional to the distance from the transmitter. Without power control, nearby mobiles could overwhelm transmissions from distant mobiles at the base station transceivers.

The domination by a nearby transmitter can prevent distant transmitter signals from ever being detected at the receiver. Even more pernicious is the so-called near-far or hidden terminal problem. This occurs when one transmitter is much closer to a receiver than another transmitter. If the nearby transmitter signal is captured successfully by the receiver, the receiver might acknowledge the successful reception of the signal. The distant transmitter might then incorrectly conclude that its signal was successfully received. Undetected errors are quite undesirable.

A base station must prevent nearby mobiles from transmitting with power levels that overwhelm other mobiles in the cell. The received signal strength (RSS) at the base station should be approximately the same for all mobiles in the cell. Often power control algorithms are based on the so-called reciprocity of RF signals (i.e., the RSS in one direction is the same as in the other direction if both transmitters are at the same power level).²³ Of course, the base station can always direct individual mobiles to use another power level. Power control is especially important at cell boundaries, to reduce the number of unnecessary handoffs and avoid interference.

2.7 Advanced Mobile Phone System (AMPS)

The Advanced Mobile Phone System is one of the earliest commercial cellular systems. AMPS technology is currently deployed throughout North America and AMPS-derivative systems are deployed in a majority of worldwide cellular markets.²⁴

AMPS was invented at Bell Labs and initially deployed in the U.S. in the early 1980’s. Ownership of the local cellular service operations was transferred from AT& T to the regional Bell operating companies (RBOCs) at the time of AT& T’s divestiture in January, 1984. Other landline telephone service providers, such as GTE, were unaffected by the divestiture and retained their own cellular operations.

To protect the consumer from potentially anti-competitive behavior by the local telephone service providers, government authorities mandated a duopoly structure for the fledgling cellular industry. This duopoly structure for cellular services has been largely imitated by other nations and has resulted in fierce competition between the service providers in many of the 734 markets defined by the FCC.

At the beginning of the cellular industry, local telephone companies (including the RBOCs) were automatically granted one of the two licenses in each market in which they provided wireline service. This is the so-called ”B” license, which can be remembered by the initial of the word ”Bell”.

The second license for each market was initially drawn by lottery and later auctioned. Initially few investors perceived their value–after all, who could compete against the local telco? Some entrepreneurs²⁵ quickly recognized the potential of these licenses and obtained as many as possible by buying out other license holders. The early days of cellular are reminiscent of the gold rush days, with pioneers rushing to buy controlling interests from lottery winners. Thus, the ”A” side, which can be remembered as ”A” for ”Alternate,” was born.

Early deployments and business deals in the cellular arena were based more on intuition than analysis. An early market analysis conducted at Bell Labs in the late 1960’s concluded that the entire nationwide cellular market would peak at about 900 thousand users. Despite analyses of this nature, pioneers were willing to bet that cellular would prove to be popular.

Over time, as the value of cellular licenses were more widely recognized, prices were driven to extreme levels. It became the accepted custom to value licenses on the basis of (potential subscriber) population or ”POPS.” The price of a cellular market is now evaluated in terms of ”dollars per POPS” normalized value, with high water marks in the neighborhood of $ 500 per POPS.²⁶

Cellular markets are defined by cellular geographic statistical areas or CGSAs. Of the 734 CGSAs comprising the U.S., 306 are in metropolitan areas and are called metropolitan statistical areas or MSAs. The remaining 428 are called rural statistical areas or RSAs. MSAs are valued more highly because of their greater density of potential subscribers.

The following subsections describe AMPS, the most widely used cellular technology.

2.7.1 AMPS Channels

The frequencies allocated to AMPS by the FCC range between 824 to 849 MHz in reverse channels (mobile to base) and 869 to 894 MHz in forward channels (base to mobile). As displayed in Table 2.2, they are not contiguous blocks because the initial 40 MHz allocation by the FCC was later extended by 10 MHz when the service’s popularity became evident. There are now a total of 416 channels available in each direction, numbered from 1 to 1024 with gaps in the numbering.

Each physical channel is 30 kHz wide and is dedicated to a single mobile station for the duration of the call while the mobile is in the current cell. Each call uses a dedicated forward channel paired with a dedicated reverse channel at a 45 MHz offset. Some of the channel pairs (21 of them) are used for control purposes in the AMPS environment. Analog frequency modulation (FM) with 8 kHz deviation is used in the traffic channels, which convey voice conversations. Binary frequency shift keying (FSK) at 10 kbps–a digital modulation technique–is used in the control channels used for signalling.

AMPS is an analog FM system, with all of the associated ramifications of such a system. AMPS channels are insecure–anyone with a channel scanner can listen to unsuspecting AMPS users.²⁷ AMPS channels can suffer from interference, which sounds like static to a user; analog signals suffering from multipath fading cannot be corrected. Finally, AMPS radio resource management is based on signal strength (rather than C/I), which can only be measured indirectly via supervisory audio tones or SAT.

2.7.2 Roaming

Since no cellular service provider covers the entire country, carriers must provide service to one another’s customers for those customers to be able to receive service whenever they are outside of their home area. This capability to receive service while in another service provider’s domain is called roaming. Intercarrier business agreements and network to network interoperation (messaging) are essential to support roaming.

The IS-41 standard has provided the technical solution to roaming between networks implemented by different equipment manufacturers. Prior to IS-41, all signalling between systems was proprietary in nature and the roaming capability had to be manually administered. In early years, intercarrier business relationships sometimes abused the customers’ need for roaming, with service providers sometimes surprising subscribers with excessive ”roaming charges.” This has cost the cellular industry much in terms of reputation and customer relations.

Despite these early business foibles and technical incompatibilities, the cellular industry is rapidly moving toward universal service, with more reasonable roaming agreements in place between service providers. Service providers who are extremely competitive with one another in some markets must simultaneously be extremely cooperative with one another in other markets.²⁸ This is becoming more important as reduced-size mobile stations encourage wider roaming.

2.7.3 AMPS Cellular Operation

AMPS cellular operation consists of call origination and call termination procedures, supported by radio resource management and mobility management functions. It is important to remember that AMPS was designed as a voice-only system, which impacts how these processes are handled. Data transmission on AMPS systems is based on this circuit-switched mode of operation and is described in Section 2.8.

When an AMPS mobile station powers up, it searches through up to 21 predefined control channels. These control channels are physically no different than AMPS traffic channels, except for how they are used–for control purposes only. Each cell utilizes a forward control channel to continuously broadcast information needed by the mobile station for registration. This information includes the system identification or SID of the MSC, which allows the mobile to know whether it is roaming.

An AMPS mobile station finds the best forward control channel it can receive (in terms of received signal strength) and announces itself or registers to the serving network via the matching reverse control channel. From that point on, the mobile remains in a passive state tuned to the control channel it selected. When the channel quality degrades (radio resource management determines this), a call event occurs or the mobile crosses a boundary between location areas,²⁹ the mobile again signals the network. This receive-only mode reduces the traffic on the reverse control channel, a shared resource.

In communicating with the network, the mobile provides two identifiers for registration, call control and validation. The first of these identifiers is the mobile identification number or MIN, which is the programmed handset phone number used to call the subscriber. This programmed identifier is associated with the subscriber and is stored in erasable non-volatile memory in the handset.

The second identifier is the electronic serial number or ESN, which is a manufactured characteristic of the mobile unit. This identifier is (in theory) permanent and associated with the physical equipment. It is 32 bits in length, with the first 8 bits identifying the manufacturer.

Both the MIN and the ESN are transmitted unencrypted by both the mobile and the network. Simple scanning receivers can be used to capture these values, which has provided many opportunities for fraudulent use of cellular services. Recently the cellular industry has instituted a subscriber-entered personal identification number or PIN as an escalation in the war on cellular fraud. But this measure has proven to be only a temporary complexification for the ”bad guys” and in early 1996 cellular service providers began deploying authentication mechanisms.

2.7.4 AMPS Mobile Call Origination

The mobile originates a call (following its owner’s depression of the ”send” button) via the reverse control channel in the cell the mobile is currently located in. The mobile ”knows” which control channel to use by information broadcast by the network on its selected forward control (paging) channel.

Access to the reverse control channel is by a CSMA³⁰ -type scheme. The mobile simply transmits its request (which includes information about the subscriber such as the MIN and ESN) and ”listens” on the forward channel for its subsequent channel assignment. The base station forwards the request to the MSC.

After validating the mobile (i.e., does the subscriber pay their bill and do we have a business/roaming agreement with the subscriber’s home service provider?) via the HLR and the VLR, the MSC selects a traffic channel pair for the mobile. If no channels are available, the MSC simply rejects the request (which results in the mobile producing the annoying ”fast busy” audible signal).

If the MSC grants a channel for the subscriber, it must then connect the call through to the destination. This is done via standard telephony procedures–the MSC simply appears to be a private branch exchange (PBX)³¹ to the PSTN. The channel grant message is relayed to the mobile via the forward control channel.

The mobile then tunes its transmitter and receiver to the assigned traffic channel pair for the duration of the call. Call and power control from this point forward are handled in-band on the AMPS traffic channel assigned to the mobile.

2.7.5 AMPS Mobile Call Termination

When an AMPS mobile is not engaged in a call, it monitors the forward control (paging) channel. A call attempt directed at the mobile (i.e., to the MIN assigned to the mobile) is received by the mobile as a page on the control channel. The page is repeated several seconds later, in case the mobile was temporarily in an RF ”hole” or otherwise unable to receive the first page. The time interval between pages is short to minimize the ringing delay experienced by the originator of the call.

The mobile responds to the page via the reverse control channel and awaits the traffic channel assignment. The mobile response is also repeated, in case the initial response collided with another mobile on the reverse control channel or suffered from bad RF conditions.

When the mobile receives the traffic channel assignment from the network (the MSC via the base station), it proceeds to that channel and produces an audible ringing tone for the subscriber. From this point forward, all further signalling between the system and the mobile is conducted in-band.

2.7.6 AMPS Radio Resource Management (RRM)

AMPS channels are controlled by the MSC. The traffic channel assignment process was described in the preceding subsections. However, there are other aspects of RRM, including power control and handoff.

Power control is handled by monitoring the received signal strength of the reverse channel at the base station, which in turn passes this and other channel quality information to the MSC. The MSC evaluates this data, including a trend analysis, to determine whether the mobile should increase or reduce its power level or be handed-off to another cell. AMPS defines eight power levels in 4 dB steps. This power level control is a means of controlling the local access point to the network for a mobile station.

Cell handoff is handled in a BCHO manner, as discussed in Section 2.4. The system controls handoffs by transmitting the SAT in-band on the forward channel. This tone is filtered by the mobile–it is outside the range of the audible channel–before reaching the subscriber’s ear, and is reflected back to the system in-band on the reverse channel.

The base station filters the reflected SAT and evaluates the quality of the reflected tone. The base station forwards SAT quality information on to the MSC. Based on RSS and SAT data, the MSC determines whether or not to initiate cell handoff procedures.

Cell handoff procedures include having neighboring cells’ base stations monitor the mobile’s reverse channel and evaluating the received signal strength. If another base station ”hears” the mobile better than the current base station, the mobile is instructed to move to a new channel pair via a ”blank and burst” message transmitted in-band by the base station. The mobile then tunes its RF transceivers to the channel pair instructed. All of these steps are orchestrated by the MSC.

The ”blank and burst” message³² sounds like static to the ear of the subscriber and is momentarily disruptive to the conversation taking place. It is also highly destructive of any data transfer which could be occurring at that point in time via modems on the cellular channel. This is one of the reasons that cellular has historically been a harsh environment for mobile data users.

Intelligent algorithms are used to prevent unnecessary and premature handoffs, especially for non-moving mobiles, mobiles located in poor in-cell coverage areas, mobiles traveling along the border between cells and situations in which no cellular channels are available beyond the cell’s boundary.

2.7.7 AMPS Mobility Management

Mobility management in AMPS networks appears to be based on the engineering assumption that most calls are originated by the mobile and seems optimized for mobiles that are usually in their home area.

Intelligent paging algorithms are employed by AMPS to reduce the collective forward bandwidth required for a page. The paging algorithm starts by paging in only a small area, based on where the mobile usually receives service (the home area) or perhaps where the mobile was last registered (i.e., the location area).³³ When a mobile receives a page, it responds and proceeds to the assigned traffic channel pair.

AMPS mobility management has been greatly enhanced by IS-41, which defines the standard for interoperation between networks. Mobility management is handled by databases known as the home location register or HLR and the visiting location register or VLR.

The purpose of the HLR is to track the location of (the VLR serving) a mobile. The HLR also contains information about each of the mobiles associated with a home area (e.g., is the mobile allowed to originate an international call, etc.). This database logically unites data describing both the subscriber and the subscriber’s equipment into a single service profile. Because this ”permanent” information is critical to serving customers, the HLR function is typically supported by multiple distributed fault-tolerant computers.

The purpose of the VLR is to track all mobile stations currently roaming in the local MSC coverage area. The VLR contains the service profile of each roaming mobile as well as other information necessary for calls terminating at the mobile station. Because the VLR function is so closely aligned with an operating MSC, it is typically collocated with or part of that MSC. Since the information it contains is of a temporary nature, fault-tolerance is less critical than for the HLR.

Because base stations periodically broadcast the SID and location area identifiers on the forward control channel, the mobile station knows immediately when it has roamed into another system or location area. An option in IS-41, known as autonomous registration, allows the mobile to register to the host MSC. This registration is forwarded by the MSC to the VLR. Another IS-41 message is then used by the VLR to notify the HLR that it is currently hosting the mobile.

The HLR passes necessary service profile information to the VLR in another IS-41 message, enabling the host system to provide service to the mobile station. Information such as whether or not the mobile is allowed to originate international calls is contained in this service profile. Since the HLR now knows the location of the mobile, more efficient paging can be used for mobile-terminated calls. The HLR also notifies any other VLR which had been previously hosting the mobile to deregister the mobile.

A mobile-terminated call attempt is always first directed to the mobile station’s HLR by the gateway MSC first contacted from the PSTN. The HLR is responsible for contacting the current serving system, obtaining a temporary local directory number (TLDN) from the serving system, and transferring the call to the TLDN. The serving system is responsible for the connection between the TLDN and the roaming mobile station.

IS-41 supports uninterrupted voice services while the mobile station moves between MSCs. This is equivalent to maintaining a session between a mobile data device and another host while the mobile host is in motion between areas. Because trunk connections are used to carry the voice traffic, the concatenated trunk length could grow as the mobile moves about while the conversation is active (i.e., the mobile ”grows a tail”). This is usually not a significant problem because most conversations are only a few minutes in duration and even fast-moving vehicles covers only so much ground in a typical call period.

HLRs and VLRs are not affected by cell handoffs, only by wider-scale mobility (between MSCs). Mobiles reregister every time they cross boundaries separating location areas. These location areas are clusters of cells large enough to minimize the number of re-registration messages (on the contended reverse control channel) while also minimizing the number of paging channels involved in a page.

2.8 Data Transmission via AMPS

The native AMPS environment is harsh for data transmission. RF modulation and coding techniques, such as vocoders, ADPCM³⁴ trunking compression and FM pre-emphasis–which are optimized for voice transmission–distort data and disallow the use of standard (landline) modems on cellular channels. Cell handoffs, channel reassignments and power level change commands are all transmitted in-band by the system, further distorting the data channel. Finally, static, signal fading and interference make cellular a noisy channel for data applications. Standard landline modems typically react by either losing data or hanging up.

Despite these challenges, the ubiquity of AMPS analog cellular coverage and demand for wireless data services have motivated development of technology supporting cellular-based data services. Today, circuit-switched cellular data is the most widely used mobile data service.

Special modem technology has been developed by AT& T Paradyne (Enhanced Throughput Cellular or ETC), Microcom (MNP10), Motorola (EC2) and others to optimize the capabilities of AMPS channels for data transmission. These enhanced cellular data protocols allow approximately 9 kbps under normal conditions, with 14.4 kbps becoming increasingly available. Data compression also increases the effective throughput enjoyed by cellular data users.

Modem pools supporting these new protocols are now being deployed by cellular service providers. These modem pools provide a gateway function bridging the specialized cellular modem protocols and standard landline modem protocols. A special code is added to the dialed digit string at the mobile, which alerts the cellular switch to connect the call to the modem pool rather than simply placing a call. The gateways allow continued interoperability of AMPS with conventional modems.

The modem pool concept has been extended with backbone packet services offered by AT& T, MCI, Sprint and others. Typical of these offerings is MCI’s Xstream Air Network, depicted in Figure 2.4, an X.25-based backbone supporting cellular modem-based data applications. Access to Xstream is via third party cellular service providers. Special 800-number access is available from anywhere to the MCI cellular modem pool, which supports both MNP10 and ETC protocols.

Another data solution for the cellular airlink is a single-sided protocol such as Air True by Air Communications, Inc. This protocol only requires support at the transmitting side (presumably the mobile) of the airlink to effectively counter the debilitating effects of the cellular environment. It recognizes network event (call control) messages for cell handoff, power control and channel changes for what they are rather than interpreting them as random noise. After the interrupting network event message has completed, the mobile transmitter resumes where it left off, increasing the effective bandwidth.

Channel characteristics are not the only challenge for AMPS-based data applications. Current cellular systems are voice-oriented and thus have usage accounting mechanisms, which are based on time of usage rather than actual data transmission. The time-of-usage billing schemes typically begin with a minimum usage of one minute. A significant departure from previous per-minute billing practices is embodied in the UPS package tracking system, supported since 1992 by a large number of cooperating cellular service providers.

In 1995, Bell South Wireless, Inc., announced a wireless telemetry solution named Cellemetry, which would be licensable by at most one cellular service provider per market. Cellemetry operates over the AMPS control channels and thus is limited by the amount of data which can be transported by cellular registration and paging messages (32 bits) and the amount of additional traffic which can be borne on the control channel without impacting the primary purposes of these control channels–cellular registration and paging.

2.9 Digital Cellular Technologies

Digital technology offers the opportunity for improved transmission in cellular systems. This is due to powerful error detection and recovery techniques, which can be used to counter the debilitating effects of noise, fading and interference. Digital technology also provides the basis for security in the forms of encryption and authentication. Finally, digital technology requires less in the way of mobile transmit power, which increases battery life in portable mobile units.

Digital cellular technologies also offer the promise of effective data transmission via cellular services. Although their vocoders prohibit the use of conventional modems, recent extensions to standards provide low-throughput data traffic in either a circuit-switched mode or via a digital control channel. Packet-switched data services are also being developed by the proponents of digital cellular standards.

However, the primary motivations for the digital cellular standards are unrelated to data. Development of the North American digital standards was motivated by the need for increased capacity in light of the 40-plus percent compounded growth rate in AMPS penetration during the 1990’s.³⁵ Overseas, development of the GSM standard was motivated by the desire to unify cellular service across European national boundaries.³⁶

Once the commitment to digital cellular voice standards was achieved in the various standards bodies, it was quickly recognized that digital services could include much more than mere capacity enhancement. Data applications, secure channels and enhanced voice services such as caller identification are now possible with the new digital standards.

Before presenting the primary digital cellular technologies, understanding the basic differences between FDMA, TDMA and CDMA is essential. As depicted in Figure 2.5, a frequency division multiple access (FDMA) system, such as AMPS, separates individual conversations in the frequency domain–different conversations use different frequencies (channels). In this depiction, the frequency domain is represented by the vertical dimension and the time domain is represented by the horizontal dimension.

Figure 2.6 shows how time division multiple access (TDMA) systems, such as IS-54/136, GSM or PDC, separate conversations in both the frequency and time domains; each frequency (channel) supports multiple conversations, which use the channel during specific timeslots. Typically there is a maximum number (3 in the example) of conversations which can be supported on each physical channel. Each conversation occupies a logical ”channel.” TDMA systems are discussed in Section 2.10 and Section 2.13.

Figure 2.7 shows how frequency-hopping code division multiple access (CDMA) systems, such as spread spectrum wireless LANs, separate conversations in both the frequency and time domains. By rotating conversations through frequencies (channels) on a synchronized basis, each conversation experiences a variety of channel conditions.³⁷ This rotation through the frequency set also tends to reduce the interference levels. These systems are discussed in Chapter 9.

Figure 2.8 shows how direct sequence CDMA systems, such as IS-95, separate conversations on the basis of something entirely different than frequency or time. It’s hard to show in a time versus frequency diagram, but we will discuss it in Section 2.14.

2.10 Europe: GSM and DCS 1800

Definition of the Groupe Special Mobile (GSM) system began in 1982, under the auspices of the Committee of European Posts and Telecommunications. Now called Global System for Mobile communications,³⁸ the goal of this time division-based digital cellular system was a unified pan-European system. In mid-1995 there were over 11 million customers using GSM worldwide; that number was expected to double by year-end 1996 with more than 140 service providers in 86 countries.

Prior to GSM, many independent analog systems were in use throughout Europe, with incompatible standards preventing intercountry roaming in many cases. Cellular usage was essentially regional in scope. The goal of GSM was to eliminate this fragmentation of the European cellular market. Since this was intended to be a ”next generation” system, it uses digital technology with the capability of supporting data applications.

GSM was originally specified in the 900 MHz band and currently runs in that spectrum (see Table 2.3). However, in 1989 the U.K. Department of Trade and Industry defined Personal Communications Network (PCN) ³⁹ to consist of GSM operating in the 1.8 GHz band. This upbanded system is referred to as Digital Cellular System 1800 (DCS 1800), with deployment anticipated by 1998.

GSM is a TDMA-based system with 8 user timeslots per frame in a 200 kHz channel. Like other TDMA systems, staggered transmit and receive timeslots allow modems to use half-duplex radios, thereby reducing their costs. The transmit/receive offset still leaves enough idle time for the mobile to participate in handovers⁴⁰ by monitoring neighbor cell channel signal strengths in a MAHO scheme, as depicted in Figure ??.

The GSM system uses GMSK⁴¹ modulation. Rate 1/2-convolutional coding with interleaving⁴² addresses Rayleigh fading. The net data rate⁴³ is 22.8 kbps with error correction in what is called full-rate mode. An additional half-rate mode at 11.4 kbps is also defined by using 16 timeslots (which are half as large as the full-rate timeslots) per frame.

A form of slow frequency hopping is used by GSM to help combat the multipath burst errors characteristic of cellular environments. Each base station has its own pattern for hopping from one carrier frequency to another from slot to slot, with mobiles using that base station following suit. This frequency hopping also reduces the incidence of cochannel interference between clusters of cells.

GSM, with slow frequency hopping and coding requires an approximately 9 dB C/I ratio for effective operation. If we assume a frequency reuse factor of 3 with 3-sector antennas and 8 users per 200-kHz bandwidth, we can estimate relative system capacity for GSM to be approximately

or 2.8 times AMPS capacity [FALC95].⁴⁴

The radio data link layer is based on a LAPD-like protocol called LAPDm. LAPDm modifications (from LAPD) include using no frame flags or bit stuffing, instead relying on a ”length indicator” field, as depicted in Figure 2.9. Also, the SAPI (SAP identifier) is included in the address field, shown in Figure 2.10. LAPDm also has no CRC bits for error detection, instead relying on lower layer block and convolutional coding for error detection and correction.

GSM mobility management is provided by specific layer entities which establish, maintain and release separate connections between the mobile station and the MSC under control of the higher connection management sublayer. These separate connections are for call control, short message service and the call-independent supplementary services. Each mobility management connection provides services such as encryption and authentication.

GSM mobility management is based on Signalling System 7 (SS7), an international intelligent network (IN) telephony standard. Each mobile is identified by a mobile station ISDN (MSISDN) number consisting of a country code (CC), national destination code (NDC) and subscriber number (SN). The MSISDN is used by a serving MSC to interrogate the appropriate HLR prior to providing service. The serving VLR provides a mobile station roaming number (MSRN)–similar in format to the MSISDN and in function to the TLDN–for temporary use in forwarding calls to the roaming mobile.

GSM provides the capability for a base station to autonomously handle handovers between coverage areas under its control without involvement from the MSC. This process is called internal connection handoff. Following handoffs, the original MSC handling the call retains control even though the call may be going through a new serving MSC.

Synchronous and asynchronous data services have been defined at 9.6, 4.8 and 2.4 kbps for both full-rate and half-rate operation. Interfaces to V.22bis and V.32 audio modems are also defined for GSM. ISDN and Group 3 facsimile interfaces are also included in the GSM system definition. Industry experts believe that the introduction of data capabilities and interfaces to PC (formerly PCMCIA⁴⁵ ) cards will spur continued exponential growth in worldwide adoption of GSM. Key to this growth is ”plug and play” interfaces which enable standard computer applications as well as vertical applications.

A connectionless packet data service called General Packet Radio Service (GPRS) is in standards development. This GSM capability will define the interworking between the cellular environment and those of X.25 and the Internet world. Two approaches are being considered–dedicated data channels (i.e., a voice channel shared by multiple data mobiles) and fast data channel setup for a single user. The data service objectives include a packet error rate of 10-4 with delay under one second. CCITT recommendation X.121 numbering is used for addressing mobile packet data in GSM.

A GSM-based datagram service called Short Message Service (SMS) is defined in support of Email and other messaging-type applications. This service allows transmission of datagrams which are up to 160 bytes in length at 300 bps on the reverse control channel. Higher data rates are available via traffic channels, which require a call setup; the resultant 9600 bps is better suited to longer messages.

2.11 Japan: PDC

The primary Japanese digital cellular technology is entitled Personal Digital Cellular or PDC and was established as a standard in 1991 by the Ministry of Posts and Telecommunications. It is a TDMA-based technology, which is replacing the analog NTT and JTACS systems, and will operate in the 800 and 1400 MHz frequency bands (Table 2.4). The motivation for PDC was similar to that of GSM–allowing roaming between different regions of the country.

PDC multiplexes three timeslots onto each carrier, like IS-54/136, but has 25 kHz channel spacing to facilitate migration to PDC from the analog systems. PDC uses p/4-DQPSK modulation, also like IS-54/136, with interleaving. The signalling rate is 42 kbps. Mobile assisted handoffs are used, like in GSM and IS-54/136. With a typical frequency reuse factor of 4, the same calculation as in Section 2.10 results in a PDC relative system capacity of 6.3⁴⁶ times AMPS.

2.12 North American Digital Standards

North America has two concurrent digital cellular standards. One is based on TDMA and has been in service since 1992. The other is based on CDMA and is imminent.⁴⁷ The CDMA standard was accepted by the TIA⁴⁸ as an interim standard in 1992, but has not yet been commercially deployed.

Having two incompatible digital standards in North America is ironic, considering the fact that the countries of Europe have long since united behind the common digital standard of GSM. Unless one of the two competing North American standards proves to be clearly superior over the other or cellular service providers agree to support only one of them, roaming subscribers will likely be restricted to the common denominator of AMPS whenever the service area they are visiting supports the ”other” digital standard.⁴⁹

Each of TDMA and CDMA has advantages and disadvantages relative to the other, which are bandied about by their respective adherents and detractors. In general it is difficult to objectively compare TDMA and CDMA systems because their underlying assumptions differ and are not easily related to one another.

Advantages of TDMA-based systems include the capability for variable bit rates (by increasing or decreasing the timeslots in use for one or more users), less stringent power control requirements (time slots reduce mobile duty cycles and thus mobiles’ ability to interfere with one another), the capability for half-duplex (less expensive) radios plus the ability to monitor alternative slots and frequencies for MAHO or MCHO operation (both due to offset transmit and receive time slots).

Advantages of CDMA-based systems include theoretically higher capacity per bandwidth, the ability to withstand noise and fading (due to the spreading of the channel) and the reduced frequency planning and complexity needed (due to shared channels and soft handoffs).

A comparison of AMPS, GSM, TDMA and CDMA transmission systems is displayed in Table 2.5.

The following sections describe the two primary North American digital cellular standards.

2.13 TDMA (IS-54/13x)

Time Division Multiple Access or TDMA was initially defined by the IS-54 standard and is now specified in the IS-13x series⁵⁰ of specifications of the EIA/TIA. Because of its heritage as the original North American digital standard it is sometimes called digital AMPS or D-AMPS.⁵¹

TDMA services were initially deployed during 1992 by McCaw, Southwest Bell, Bell South and others. Although initial customer adoption was slow, there were an estimated half million TDMA subscribers by early 1995. This number is expected to grow dramatically in coming years, especially with new generation vocoders (which improve the perceived voice quality). Because TDMA physical channels are the same as the physical channels of AMPS, TDMA can be easily migrated into and coexist with AMPS systems in a dual mode manner.

TDMA subdivides each of the 30 kHz AMPS channels into 3 full-rate TDMA channels, each of which is capable of supporting a single voice call.⁵² In the future, each of these full-rate channels will be further subdividable into two half-rate channels, each of which–with the necessary coding and compression–could also support a voice call. Thus, TDMA could provide 3 to 6 times the capacity of AMPS traffic channels, with a corresponding gain in trunking efficiency. A similar calculation to that of previous sections yields an estimate of 3.5 to 6.3 times the capacity of an AMPS system [FALC95].

Like AMPS, some of the digital channels are designated as control channels, called digital control channels or DCCH. These control channels serve the same purpose as in AMPS–paging and call control. Three forward-direction call setup control channels are used. The ”A-stream” is used to page mobiles with even-numbered MINs. The ”B-stream” is used to page mobiles with odd-numbered MINs. The ”B/I-stream” indicates the busy/idle status of the reverse control channel, control of which is contested by mobiles wishing to originate calls.

Because of its time-division nature, by offsetting corresponding forward- and reverse-direction time slots, TDMA allows half-duplex phones to be used. This has the benefit of reducing cost and power consumption (i.e., battery size) of the mobile station, but with an increase in complexity due to the variable power envelope. It also allows the monitoring of control channels for out-of-band signalling during a call. Finally, the half-duplex operation allows mobiles to monitor the quality of channels used in neighboring cells in order to assist handoffs.

Originally, TDMA used parametric coding voice digitization, which is based on mathematical models of human vocal sounds. This prohibited the use of analog facsimile and modems due to the resultant distortion of modem signals (which are unlike human voice). Due to complaints of voice quality, the vocoders specified for TDMA have been upgraded with the 1995 standards revision.

TDMA traffic channels use p/4-DQPSK modulation at a 24.3-kbaud channel rate. This results in an effective 48.6 kbps data rate across the six time slots comprising one frame in the 30-kHz physical channel. TDMA standards specify RS-232 and AT-command set-capable mobile units which can use the system at a full-rate data speed of 9.6 kbps, which can be effectively doubled with V.42bis data compression. A triple-rate data speed of 28.8 uncompressed (57.6 kbps compressed) is also specified. Gateways for facsimile and landline modems can be installed at MSCs by TDMA service providers.

A capability called short messaging service (SMS) has been specified in IS-136 to use the DCCH for short messages. This two-way service can deliver messages of up to 256 characters to the display on a subscriber’s phone. Similar services are also specified for CDMA and N-AMPS⁵³ systems.

A very recent packet data initiative has been underway under the auspices of the TDMA Forum, the trade association for TDMA technology participants. The approach favored by the committee working on packet data services uses a dynamic time slot assignment with reservation algorithm which melds directly into the existing TDMA standard to provide CDPD-type services over TDMA channels.

In this proposed standard, all of the usual capabilities are supported in addition to variable bandwidth, which is potentially very large if enough TDMA channels are momentarily available for this purpose. Also specified is an efficient MAC layer ARQ mechanism plus the capability for a mobile unit to monitor both voice and data services simultaneously [CHAN96].

2.14 CDMA (IS-95,99)

Code Division Multiple Access or CDMA was introduced as a cellular standard by QUALCOMM, Inc., in 1989. Based on technology initially discovered during World War II⁵⁴ , CDMA was accepted as an alternative North American digital cellular standard (IS-95) by the TIA in 1992 and has undergone development in the standards process since then. At year-end 1995 there were still no commercially-available CDMA systems.

The basic idea of spread spectrum is to rely on something other than time division or geographic attenuation of a signal to prevent cochannel interference. This makes sense because there are general tendencies of signal strength attenuation, but no absolute rules; topography, weather, foliage, presence of reflectors, etc., are all major factors determining the strength of a received signal.

There are two flavors to spread spectrum technology. One flavor is called frequency hopping spread spectrum and is discussed in Chapter 9. The second flavor of spread spectrum is called direct sequence spread spectrum, which forms the basis of the IS-95 physical layer.

CDMA addresses the two basic problems with radio systems–multipath fading and interference from others in the cellular environment. Both of these challenges are mitigated via the frequency diversity introduced by the wide bandwidth used in CDMA. No single source of interference can impact more than a subset of the spread spectrum in use. CDMA redistributes a base signal across a broad bandwidth under control of digital circuitry.

In CDMA, average interference limits system performance rather than worst-case interference as in FDMA- and TDMA-based systems. Thus, CDMA systems reuse the same frequency in neighboring cells. It’s a good thing, because CDMA RF channels are large–1.25 MHz–and thus there are relatively few of them.

Cellular CDMA systems code speech in a compact 8 Kbps format and transmit a basic data rate of 9600 bps in a spread format of 1.2288 Mbps called ”Mchips per second.” This spreading factor of 128 resulting in a coding gain of 21 dB, which combats many of the vagaries of RF transmission. The spreading mechanism differs between the forward and reverse channels because the capabilities of transmitter and receiver differ on the mobile and system sides of the airlink. Different frequencies are used in forward and reverse directions also, for a limited form of frequency division duplexed or FDD operation.

CDMA uses a soft base-controlled handoff for mobiles transitioning between cells. This improves the quality of service for both voice and data applications. CDMA enjoys somewhat reduced complexity in the network–frequency planning and cell handoff processes–for somewhat increased complexity on the radio link side of the system. This is an interesting approach.

Estimating the capacity of CDMA systems objectively is difficult because of the number of assumptions required.⁵⁵ QUALCOMM claims a relative capacity of 14 times AMPS capacity [VITE95]. A better estimate might be half that value.

The size of CDMA channels (1.25 MHz) will make migration to dual mode cellular operation (i.e., CDMA and AMPS) more of a quantum leap than an evolutionary process. How this will be accomplished–removing large numbers of AMPS channels from AMPS service to CDMA service–in the face of already overloaded systems will be an interesting challenge to the CDMA service providers.

Data services in CDMA systems have been specified for facsimile and asynchronous data applications. Both of these are included in the IS-99 standard and are circuit-switched in nature. A packet-switched service has been defined for CDMA systems with IS-667. Current 14.4 Kbps data rates could be augmented under a proposed Extended CDMA specification to support 76.8 Kbps data streams.

2.15 PCS: Back to the Future?

A number of trends in wireless communications are becoming evident. First among these is the imminent deployment of Personal Communications Services (PCS) by licensed service providers. Auctions held in 1994-96 resulted in the assignment of licenses for both narrowband PCS-based services and broadband PCS-based services in the U.S.⁵⁶ The objective of PCS is to offer so-called ”next generation” digital cellular-like features, as well as offer competition for local loop services. Despite its hype, PCS is simply cellular at higher frequency.

The creation of a PCS industry has been driven by a combination of demand and technological evolution. Early cellular ”mobiles” were large and bulky, requiring an automobile for transport. Early cellular systems were designed with this vehicular bias and were expected to serve no more than four or five million subscribers by the mid-1990’s. As the size of mobile equipment has decreased to that of portable handsets, the costs of providing service has similarly decreased and cell phones have become almost a consumer staple. PCS is aimed at meeting this growing demand for anywhere anytime communications.

The narrowband PCS services will be offered via 10 nationwide licenses in the 930 MHz region. The inbound and outbound channel sizes vary between 12.5 and 50 kHz, depending on the particulars of the license. The target service offering for narrowband PCS is advanced messaging (e.g., acknowledged or two-way paging). A variety of technologies are under development to provide these services. Although the bandwidth for narrowband PCS prohibits general-purpose data networking applications, there is certainly the opportunity for specialized applications, such as telemetry, credit card verification, etc. These services and technologies are discussed in Chapter 9.

The broadband PCS services will initially be offered via 3 30-MHz licenses per market in the 1850-1990 MHz region (Table 2.6). Aimed at all types of voice and data communications, PCS services will compete with cellular and other existing wireless services. Existing technologies such as CDMA, TDMA and GSM (called PCS-1900 in this context) are targeted for broadband PCS; the additional spectrum created by PCS will also encourage enhancements of these standards to better support data services. New technologies are also likely to appear.

The current distinction between market segments is likely to disappear with the new technologies and services. A continuum of services–paging, two-way paging, short messaging, data at varying bandwidths–will likely replace the current paging and data segments. With existing cellular service providers and partnerships expanding their coverage area via the broadband PCS auction, the need for cooperation and interoperation between service providers is likely to become less important than it currently is in cellular.

Western Europe is expected to award up to 4 PCN licenses per country beginning in 1996, based on the DCS1800 standard, to compete with the GSM duopoly in these countries. However, the European Commission policy on PCN is vague, giving wide scope for national discretion in deciding who should obtain or bid to obtain PCN licenses. Also, licensing procedures vary between countries. The European Radiocommunications Committee (ERC) is likely to allocate 1710 to 1785 and 1805 to 1880 MHz bands for PCN use by January 1998.

2.15.1 PCS Licensing

Like AMPS, licenses for North American PCS are assigned on a per market basis. However, rather than considering only local markets, the FCC has also defined PCS markets which are regional in scope. These fifty-one regional markets are referred to as Major Trading Areas or MTAs. The A and B blocks of PCS licenses have been awarded on a per MTA basis.

The C- through F- blocks of PCS licenses will be awarded on the basis of the local markets, called Basic Trading Areas or BTAs. There are 491 BTAs defined in the U.S. Canada and Mexico are likely to follow suit in defining PCS licenses on the basis of MTAs and BTAs for commonality amongst North American mobile service providers.

The magnitude of the bets placed by the ”winners,” coupled with the fact that the ”winners” have to underwrite the costs of relocating existing users of these frequencies (i.e., point-to-point microwave applications) to new frequencies, will exert great pressure on the service providers to get revenues flowing. The PCS service providers are free to use any air interface and system architecture, so long as transmit power levels, etc., are within specified ranges. The need for rapid service deployment will encourage the ”winners” to use the existing digital cellular standards.

2.15.2 PCS Standards

Probably the most controversial aspect of PCS is the continuation of the digital standards battle from the cellular industry; with the addition of GSM as a contender one could argue that the holy wars have escalated. With the possibility of one or more service providers offering nationwide service between their existing cellular licenses and the new PCS licenses, proprietary standards could also emerge. In some cases previously-supported standards and licenses from their cellular markets are being dropped by service providers in favor of the standards and licenses selected by PCS partnerships in which they are involved to avoid conflicts of religion (technical standards) and geography (licenses).⁵⁷

In any case, as in digital cellular, the multiplicity of standards will limit the capability for ”roaming” in another service provider’s coverage area. As always, the old standby–AMPS in the 800 MHz bands–will be the common denominator. Most portables are likely to continue to support this analog standard to prevent subscribers from being limited to ”islands” of mobile services.

At this time (mid-1996), the Joint Technical Committee (JTC) of the TIA and the T1/T45 engineering group have approved four technical airlink standards for PCS. They are CDMA, TDMA, GSM and a composite CDMA/TDMA/FDMA standard proposed by Omnipoint. All of the standards running in the 800-900 MHz bands have been modified in the appropriate ways to support ”up-banded” operation. Each of the standards has its supporters and detractors.

2.15.3 PCS Challenges

The primary challenges for PCS service providers–once the standards decision has been made–are the relocation of current users of PCS frequencies to other frequencies and site acquisition. Relocating the current spectrum users–called ”incumbents”–is projected to cost the PCS ”winners” over $ 1 billion. Each market must be negotiated separately. According to the rules established by the FCC, the incumbents have provisions to return to the 1900 MHz frequencies if they are not satisfied with the new higher-frequency microwave operations. Up to ten thousand such microwave links must be relocated.

Site acquisition has always been a challenge for cellular service providers and will be for PCS service providers as well. Many of the best cell sites have already been taken by existing cellular service providers and zoning board approvals are getting harder to come by. The higher frequencies and lower power levels to be used for PCS dictate a greater number of cell sites than in conventional cellular systems. ⁵⁸

The general idea of PCS is for base stations to be located on utility poles and billboards, etc., which will greatly reduce the costs of acquiring real estate. However, this will be mitigated by the additional infrastructure equipment required. Meanwhile the existing cellular carriers aren’t exactly standing still; both in terms of additional deployment and customer penetration they have been extremely active.

The emphasis of PCS services is on small, low power mobile devices. The user is assumed to be a pedestrian, rather than an occupant of a vehicle–the design point for early cellular services. PCS cell sizes are small–less than one kilometer in diameter for small low power mobile devices. Three-dimensional coverage considerations apply, as opposed to the conventional cellular ”flatland.”

However, with the threat of increased competition from PCS, existing cellular service providers are increasingly offering services that are indistinguishable from PCS except for the frequencies in use. The decreasing prices paid by subscribers will force lower margins and a continuation of the mergers between service providers. It has been estimated that in the end there will be at most three nationwide service providers for combined cellular and PCS services.

2.16 Summary

This chapter has presented an overview of cellular systems. Although these systems have traditionally been voice-centric in terms of services and operation, they are now being extended to better support data applications. The most significant of these extensions is Cellular Digital Packet Data, which we introduce in the following chapter.

Chapter 3
Overview of CDPD

This chapter presents an overview of Cellular Digital Packet Data technology–its history, objectives, services and architecture. CDPD is an open system definition which uses digital transmission on analog cellular (AMPS) channels to provide mobile packet-switched data services. CDPD is also a system concept, with a layered architecture which supports evolution to future technologies.

Many aspects of CDPD are generic in the sense that any wide area data network which supports mobility will share these aspects–both positive and negative. Other aspects of CDPD are unique to the cellular industry which is CDPD’s heritage.

The purpose of CDPD is to provide mobile access to the services available via standard connectionless data protocols such as IP and CLNP. CDPD could be considered to be a wireless extension to the Internet which is available anywhere. As we shall see, CDPD could be easily enhanced to support other connectionless network protocols, such as IPv6¹ .

3.1 CDPD Background

During 1991 IBM and McCaw Cellular Communications, Inc.² , began a collaborative effort to determine the feasibility of overlaying a digital packet-switched data network on the North American AMPS analog cellular system. This joint venture resulted in a proof of concept implementation at McCaw’s headquarters in Kirkland, Washington, at the end of the year. The technology was named ”Celluplan II” by IBM, the provider of the initial conceptual framework for the technology.³

3.1.1 CDPD Prototypes

The initial prototype system used a private partition of cellular channels in a single cell to demonstrate the concept of frequency-hopping, also called channel-hopping⁴ . In this demonstration, the RF coverage exceeded that provided by AMPS because of the digital GMSK modulation and the Reed-Solomon (63,39) forward error correction coding employed.

Following the demonstration’s success, plans were made for extending the scope of the project to include a field trial of a larger system. The technology was renamed ”Data Over Cellular” and again renamed Cellular Digital Packet Data or CDPD. From this point forward the prototype CDPD effort was dominated by schedules which could be charitably characterized as highly aggressive and accompanied by hyperbole to match.

In order to standardize the technology and increase the geographic coverage of the eventual service, McCaw enlisted the support of other large cellular service providers. The public announcement of CDPD and its backing by eight of the largest North American wireless services providers was made in April, 1992. In May, 1992, these wireless services providers (Ameritech, Bell Atlantic, GTE, McCaw, Nynex, PacTel, Southwest Bell, US West) staged a CDPD Technical Conference in Santa Clara.

The Santa Clara conference drew more than 600 attendees, reflecting the widespread interest in mobile data communications. Copies of a preliminary technical specification (Release 0.1) were distributed and plans for an upcoming field trial in the Bay Area were disclosed.

The early CDPD system architecture was telecommunications-oriented. A modified version of SCCP from SS7 provided the connection-oriented transport service, then considered necessary for support of mobile devices. This architecture required gateway services⁵ to interconnect with the rest of the data networking world, not unlike the competing RAM and Ardis systems. The RF channel used the same GMSK modulation with Reed-Solomon (63,39) forward error correction as in the earlier demonstration system. The RF MAC sublayer was specified with both polled and contention-based modes of shared channel operation.

The so-called ”field trial” took place in the Bay Area during the latter half of 1992 and validated the radio resource management operation of the CDPD overlay on cellular systems. Channel hopping, cell transfer and interference avoidance were all exhaustively tested. Suspicious police often followed rented antenna-clad Cadillacs occupied by test personnel and equipment slowly cruising Camino Real in the dead of night, when potential cellular voice customer impact would be minimized.

The results of the trial were sufficient to convince seven of the cellular service providers supporting the effort (Ameritech, Bell Atlantic, GTE, McCaw, Nynex, PacTel, Southwest Bell) to continue the development of the technical specifications.

The early focus of the technical specifications effort was on the RF aspects of the system. The end result was a Reed-Solomon (63,47) forward error correction designed to increase the effective user bandwidth of the GMSK-modulated bitstream and a contention resolution scheme, similar to Ethernet, which provides both collision avoidance and collision detection. This MAC protocol is called slotted non-persistent Digital Sense Multiple Access (DSMA) with collision detection.

3.1.2 ”CDPD Lite”

Although the 1992 CDPD specification team focus was on robust airlink protocols, work continued on the overall system architecture. In the fall, several members of the team concluded that the telephony-based system architecture was inappropriate for the services to be provided by CDPD⁶ . During the first week of December, a five-person subteam designed an alternative architecture, informally dubbed ”CDPD Lite.”

The ”CDPD Lite” architecture was based on existing data networking standards and on the open connectionless Layer 3 protocols (IP and CLNP) that were available. This open architecture eliminated the need for gateways and leveraged conventional network recovery mechanisms to help support transient mobility. Its mobility management scheme was based on the early work of the IETF Mobile IP task force. This architecture, elegant in its simplicity, was later adopted as the ”official” CDPD architecture by the group of seven cellular service providers who continued to support the effort.

The first half of 1993 saw the completion of the Bay Area ”field trial.” Preliminary specification releases of the new CDPD architecture were published in March (Release 0.8) and May (Release 0.9). The first official release of the specification (CDPD System Specification Release 1.0 [CDPD93]) followed in July. All of these published releases embodied the ”CDPD Lite” architecture.

The second half of 1993 saw the initial development of legitimate CDPD infrastructure and mobile devices. Separate efforts by McCaw and a collaboration of other cellular service providers resulted in two demonstrations of CDPD operation at the large Comdex trade show in Las Vegas in November. This rapid four-month specification to demonstration timeframe reflects the benefits of the open CDPD architecture.

3.1.3 CDPD Forum

In 1994, the cellular service providers behind the CDPD specification development created the CDPD Forum to enlarge the base of support for CDPD. This trade association of service providers, infrastructure and mobile unit vendors, and software and applications developers continues to have as its objective the support and promotion of CDPD as a basis for mobile data applications.

The CDPD Forum supported the development of CDPD System Specification and Implementor Guidelines Release 1.1 [CDPD95] during 1994. This release was published in January, 1995, and includes enhancements to the radio resource management procedures, the Mobile Data Link Protocol (MDLP), accounting and multicast capabilities as well as protocol test specifications and implementation guidelines. The primary purpose of Release 1.1 was to finish the definition of incomplete or unclear capabilities of CDPD Release 1.0.

Release 1.1 also restructured the CDPD specification into a System Specification and Implementor Guidelines. This restructuring provided a better distinction between stable protocols (which were placed into the System Specification) and other Parts⁷ which might be less mature or more implementation-specific or otherwise unsuitable for a system standard (and which were placed into the Implementor Guidelines⁸ ).

In 1995, the CDPD Forum developed certification test plans for mobiles and initiated a selection process for an agency to conduct these tests. Further extensions to CDPD were contributed by Forum members and ratified by the Technical Steering Committee in the areas of circuit-switched access to CDPD services and limited size messaging. By mid-1995, the CDPD Forum had almost 100 member companies, reflecting the diverse and dynamic interests in CDPD technology. It continues to actively support the contributions of members to CDPD standards and implementor guidelines development. A standards track process has been developed which is loosely modeled after that of the IETF.

3.1.4 CDPD Service Providers

Separately, a Service Provider Corporation was created in 1995 to provide a forum for resolving inter-service provider concerns. This SPCo manages the activities of the CDPD Network Information Center (NIC), which include administration of CDPD network address blocks, DNS names and unique CDPD identifiers amongst the service providers.

Over the course of the year, CDPD service providers deployed services and expanded the customer base for those services. At year-end 1995, CDPD services were offered in over thirty markets, as depicted in Figure 3.1. Inter-service provider testing and interconnection was also well underway, with agreements and interoperation amongst the major service providers in place by early 1996.

Time will tell just how successful CDPD will be in terms of commercial adoption. In any case, the technology is likely to influence future mobile data systems. Both the CDMA and TDMA Forums are developing technical specifications for packet data services whose mobility management mechanisms are identical to and will interoperate with CDPD; this will allow service providers to offer CDPD services via multiple airlinks.

3.2 Relationship of CDPD to other Cellular Data Initiatives

When CDPD was first conceived and specified in the early 1990’s, the CDMA (IS-95) and TDMA (IS-54, now IS-136) standards efforts for North American cellular voice services were well underway. In fact, TDMA digital voice services were already being deployed in a number of markets. It was a foregone conclusion that both of these competing digital voice standards would eventually support data services as well.

Unfortunately, the schism between the North American cellular service providers which support CDMA and those which support TDMA shows little sign of closing. In early 1996, the only common North American cellular standard is still the analog AMPS (Advanced Mobile Phone System) standard. The common denominator for data services likewise remains CDPD.

In the future, it is likely that both CDMA and TDMA will be deployed throughout North America, thanks in large part to the new PCS spectrum. These standards are being extended to support data services, both in circuit-switched and packet-switched modes. The packet-switched modes are based on a CDPD system design–all that is changed is the necessary radio modulation techniques and protocols, primarily at Physical and MAC Layers. So rather than replacing CDPD, these digital cellular standards will instead adopt CDPD for their base data services-providing architecture.

Another more recent adoption of CDPD architecture is embodied in the new personal Air Communications Technology (pACT) announced by AT& T Wireless Services and others. This two-way messaging technology modifies CDPD to use the narrowband PCS channels auctioned in 1994–another example of CDPD operating over alternative airlinks. pACT is discussed in Chapter 9.

CDPD architecture was conceived with this kind of extensibility in mind. CDPD is more than an airlink specification–it is a system architecture for mobile data services which can support multiple RF technologies. One of the more common misperceptions in the trade press has been the eventual ”replacement” of CDPD by emerging standards which were based on the new RF technologies.⁹

Another initiative is that of the Portable Computer and Communications Association (PCCA), which has defined standard APIs for the mobile computing industry since 1993. Their recent STD-201 wireless modem standard is based on the commonly-used Microsoft Network-Device Driver Interface Specification (NDIS), and will allow software developers to support wireless modes of operation to Windows applications. STD-201 complements the earlier STD-101 standard, also developed in the PCCA, which defined hardware- and network-independent extensions to the popular Hayes AT modem command set for wireless operation. CDPD is one of the primary target networks for these interface specifications.

3.3 CDPD Services and Characteristics

CDPD is an enabling technology which provides support for mobility in the WAN environment. To accomplish this, CDPD provides three kinds of services–network services, network application services and network support services. As Figure 3.2 depicts, these services provide support for applications ranging from stationary vending machines to mobile vehicle tracking. The services also support applications ranging from telemetry (extremely low data rates) to interactive PC-based applications such as remote server access.

3.3.1 CDPD Network Services

Network services are data transfer services–the capability of moving data from one location to another. This is the basic service type offered by CDPD. It neither adds value nor content to what the user intends; it is simply data carriage to and from a mobile device.

In terms of data networking, CDPD provides support for routable connectionless peer-to-peer Layer 3 protocols, such as IP and CLNP. Other Layer 3 protocols, such as IPv6 are likely to be supported in the future as well. By definition, Layer 3 is the layer responsible for getting data from one point to another across one or more networks.

Part of basic network service is the provision of access to systems and services–both public and private–external to CDPD. These services can be data resources or networks. In most cases a mobile user sends data to and receives data from a resource external to the CDPD network. CDPD simply provides a means of accessing that external resource and could simply be regarded as an extension of existing data communications networks which are IP- or CLNP-based.

Basic CDPD network services are summarized in this overview chapter and described in more detail in Chapter 4 (mobility management) and Chapter 5 (network access).

3.3.2 CDPD Network Support Services

Network support services are the services necessary to support the operation of a mobile data network. Network support services include things such as network management, usage accounting and security, which are necessary to operate any network. CDPD network support services also include things such as mobility management and radio resource management, which are necessary for mobile wireless WANs.¹⁰

In theory, an end-user of the system could use the network services while perfectly oblivious to the existence of the network support services (at least as long as these services are running correctly, save accounting!). In many cases the support services could be considered to add ”intelligence” to the system; an example is the fault recovery actions taken by network management when an exceptional condition arises.

CDPD network support services are described in Chapter 6 (security) and Chapter 7 (other support services).

3.3.3 CDPD Network Application Services

Network application services are services which add value or content to a user’s activities above and beyond basic data carriage. The end-user is typically quite aware of these value-added services and often must explicitly subscribe to them. These services in CDPD could leverage off of the mobility of the user, such as subscriber location services or limited size messaging services. It is also possible that these services could be independent of mobility, such as advanced messaging capabilities.

The CDPD specification includes technical descriptions of mobility-enhanced value-added services. Other CDPD network application services could be based on the intrinsic broadcast and multicast capability defined in the CDPD System Specification.

Some of these network application services are described in Chapter 8 (limited size messaging).

3.4 CDPD Design Goals and Considerations

A number of design goals and considerations influenced the architecture of CDPD. These are technical objectives only and are not necessarily indicative of the business objectives of CDPD service providers or other industry participants.

Many of these technical objectives reflect the lessons learned by cellular service providers over their first decade. Since the CDPD initiative originated in the cellular community, these objectives largely reflect the interests of the cellular service providers in developing an open and interoperable standard for mobile data services.

As always, full enjoyment of these applications depends on proper implementation by vendors and intelligent operation by service providers.

3.4.1 Location Independence

The operation and appearance of basic CDPD services to an end-user (called a subscriber) or application is intended to be independent of the service provider and location at which those services are made available. If a user is receiving service while located in a different CDPD service providers’ coverage area, there should be little, if any, impact on the user or application. This seamless ”visiting” should not be confused with ”roaming” in the cellular world, which as we have seen has had many bad connotations.

However, CDPD also allows service providers to differentiate their respective service offerings by offering services which add value above and beyond the baseline CDPD services. By defining as little as possible, the CDPD System Specification leaves plenty of invention up to the service providers. Over time, service providers will likely differentiate their CDPD service offerings with these additional capabilities, as well as by the level of service they provide.

3.4.2 Application Transparency

Applications do not have to be modified in order to use CDPD; an explicit goal of CDPD is to have minimal impact on end-devices and applications. CDPD services should be accessed via industry-standard application program interfaces (APIs), which are identical to those employed in conventional networks. According to the CDPD System Specification, ”the CDPD Network design shall ensure that no impact is exerted on Transport and higher protocols.”

However, it is possible that timers (such as TCP restart timeouts, etc.) might have to be altered somewhat for optimal CDPD (or other wireless services) usage. Otherwise, CDPD is intended to be fully compatible with existing data networking applications. Time and again this objective has been demonstrated with new applications running immediately on CDPD with no more effort required than on a conventional LAN. We do this constantly.

However, this application transparency does not prevent additional services and applications, which could not be provided by conventional data networks (such as remote telemetry or location services), from being supported by CDPD. Many of these services which are exclusive to mobile solutions are the economic raison d’etre for CDPD for end-users (and thus also for service providers). An example is the limited size messaging capability introduced in the CDPD Forum in mid-1995.

3.4.3 Multiprotocol Support

CDPD was intended from the outset to support more than a single connectionless Layer 3 protocol. This was an important consideration because there were concerns during the early 1990’s that IP Class B address blocks would be fully assigned in the near future. The impending shortage of address space (among other reasons) motivated the creation of the IPng committee by the IETF [BRAD96]. The resulting IPv6 protocol standard was drafted in 1994. This protocol will likely also be supported by CDPD as its implementation and usage become widespread.

CLNP was also supported from the beginning of the ”CDPD Lite” architecture. Support for CLNP is necessary to support several of the OSI applications, such as X.700 (CMIP) for network management and X.400 (message transport) for accounting information exchange between service providers. CLNP is also key to the intersystem NPDU redirection which lies at the heart of CDPD mobility.

3.4.4 Interoperability

The CDPD System Specification and Implementor Guidelines provide the information necessary to ensure interoperability between the equipment and software provided by multiple vendors. This in turn minimizes the infrastructure and device costs in a competitive environment. Interoperability between service providers is also supported by one of the three well-defined interfaces in the CDPD specification. Abstract test suite definitions further support interoperability between equipment and software providers and CDPD service providers.

3.4.5 Minimal Invention

One of the goals of the CDPD specification effort was to minimize the overall risk to the fledgling industry by minimizing the invention in CDPD. The fast schedule required that off-the-shelf technology be utilized wherever possible. The bulk of the ”invention” in CDPD consists of combining existing technology in a new way, mostly over the RF airlink. CDPD service providers have the ability to reuse existing cellular facilities to support data as well as voice services.

3.4.6 Optimal Usage of RF

Although CDPD architecture is holistic in nature, a key design goal was to make efficient use of the radio channel. Since this airlink is the most precious resource in the system, CDPD design trade-offs consistently favored airlink efficiency at the potential expense of other resources of the system. An example of this is the extensive application of standard data compression (V.42bis) technologies to conserve over-the-air transmission at the expense of greater computational loads at either side of the airlink. In Moore’s Law we trust.

The raw signalling rate of the airlink is 19.2 Kbps; with Reed-Solomon (63,47) FEC, this amounts to a maximum effective bandwidth of 14.4 Kbps full-duplex before considering channel control overhead. Since the inbound channel is shared via a contention-based access scheme, which further imposes overhead on both the inbound and outbound channels, it is essential that the airlink be used effectively.

Another concern in CDPD development was not ”pushing the envelope” of RF technology too far. One of the obvious ways to get around the constraint of the narrow airlink ”pipe” would be to employ more sophisticated RF modulation techniques. However, doing so would have increased the cost for mobile devices beyond what commercial users would bear.¹¹ GMSK is already used in GSM systems around the world and provides the 19.2 Kbps data transmission rate over the air, which provides the basis for CDPD.

3.4.7 Evolutionary Design

The data networking world is evolutionary and so must CDPD be. The definition of CDPD architecture strictly in terms of OSI reference model layers coupled with the limited scope of the system definition (i.e., OSI layer 3 and below) allows for evolution of CDPD as well as network technologies supported by CDPD. New applications and transport protocols are immediately operable on CDPD systems because of its support for native IP.

The recent introduction of a hybrid architecture of cellular circuit-switched airlink in the CDPD Forum exemplifies the evolutionary nature of CDPD architecture. Similarly, anticipated support for IPv6 (as it becomes widely adopted) will also be straight-forward. Other areas of anticipated evolution include the airlink link layer protocol (MDLP) and airlink security (encryption and authentication); evolution in these areas is supported with version codes, command types, etc.

3.4.8 Open

CDPD has always been intended to be an open system, free from all proprietary technology. In our opinion, the only known aspect of CDPD architecture involving previous intellectual property rights is the use of public key cryptography techniques which underlie security across the airlink. CDPD is based on open standards and protocols; the limited invention in CDPD is also open and freely available.

An open standard provides the basis for multi-vendor interoperability and the resulting economic benefits of competition. It also encourages multiple vendors to participate and compete in the marketplace, driving down costs. Multiple developers working on a common problem are likely to produce the best solution. These benefits are rarely achieved by proprietary solutions.

3.4.9 Secure

CDPD is intended to provide data networking services which are no less secure than conventional WANs. (Of course, the trade press has popularized the notion of insecurity of the Internet, so maybe this isn’t such a great objective!) The security capabilities provided by CDPD are integral to the system, not later add-ons.

These security capabilities include confidentiality for both users and their data (via data encryption and the use of temporary IDs for users), user authentication and the key management necessary to support these capabilities. The design requirement for CDPD was to prevent casual ”eavesdropping” of users across the airlink; thus the users of CDPD are no less secure than users of conventional networks. Of course, as a public shared network, end-to-end encryption by applications is always encouraged.

The security capabilities of CDPD are limited to the airlink interface. The specification team always felt that the other primary interfaces of CDPD would be able to leverage off the work being done by network security experts. Network security is an issue which transcends mobile networks; mobility only exacerbates the challenges of key management, authentication, etc.

3.4.10 Simple

The architecture of CDPD is elegant in its simplicity. In particular, the mobility management aspect of CDPD is quite straight-forward, by design. [KRIS95] states that simplicity ”is one of the most important attributes for a routing protocol.” Simplicity in design allows for more rapid development and more reliable operation. The rapid CDPD specification draft to service deployment timeframe indicates the value of simplicity.

3.4.11 Transparent to the Existing Cellular Voice Network

CDPD has always been intended to be an overlay service on the existing cellular voice infrastructure. Maintenance of a high quality cellular voice service is of paramount importance to the cellular voice service providers who provided the initial funding for CDPD specification development. Therefore, it is essential that the introduction of CDPD service not negatively impact the basic cellular voice service offered by cellular service providers.

There are two general concerns about the CDPD overlay. First is the voice service degradation which could result from RF interference by CDPD RF transmitters. Because CDPD is transparent to the AMPS network (i.e., the AMPS system is unaware of and does not require modifications for CDPD), CDPD must operate in a way which does not interfere with the cellular voice system. This has proven to be a significant constraint on CDPD design, influencing features such as channel hopping.

The second general concern about the CDPD overlay is the removal of cellular voice capacity required to dedicate AMPS channels to CDPD service. This concern is addressed by the channel hopping capability of CDPD in which a CDPD base station utilizes currently-idle AMPS channels to provide CDPD services. An RF ”sniffer” is included in the cell site infrastructure to enable the CDPD channel stream to ”hop” to a new physical AMPS channel whenever cellular energy is detected on the channel used by CDPD. An anticipatory algorithm could also be used to minimize the frequency of involuntary channel hops.

The channel hopping mode of operation relies on proper cellular engineering practices, in which it is assumed that the busy hour call blocking probability is approximately two percent.¹² Using an Erlang B distribution (a standard engineering practice in telephony), one can see that approximately twenty to thirty percent of all channels must be available at any instant (on average) to provide the two percent blocking service level of quality. It is this set of idle channels that CDPD is intended to operate with in the channel hopping mode.

3.5 The CDPD Architectural Approach

The approach taken by the CDPD specification team assumed that all objects could be defined by their interfaces and functions. Initially this philosophy was applied to the entire system.

Upon examination of the basic design goals and considerations of the CDPD network, the specification team found that they could be satisfied with a network providing an ”over-the-air” interface to the mobile devices, an external interface to land-line hosts, and an inter-service provider interface to link multiple cooperating CDPD service providers.

This philosophy led to the development of the ”CDPD cloud”¹³ depicted in Figure 3.3. With this model, it is necessary and sufficient to fully define the CDPD network by specifying the detailed interface to the mobile devices, external networks and other CDPD networks. Within the CDPD network ”cloud”, the necessary support functions of mobility management and data delivery can be defined separately.

Although this ”cloud” approach addresses the stated requirements of the CDPD network, it does not address the practical side of developing a new network service. Any new network service deployment requires the development of new network infrastructure equipment. The technical specification of the new service must also define these components.

While the ”cloud” approach to network system specification defines all the necessary information for network equipment development, it may result in vastly different internal network architectures. Different equipment vendors may conceive of different sets of equipment to provide the same functionality and interfaces. This is true regardless of the level of detail attained in the system specification, assuming it doesn’t go so far as to specify an actual implementation.

Some of the service providers expressed concerns about the ”cloud” approach of system specification. They recognized that this approach could result in networks that interoperate (over the I - Interface) but cannot share internal components. They were concerned that the RF equipment of one vendor would only operate with network routing equipment from the same vendor. This type of limitation would severely restrict a service provider’s flexibility in equipment vendor selection. Indeed, most of these service providers have already lived under these types of captive marketing approaches in the cellular telephony world. They did not want this vendor-dependence to continue.

It was the service provider’s initial discomfort with this aspect of the CDPD ”cloud” that drove the specification team to then define individual components within the CDPD network. Some of these components are depicted in Figure 3.4.

3.6 The Three Key CDPD Interfaces

CDPD architecture defines three key external interfaces, as depicted in Figure 3.4. Because these interfaces–”A”, ”E” and ”I”–form the logical boundaries for a CDPD service provider’s network, they are essential to the proper operation of CDPD.

Other lesser interfaces are defined within the CDPD service provider ”cloud.” However, since these internal interfaces are under the control of a single CDPD service provider, their specifications could be considered to be recommendations rather than requirements.

Although this may sound contrary to our previous discussion, flexibility is required in internal interfaces because of the different network implementation approaches favored by the various CDPD service providers. It is important to differentiate between technical specification and implementation requirements.

Each of the interfaces defined in the CDPD specification includes a profile representing the protocols supported or required at each layer in the OSI Reference Model. An example of these profiles is displayed in Figure 3.5. Well-defined primitives at each layer request services of the layer below; the services provided to each layer consists of the collective set of services provided by all of the underlying layers.

3.6.1 The A-Interface

The A-Interface or airlink is the interface between the CDPD mobile device and the CDPD network and contains much of the ”invention” of CDPD. This is the point at which CDPD network services are accessed by a subscriber and is described in detail in Chapter 5 (network access). It is defined by Parts 400 through 409 in the CDPD System Specification. Although the airlink receives much attention, this is only one part of the overall CDPD system architecture.

3.6.2 The E-Interface

The External or E-Interface of CDPD is the means by which CDPD interoperates with the rest of the world and is key to the provision of CDPD network services. Conventional data networking protocols are used for data carriage between CDPD and external data services such as the Internet, VANs, wide area transport providers or private networks.

The Layer 3 protocols supported at the E-Interface include the same connectionless protocols as within CDPD–IP and CLNP. IPv6 is likely to be supported at a later time as it matures. Other protocols, such as APPN (via MPTN) or IPX, could be supported either via encapsulation (say within IP packets) or via protocol translation gateways. Either of these techniques is outside the scope of the CDPD specifications.

Border gateway protocols such as BGP-4 and IDRP are also recommended at the E-Interface. This is necessary because the E-Interface specifies a boundary between two autonomous systems–that of a CDPD service provider and that of an external party. Initially, static routing is likely to be employed at the E-Interface for CLNP traffic.

The E-Interface is intended to be no different than the interface to any other autonomous system. All of the issues of security, routing, name-to-address translation, etc., apply. The fact that the CDPD service provider supports mobility is transparent to the E-Interface, by design.

3.6.3 The I-Interface

As we discussed in Chapter 2, the North American cellular service environment is partitioned into markets which are served by a multiplicity of service providers. Seamless nationwide coverage (a goal of CDPD) requires these service providers to be capable of supporting each others’ customers. Seamless coverage also requires the capability for a transparent and smooth transfer from one system to its neighbors.

The Inter-service provider or I-Interface defines the means by which the CDPD service providers can collectively provide a seamless nationwide service. From a purely technical viewpoint, there is nothing preventing CDPD service from being offered worldwide.

The I-Interface supports the same protocols as the E-interface plus the Mobile Network Location Protocol or MNLP. This protocol is the means by which mobile users from one system are supported by another system and is a key piece of the CDPD mobility management scheme. Additional protocols for network management (CMIP) and accounting (X.400-based) are also defined for the I-interface. All of the protocols across the I-interface are based on CLNP.¹⁴

3.7 CDPD Network Elements

The successful operation of any communications network requires cooperating system components. These network components or entities perform predefined and a priori agreed upon functions. The components must also communicate with each other in a predefined manner.

The component entities defined by the CDPD architecture (see Figure 3.6) include several that are unique to CDPD and others which are standard ”off the shelf” components Since the CDPD network is an overlay on the existing cellular network, it only made sense to leverage off the existing infrastructure. The CDPD network model defines component elements reflecting the cellular network, and is illustrated in Figure 3.6.

In a typical cellular telephone network, cell sites are deployed throughout the coverage area. At each cell site, a base station transmits and receives RF signals from the cellular telephones through an antenna tower. The demodulated signals are then digitized and placed on voice communications channels of a multiplexing channel. These are typically 1.544 Mbps T1 connections which can carry up to 24 voice calls, each one occupying a 56 kbps digital channel. The voice circuits terminate at a mobile switching center (MSC), which interfaces with the land-line telephone network.

The design of the CDPD system is aimed at minimizing the changes in network operation and maximizing the reuse of the existing cellular network infrastructure. Given this constraint, the next specification team objective was using the system’s resources effectively. Without a doubt, the most precious resource is the radio spectrum. Beyond the radio resource, the precious system resources to consider are the physical plant and the communications links.

Many people may not immediately realize that the collection of cell sites constitute a significant investment by cellular service providers.¹⁵ Many cell sites have to be constructed while others involve ongoing leasehold arrangements. In addition, the antenna tower space is also tied with space, height, zoning and licensing issues. Furthermore, each base station typically has a T-1 circuit dedicated to carrying its traffic.

The CDPD system architecture reuses these components by specifying a piece of additional infrastructure equipment at the existing cell sites. This new piece of equipment, named the Mobile Data Base Station (MDBS), can use the existing antenna feeds and towers. Furthermore, the functionality of the box has been limited to allow small compact designs that can fit within most existing cell sites. Recent designs can be deployed in microcell environments.

The MDBSs handle the RF communications to and from the mobile devices and relay the data to and from more centralized CDPD network infrastructure equipment. This communication path can reuse channels on the existing cellular communications links. In most cell sites, there are more than enough channels to justify the deployment of a T1 connection. However, there are typically a few unused channels in these T1 links. The MDBS can use the idle channels for CDPD data. Once all the CDPD data channels are brought back to the MSC site, they can be ”groomed” off to the CDPD central infrastructure equipment, the Mobile Data Intermediate System (MD-IS).

With this design approach, the CDPD specification team defined the conceptual reference model depicted in Figure 3.7 . In the following sections, each of the identified components are described.

3.7.1 The Mobile End System (M-ES)

In OSI terminology, an End System or ES is a network node which is the ultimate source and destination of NPDUs. This is the same as a host in Internet terminology. The Mobile End System or M-ES is any network host which happens to be mobile (i.e., CDPD radio and software-equipped). Example M-ESs could include telemetry devices, personal communicators and personal computers. Even communicators in vending machines (which don’t actually move) could be considered to be M-ESs.

The M-ES is a network device with protocols specified up to and including Layer 3. The M-ES has a Layer 3 address which is globally unique in both the CDPD and conventional networking environments. M-ESs are true mobile hosts and CDPD networks are extensions of IP-based (and CLNP-based) networks, such as the Internet. There is no need for gateways as in other wireless packet data services.

Applications on the M-ES access the network via conventional means–sockets, TLI, NDIS and ODI are a few example APIs. Standard APIs and protocols are emphasized, especially at the M-ES; this allows immediate portation of applications from existing PCs to CDPD.

Both full and half duplex M-ESs are supported by CDPD. This allows low-cost devices with only a single radio to provide mobile data services to low traffic generating applications, such as the previously-mentioned vending machines.

The M-ES architecture consists of three distinct functional blocks, as illustrated in Figure 3.8 : the subscriber unit, the subscriber identity module and the mobile application subsystem.

The Subscriber Unit (SU) constitutes the portion of the device that establishes and maintains data communications with the CDPD network infrastructure. It achieves this through proper execution of the CDPD airlink protocols, which extend from the physical layer to the network layer. Also included are administrative layers above the network layer and the layer management entities necessary to ensure proper cooperation between the M-ES and the network.

The Subscriber Identity Module (SIM) is the repository of identity and authentication credentials for the network address in use at the M-ES. Every subscriber device must have the proper authentication credentials and identity. This function was separated from the rest of the M-ES functions to enable implementation of removable SIM cards as in GSM. This is in consideration for users that may find it easier to carry a small smart card with all the necessary identity information than to carry a complete Mobile End System¹⁶ . The specification used to define the SIM is based on the appropriate GSM standards.

The Mobile Application Subsystem (MAS) is the portion of the M-ES that contain all protocols at the network layer and above. This is what gives the M-ES something interesting to communicate with. The application is whatever needs mobile network connectivity–Email, remote database access, vending machines, etc.

M-ESs span a wide range of feature sets and form factors. Some of the units currently available support the CDPD protocols only, while others also provide AMPS cellular modem capability. Still others support voice communications with the addition of a handset.

Along with varying form factors, M-ESs also come with different power source options. Some require large external power sources such as would be available in a vehicle. Others supply their own power through internal batteries. Still others draw on the power source of a laptop or notebook computer.

3.7.2 The Mobile Data Base Station (MDBS)

The Mobile Data Base Station or MDBS is the system end of the MAC sublayer over the airlink. The MDBS arbitrates activities on the channels it hosts at the MAC sublayer much like an Ethernet hub. It relays frames at the LLC sublayer. This device is physically located at cell sites.

The MDBS is the network infrastructure device that bridges the different media between the Mobile End System and the CDPD network. The MDBS communicates with the M-ESs through the airlink physical interface. It performs all the necessary modulation of the data bits onto the RF channel. It also demodulates the RF signal into digital bits of data. These operations are carried out within the specifications and rules required to operate on the cellular frequency bands.

In a CDPD system, multiple mobile devices share the use of a single radio channel. To ensure proper sharing of the RF channel, a medium access control mechanism is used to arbitrate access to that channel. An MDBS is an active participant in the CDPD medium access control scheme, Digital Sense Multiple Access (DSMA). Once the data stream is successfully received and decoded by the MDBS, it relays the Link Layer frames between the M-ES and the MD-IS.

The MDBS is Layer 3-addressable for network management and radio resource management purposes. In terms of user data, the MDBS is little more than a Layer 2 relay between the RF and the conventional networking worlds. For user traffic at Layer 3, the MDBS is a ”phantom” element.¹⁷

3.7.3 The Mobile Data Intermediate System (MD-IS)

The Mobile Data Intermediate System or MD-IS is the focal point of CDPD mobility management. It has two functions in its role as a mobility-enabling router–the mobile serving function and the mobile home function.

The mobile serving function or MSF of the MD-IS provides the system end-point of the LLC sublayer MDLP link, opposite the mobile. This connection-oriented link serves as the foundation for the registration of mobiles to the system. When a mobile announces itself to the system, it is the mobile serving function which receives this registration request.

The mobile home function or MHF provides the anchor for the mobility of the M-ES. Whenever some network entity sends packets destined to the M-ES, the packets are routed in the conventional manner to the mobile home function, which then forwards them to the mobile serving function (which could be located in another MD-IS), based on previously exchanged messages between the mobile serving and mobile home functions.

The MD-IS is the most important data networking entity in the CDPD system. These devices are responsible for most of the mobility management functions of the network. The MD-ISs perform the functions necessary to track the local access point of the mobile devices. In other words, the MD-IS deals with the determination of and tracking of the exact radio coverage cell each M-ES is operating from.

The MD-IS is responsible for presenting an interface to the external networks on behalf of all the M-ESs in the CDPD network. This interface is necessary to ensure that hosts wishing to communicate to the any M-ES can traverse the external networks and enter the CDPD network at the proper point of presence.

The MD-IS is also responsible for routing all network traffic to the appropriate M-ES destination. The MD-ISs within a CDPD network must cooperate to ensure this task is achieved irrespective of whether the M-ES is in a local area, or if it is at the far end of the continent.

Finally, since the CDPD network is a commercial public data network, the routing nodes–the MD-ISs–must also perform the necessary administrative functions such as usage accounting.

The MDBS-to-MD-IS Interface

The MDBS/MD-IS Interface is an internal interface between the MDBS and the MD-IS. Since this interface is internal to the CDPD ”cloud,” it is recommended rather than specified. Proprietary solutions have no need to adhere to this interface definition. However (CDPD service providers beware!), allowing a proprietary solution here is tantamount to losing control of one’s system architecture.

This interface is specified for the sole purpose of an open system definition. This allows a multiplicity of vendors for each side of the interface in a more or less plug-’n-play manner. There was much discussion amongst members of the CDPD specification team regarding the need for any specification of this interface. In the end, this interface definition met the needs of the CDPD service providers wishing to order equipment from their vendors.

3.7.4 The Intermediate System (IS)

The Intermediate System or IS is a fancy name for a network router¹⁸ . It is standard OSI terminology and function. In CDPD, ISs handle packet forwarding for both IP and CLNP connectionless Layer 3 protocols, just as in conventional data networks. Typically, a packet traversing between networks will be handled by several ISs on its journey. MD-ISs must also support IS functionality.

In addition to the packet forwarding protocols–IP and CLNP–supported by the ISs, they must also support routing information exchange protocols (in order to function as routers, unless static routing only is used–a bad choice for dynamically changing networks!). The internal routing information exchange protocols defined by the CDPD specification include OSPF for IP and IS-IS for CLNP. External gateway routing information exchange protocols include BGP-4¹⁹ for IP and IDRP for CLNP. It is wise, whenever possible, to have ISs supporting integrated IS-IS to prevent the ”ships in the night” phenomenon, as described in [PERL92].

The CDPD network is a multi-protocol network. This means that the CDPD network is intended to support routing of multiple network layer protocols, including IP and CLNP. The architecture of CDPD allows future extensions for additional connectionless network protocols, such as IPv6. All ISs used in the CDPD network must also support multiple network layer protocols.

For the purpose of providing interconnection between a CDPD network and external networks, and between two CDPD networks, some level of security protection is valuable. For this requirement, the ISs at the borders of the CDPD network are further required to provide security filtering and access control functions, whose definition is beyond the scope of the CDPD System Specification.

3.7.5 The Fixed End System (F-ES)

The Fixed End System ²⁰ or F-ES is a conventional network node, which could be either external to the CDPD service provider network, such as a transport layer peer of an M-ES, or internal, such as the servers which provide network support or application services. The only purpose in explicitly defining an F-ES is to distinguish between a conventional network host and a mobile host.

External F-ESs are the hosts that most CDPD subscribers should be familiar with. They are typically the systems hosting the applications the mobile user wishes to access. Some examples of these F-ESs might include: an inventory system at a home office, an electronic mail server for a road warrior and a user’s favorite World Wide Web site. The most significant thread through these F-ES descriptions is that they are simply common hosts that support standard network layer protocols.

Internal F-ESs are hosts much like external F-ESs. The primary difference is that internal F-ESs operate within the boundaries of the CDPD network. As such, they conceivably are under the control of the service provider and can thus be presented with additional internal network data. Such data may include usage accounting information, mobile location information, subscriber authentication information, etc.

Internal F-ESs allow CDPD service providers to operate administrative servers and value added servers without the need for non-standard communications protocols. This capability is representative of the CDPD network architecture’s flexibility. This flexibility results from the use of standard network protocols and adherence to the ISO Open System Interconnect reference model.

The Accounting Server (AS)

The Accounting Server or AS is an internal F-ES which serves two basic functions–collection and distribution of usage accounting data. Subscriber activity is recorded at the serving MD-IS over a configurable time period, then flushed to the AS. The AS then collates the detailed accounting records and distributes them to the appropriate peer accounting servers.

The distribution of detailed accounting records to home accounting servers is called the serving accounting distributor or SAD function. The reception and redistribution of these records at the home accounting server is called the home accounting distributor or HAD function. Other AS functions are defined in CDPD to enable near real-time accounting, separate accounting for large customers and separation of accounting (business) from network (operation) customer relationships.

The usage accounting capability is described in Chapter 7 and defined in Part 630 of the CDPD System Specification and Part 1023 of the CDPD Implementor Guidelines.

The Authentication Server

The authentication server is an internal F-ES which is intended to support the authentication function in CDPD. Since this function can be implemented in many different ways, it is not required to be a separately addressable entity; it could in fact be contained within the MD-IS. Authentication is discussed in Chapter 6 and defined in Parts 406 and 640 of the CDPD System Specification.

The Directory Server

The directory server is an internal F-ES which provides support for directory services within the CDPD network. Directory services are used primarily for network management and other support services. Like the authentication server, this can be implemented in many different ways. Depending on the needs of a CDPD service provider and its customers, the directory server could support either DNS or X.500 capabilities, or both. Directory services are described in Chapter 7 and defined in Part 610 of the CDPD System Specification.

The Network Management System

The network management system is the means by which a CDPD service provider operates the CDPD network. Network management includes configuration management, fault management, performance management and other functions. Like other servers, it can be implemented in many different ways. Typically network managers run in stand-alone processors because of the resource-intensive nature of their activities. Network management is discussed in Chapter 7 and defined in Parts 700 through 750 of the CDPD System Specification.

The Message Transfer System

The message transfer system is the means by which CDPD accounting and other messaging functions are supported. CDPD accounting is supported by an X.400 messaging model, which is embodied in the message transfer system entity. It can be implemented in many ways and is discussed in Chapter 7.

3.8 CDPD Mobility Management

The central capability of CDPD is its ability to get data to a mobile device or application regardless of its location. This capability is called mobility management. CDPD mobility management is provided by two protocols: the Mobile Network Registration Protocol or MNRP and the Mobile Network Location Protocol or MNLP .

MNRP is the means by which a mobile announces its presence (including authenticating itself) to the serving system (via the End System Hello or ESH message). MNLP is the means by which the serving system notifies the home system for the mobile that it is providing service to the mobile.

The CDPD mobility management scheme is based on triangular routing of NPDUs via encapsulation from a home MD-IS (the mobile home function) to the current serving MD-IS (the mobile serving function). This packet redirection is displayed in Figure 3.9.

By using the home system as the network anchor for the M-ES, the outside world of networked applications can always access the mobile the same way they access any host–conventional routing technology is sufficient to get data packets to the home system. The CDPD mobility management scheme takes over from there, forwarding the data packets via encapsulation to the current system serving the M-ES. This solution is identical to an early Mobile IP Task Force solution, which is elegant in its simplicity.

Although there are pathological cases in which this redirection of packet traffic is highly inefficient, the CDPD specification team elected this solution because the speed of WAN networks is ever-increasing and their costs decreasing. A solution which might be more efficient in the pathological cases would be generally less efficient in accommodating moving users in the cellular environment.²¹

The CDPD mobility management scheme is defined by Parts 500 through 507 of the CDPD System Specification and is described in Chapter 4.

3.9 CDPD Radio Resource Management

Closely associated with mobility management is radio resource management or RRM, the process which ensures that the optimum radio channel is used by an M-ES. Optimum means that the channel provides the most reliable performance while not interfering with either other CDPD or cellular users. Non-interference with cellular voice customers is of paramount importance, since CDPD is a service overlaid on the existing cellular voice network.

Contrary to cellular voice systems, in which cell handoff is controlled by the system,²² the CDPD system requires that the mobile device manage its own use of radio resources. This is necessary because of the bursty nature of the packet-switched access to CDPD services; the mobile generally does not transmit long enough for the system to accurately measure its RF reception from multiple MDBSs.

The CDPD system periodically broadcasts information to assist the mobile in evaluating its current channel against alternative channels provided in adjacent cells. Prior to transmitting, the M-ES evaluates alternative channels to ensure that it uses the best channel. In a properly tuned system, the best channel will be in the same cell as the channel assigned to a similarly-located cellular voice handset.

The radio resource management scheme in CDPD also controls the power level used by the mobiles by broadcasting a so-called power product. The theory is that RF signals are reciprocal between two points; if one transceiver informs the other transceiver of the power level it is transmitting at, the second transceiver can calculate the power level it must use, based on the received signal strength.

CDPD radio resource management is defined by Part 405 of the CDPD System Specification [CDPD95] and is described in Chapter 5.

3.10 CDPD Security

CDPD provides security across the airlink sufficient to prevent casual eavesdropping and fraud. Up to 128 encryption techniques can be supported by the system; currently, only the RSA RC4 stream cipher is specified [RSA-92]. A variation of the Diffie-Hellman electronic key exchange is used to dynamically create the encryption keys to be used by the system and the mobile across the airlink.

Once in encrypted mode, the mobile must register its network address along with authentication credentials before it can receive services from the system. The credentials are compared by the M-ES’s home system with its record for that M-ES network address. This procedure establishes the authenticity of the M-ES network address to the system. The CDPD security scheme could be enhanced in the future to provide capabilities such as authentication of the system by the M-ES.

CDPD security management is defined by Part 406 of the CDPD System Specification and is described in Chapter 6.

3.11 CDPD Accounting

As mentioned earlier, CDPD defines an accounting scheme which is very general and capable of supporting many kinds of service provider and customer relationships. The basic idea is to functionally separate accounting (business) relationships from networking relationships–i.e., the accounting home for a CDPD subscriber need not be the same as the networking home for that subscriber.

The flow of accounting information is capable of supporting mobile users, while minimizing the coordination between service providers. Usage information such as packets and bytes transferred at Layer 3 is captured by the serving MD-IS. Each CDPD service provider only needs to know the presentation title of the HAD associated with each block of Layer 3 addresses which are in use by M-ESs. X.400 provides the basis for ensuring reliable delivery of accounting data from one accounting server (the SAD) to another (the HAD).

CDPD accounting is defined by Parts 630 and 1023 in the CDPD System Specification and Implementor’s Guidelines, res