Second Generation (2G) Cellular Systems

4 Second Generation (2G) Cellular Systems 4.1 Introduction As was mentioned in the previous chapter, the era of mobile telephony began with the development and operation of the First Generation (1G) of cellular systems in the late 1970s Although these systems have found widespread use and are still used nowadays, the evolution of technology has enabled the industry to move to Second Generation (2G) systems, the successors of 1G systems 2G systems overcome many of the deficiencies of 1G systems mentioned in the previous chapter Their increased capabilities stem from the fact that, contrary to 1G systems, 2G systems are completely digital Compared to analog, digital technology has a number of advantages: † Encryption Digitized traffic can be easily encrypted in order to provide privacy and security Encrypted signals cannot be intercepted and overheard by unauthorized parties (at least not without very powerful equipment) On the other hand, powerful encryption is not possible in analog systems, which most of the time transmit data without any protection Thus, digital systems provide an increased potential for securing the user’s traffic and preventing unauthorized network access † Use of error correction In digital systems, it is possible to apply error detection and error correction techniques to the user traffic Using these techniques the receiver can detect and correct bit errors, thus enhancing transmission reliability This obviously leads to signals with little or no corruption, which of course translates into (a) better voice call qualities, (b) higher speeds for data applications, and (c) efficient spectrum use, since fewer retransmissions are bound to occur when error correction and error detection techniques are used Furthermore, digital data can be compressed, which increases the efficiency of spectrum use even more It is actually this increased efficiency that enables 2G systems to support more users per base station per MHz of spectrum than 1G systems, thus allowing operators to provide service in high-density areas more economically † In analog systems, each RF carrier is dedicated to a single user, regardless of whether the user is active (speaking) or not (idle within the call) In digital systems each RF carrier is shared by more than one user, either by using different time slots or different codes per user Slots or codes are assigned to users only when they have traffic (either voice or data) to send 112 Wireless Networks The movement from analog to digital systems was made possible due to the development of techniques for low-rate digital speech coding and the continuous increase in the device density of integrated circuits Contrary to 1G systems, which employ FDMA for user separation, 2G systems allow the use of Time Division Multiple Access (TDMA) and Code Division Multiple Access (CDMA) as well Since the standards that will be discussed in this chapter employ either TDMA or CDMA (sometimes with a combination with FDMA), we briefly revisit the three approaches In order to accommodate various nodes inside the same cellular network, FDMA divides the available spectrum into subbands each of which are used by one or more users Each user is allocated a dedicated channel (subband), different in frequency from the channels allocated to other users When the number of users is small relative to the number of channels, this allocation can be static, however, for many users dynamic channel allocation schemes are necessary In cellular systems, channel allocations typically occur in pairs Thus, for each active mobile user, two channels are allocated, one for the traffic from the user to the Base Station (BS) and one for the traffic from the BS to the user The frequency of the first channel is known as the uplink (or reverse link) and that of the second channel is known as the downlink (or forward link) For an uplink/downlink pair, uplink channels typically operate on a lower frequency than the downlink one in an effort to preserve energy at the mobile nodes This is because higher frequencies suffer greater attenuation than lower frequencies and consequently demand increased transmission power to compensate for the loss By using low frequency channels for the uplink, mobile nodes can operate at lower power levels and thus preserve energy Due to the fact that pairs of uplink/downlink channels are allocated by regulation agencies, most of the time they are of the same bandwidth This fact makes FDMA relatively inefficient since in most systems the traffic on the downlink is much more heavier than that in the uplink Thus, the bandwidth of the uplink channel is not fully used TDMA is the technology of choice for a wide range of second generation cellular systems such as GSM, IS-54 and DECT TDMA divides a band into several time slots and the resulting structure is known as the TDMA frame In this, each active node is assigned one (or more) slots for transmission of its traffic Nodes are notified of the slot number that has been assigned to them, so they know how much to wait within the TDMA frame before transmission Uplink and downlink channels in TDMA can either occur in different frequency bands (FDD-TDMA) or time-multiplexed in the same band (TDD-TDMA) The latter technique obviously has the advantage of easy trading uplink to downlink bandwidth for supporting asymmetrical traffic patterns TDMA is essentially a half-duplex technique, since for a pair of communicating nodes, at a specific time, only one of the nodes can transmit Nevertheless, slot duration is so small that the illusion of two-way communication is created The short slot duration, however, imposes strict synchronization problems in TDMA systems This is due to the fact that if nodes are far from one another, the propagation delay can cause a node to miss its turn In order to protect inter-slot interference due to different propagation paths to mobiles being assigned adjacent slots, TDMA systems use guard intervals in the time domain to ensure proper operation Instead of sharing the available bandwidth either in frequency or time, CDMA places all nodes in the same bandwidth at the same time The transmissions of various users are separated through a unique code that has been assigned to each user All nodes are assigned a specific n-bit code The value of parameter n is known as the system’s chip rate The various codes assigned to nodes are orthogonal to one another, Second Generation (2G) Cellular Systems 113 meaning that the normalized inner product of the vector representations of any pair of codes equals zero Furthermore, the normalized inner product of the vector representation of any code with itself and the 1s complement of itself equals and 21, respectively Nodes can transmit simultaneously using their code and this code is used to extract the user’s traffic at the receiver Obviously, the receiver knows the codes of each user in order to perform the decoding The use of TDMA or CDMA in cellular systems offers a number of advantages: † † † † Natural integration with the evolving digital wireline network Flexibility for mixed voice/data communication and the support of new services Potential for further capacity increases as reduced rate speech coders are introduced Reduced RF transmit power (which obviously translates into increasing battery life in handsets) † Reduced system complexity (mobile-assisted handoffs, fewer radio transceivers) 4.1.1 Scope of the Chapter The remainder of this chapter describes several 2G standards D-AMPS, the 2G TDMA system that is used in North America and descends from the 1G AMPS is described in Section 4.2 CdmaOne, which is the only 2G system based on CDMA is discussed in Section 4.3 The widely used Global system for Mobile Communications (GSM) is described in Section 4.4 Section 4.5 describes IS-41, which is actually not a 2G standard but rather a protocol that operates on the network side of North American cellular networks Section 4.6 is devoted to data transmission over 2G systems and discusses a number of approaches, including GRPS, HSCSD, cdmaTwo, etc Furthermore, Section 4.6 discusses the problems faced by TCP in a wireless environment, mobileIP, an extension of the Internet Protocol (IP) that supports terminal mobility and the Wireless Access Protocol (WAP) Section 4.7 discusses Cordless Telephony (CT) including the Digital European Cordless Telecommunications Standard (DECT) and Personal Handyphone System (PHS) standards The chapter ends with a brief summary in Section 4.8 4.2 D-AMPS In an effort to increase the performance of AMPS a standard known as D-AMPS (standard name is IS-54) was developed D-AMPS maintains the 30-kHz channel spacing of AMPS and is actually an overlay of digital channels over AMPS D-AMPS was designed in a way that enables manufacturing of dual-mode (AMPS and D-AMPS) terminals Thus, the development of D-AMPS has led to a hybrid standard This is necessary to accommodate roaming subscribers, given the large embedded base of AMPS equipment The main difference between AMPS and D-AMPS is that the latter overlays digital channels over the 30 kHz carriers of AMPS Each such digital channel can support three times the users that are supported by AMPS with the same carrier Thus, D-AMPS can be seen as an overlay on AMPS that ‘steals’ some carriers and changes them to carry digital traffic Obviously, this does not affect the underlying AMPS network, which can continue to serve regular AMPS users In fact, each D-AMPS MS initially accesses the network via the traditional AMPS analog control channels Then the MS can make a request to be assigned a 114 Wireless Networks digital channel and if such a channel is available, it is allocated to the D-AMPS MS; otherwise the MS will operate in AMPS mode Finally, as far as handoffs are concerned, D-AMPS supports Mobile Assisted Handoff (MAHO) MSs make measurements of the signal strength from various neighboring BSs and report these measurements to the network, which uses this information to decide whether a handoff will be performed, and to which BS The difference with AMPS is that in AMPS, MSs not perform signal strength measurements Rather these measurements are made by the BSs as can be seen in Chapter from the sequence of events that describes a handoff in AMPS Both D-AMPS and its successor IS-136 support voice as well as data services Supported speeds for data services are up to 9.6 kbps 4.2.1 Speech Coding D-AMPS utilizes Vector-Sum Excited Linear Predictive Coding (VSELP) This method breaks the PCM digitized voice bit-stream into parts corresponding to 20 ms speech intervals Each such bitstream forms the input to a codebook whose output replaces the input bitstream with the codeword that is closest to the actual value of the input bitstream This codeword is what will be transmitted over the wireless link Each codeword will be later provided with protection against the fading wireless environment This protection comprises: (a) a CRC operation on the most significant bits of each speech coder output; (b) convolutional coding to protect the most vulnerable bits of the speech coder output; and (c) interleaving the contents of each coder output over two time slots Each digital channel provides a raw bit rate of 48.6 kbps, achieved using p/4 DQPSK 4.2.2 Radio Transmission Characteristics D-AMPS operates at the same frequency band with AMPS Uplink digital channels occur in the 824–849 band and downlink ones in the 869–894 band Each digital channel is organized into 40 ms frames and each frame comprises six 6.67 ms time slots Each user can use either slots (either and 4, and or and 6) or slot within each frame The first configuration is used with the full-rate voice codec, producing transmission of actual voice information up to 7.95 kbps (5.05 kbps with Forward Error Correction (FEC)) The second configuration is used with the half-rate voice codec producing transmission of actual voice information up to 3.73 kbps (2.37 kbps with FEC) The corresponding values for data speeds are 9.6 without FEC and 3.4 kbps with FEC The overall access method is shown in Figure 4.1 It can be seen that the uplink and downlink slots have a slightly different internal arrangement The slot parts are described below: † The training part This part has enables the MS and BS to ‘learn’ the channel This is because a signal is bound to arrive at the receiver over a number of paths due to reflections from objects in the environment Thus, equalization is used to extract the desired signal from the unwanted reflections The IS-54 standard also provides for an adaptive equalizer to mitigate the intersymbol interference caused by large delay spreads, but due to the relatively low channel rate (24.3 kbaud), the equalizer will be unnecessary in many situations Second Generation (2G) Cellular Systems Figure 4.1 115 Structure of IS-54 slot and frame † The traffic (data) parts These parts carry user traffic, either voice or data-related As the channels are digital, user traffic can be encoded or encrypted, thus the whole traffic part is not always entirely dedicated to the transfer of user data but also contains encryption/ coding overhead † The guard part This provides guard intervals in the time domain in order to separate a slot from the previous slot and the next slot The need for these parts is due to propagation delay, which can cause a node to miss its slot when nodes are very far from one another † The ramp bits These are used to ramp up and down the signal during periods where the signal is in transition † The control parts These carry control signaling via the channel shown in parentheses Uplink and downlink frames are offset in time by 8.518 ms As the uplink and downlink occur in different carriers, this offset allows an MS to operate at half-duplex mode since with this arrangement MSs never transmit and receive at the same time 4.2.3 Channels D-AMPS reuses the AMPS channels described in Chapter However, it also introduces some new digital channels The channel definitions for AMPS are as follows: † Forward Control Channel (FOCC) Same as AMPS † Forward Voice Channel (FVC) Same as AMPS The analog channel carrying voice traffic from the BS to the MS † Forward Digital Traffic Channel (FDTC) This is a BS to MS channel carrying digital traffic (both user data and control data) It consists of the Fast Associated Control Channel (FACCH) and Slow Associated Control Channel (SACCH) FACCH is a blank-and-burst operation, meaning that the traffic channel is pre-empted by control signaling SACCH is a Wireless Networks 116 continuous channel also associated with control signaling However, it differs from FACCH in that a certain amount of bandwidth is allocated a priori to SACCH † Reverse Control Channel (RECC) Same as AMPS † Forward Voice Channel (RVC) Same as AMPS The analog channel carrying voice traffic from the MS to the BS † Reverse Digital Traffic Channel (RDTC) This is an MS to BS channel carrying digital traffic (both user data and control data) It consists of a FACCH and SACCH 4.2.4 IS-136 IS-136 is an upgrade of AMPS that also operates in the 800 MHz bands However, there are planned upgrades to the 1900 band While D-AMPS is a digital overlay over AMPS, IS-136 is a fully digital standard IS-136 has much in common with GSM (such as convolutional coding, interleaving, etc.) However, their air interfaces are incompatible Due to the similarities between GSM and IS-136, we not make a detailed presentation of the former Rather, we present the organization of the air interface of IS-136, which as can be seen from Figure 4.2 builds on top of that of D-AMPS Figure 4.2 Structure of IS-136 slot, frame and multiframe Second Generation (2G) Cellular Systems 117 4.3 cdmaOne (IS-95) In 1993 cdmaOne, a 2G system also known as IS-95, has been standardized and the first commercial systems were deployed in South Korea and Hong Kong in 1995, followed by deployment in the United States in 1996 cdmaOne utilizes Code Division Multiple Access (CDMA) In cdmaOne, multiple mobiles in a cell, whose signals are distinguished by spreading them with different codes, simultaneously use a frequency channel Thus, neighboring cells can use the same frequencies, unlike all other standards discussed so far cdmaOne is incompatible with IS-136 and its deployment in the United States started in 1995 Both IS136 and cdmaOne operate in the same bands with AMPS cdmaOne is designed to support dual-mode terminals that can operate either under an cdmaOne network or an AMPS network cdmaOne supports data traffic at rates of 4.8 and 14.4 kbps 4.3.1 cdmaOne Protocol Architecture Figure 4.3 shows the protocol architecture of the lower two layers of cdmaOne and its correspondence to the layers of the OSI model Layer obviously deals with the actual radio transmission, frequency use, etc These issues will be discussed briefly in the next subsection Layer offers a best effort delivery of voice and data packets The MAC sublayer of this layer also performs channel management This sublayer maintains a finite-state Figure 4.3 cdmaOne protocol architecture Wireless Networks 118 Figure 4.4 cdmaOne MAC states machine with the two states shown in Figure 4.4 Reflecting the status of packet or circuit data transmissions, a different machine is maintained for each transmission cdmaOne mobiles maintain all their channels and go to the dormant state after a ‘big’ timeout (big period during which the MS is idle) In this state, mobiles not maintain any channels Thus, there exists no mechanism for sending user data while in the dormant state; rather the mobile must request channel assignment, thus incurring an overhead for infrequent data bursts Upon having traffic to send, they return to the active state where channels are assigned to the mobile Finally, data originating from different sources are multiplexed and handed for transmission to the physical layer 4.3.2 Network Architecture-Radio Transmission As mentioned above, cdmaOne reuses the AMPS spectrum in the 800 MHz band cdmaOne uses a channel width of 1.228 MHz both on the uplink and downlink Therefore, 41 30 kHz AMPS channels are grouped together for cdmaOne operation A significant difference between cdmaOne and the other cellular standards stems from the fact that in cdmaOne, the same frequency is reused in all cells of the system This leads to a frequency reuse factor of and is due to the fact that cdmaOne identifies the transmissions of different mobiles via the different spreading codes that identify each mobile Both cdmaOne BSs and MSs utilize antennas that have more than one element (RAKE receivers) in order to combat the fading wireless medium via space diversity The use of CDMA for user separation imposes the need for precise synchronization between BSs in order to avoid too much interference This synchronization problem is solved via the use of the Global Positioning System (GPS) receivers at each BS GPS receivers provide very accurate system timing Once the BSs are synchronized, it is their responsibility to provide timing information to the MSs as well This is achieved by conveying from the BSs to the MSs a parameter identifying the system time, offset by the one way or round-trip delay of the transmission In this way, it is ensured that BSs and MSs remain synchronized Finally, as far as the network side is concerned, cdmaOne utilizes the IS-41 network protocol that is described in a later section 4.3.3 Channels 4.3.3.1 Downlink Channels Downlink channels are those carrying traffic from the BS to the MSs The cdmaOne downlink is composed of 64 channels These logical channels are distinguished from each other by using different CDMA spreading codes, W0 to W63 The spreading code is an orthogonal code, or called Walsh function The cdmaOne downlink comprises Second Generation (2G) Cellular Systems 119 common control and dedicated traffic channels, the most important of which are summarized below † Pilot channel This channel provides the timing information to the MS regarding the downlink and signal strength comparisons between BSs The actual content of the pilot channel is a continuous stream of 0s at a rate of 19.2 kbps † Sync channel This optional channel is used to transmit synchronization messages to MSs The sync channel is usually present, but may be omitted in very small cells In that case, a mobile will get synchronization information from a neighboring cell The channel operates at a rate of 1200 bps † Paging channel This is an optional channel There are up to seven paging channels on the downlink which can carry four major types of messages: overhead, paging, order, and channel assignment This channel operates at one of the following data rates: 2400, 4800, or 9600 bps † Traffic channels Traffic channels carry user data, at 1200, 2400, 4800, or 9600 bps All traffic channels are spread by a long code (PN code), which provides discrimination between mobile stations Except for the pilot channel, all channels on the downlink are coded and interleaved The vocoder uses the Code Excited Linear Predictive (CELP) algorithm The vocoder is sensitive to the amount of speech activity present on its input, and its output will appear at one of four available rates The bit rate of the vocoder changes in proportion to how active the speech input may be at any time The rate may vary every 20 ms The output of the vocoder is first encoded by the convolutional encoder into a constant 19.2 ksps (1000 symbols/second) binary stream, each data bit is represented by two symbols, with one redundancy bit inserted (rate 1/2) The output of the convolutional coder is input to a repetition function, which is used to repeat the data pattern of reduced rates (1200, 2400, or 4800 bps) to form a constant output rate of 19.2 ksps The encoded binary stream is then interleaved randomly by the interleaver (at an interval of 20 ms) into frames (frame interleaving) The purpose of using interleaving is to combat the multipath fading environment, which causes burst errors on the radio channel The output of the interleaver is then modulo-2-added to a 19.2 kcps (1000 chips/second) scrambling code from a 1/64 decimator The decimator selects every 64th bit from a ‘long code’ generator running at 1.2288 Mcps The ‘long code’ generator creates a very long codes (2 42 bits) based on the user-specific information, such as the Mobile Identity Number (MIN) or the user’s Electronic Serial Number (ESN) Long codes provide a very high level of security, because of the long length This information is also made available to the network when the MS sends its handshaking information to the BS After modulated by a long code, the resulting 19.2 ksps data stream is spread by a Walsh function running at a rate of 1.2288 Mcps Walsh spreading provides every channel with a unique identification number Finally, the spread 1.2288 Mcps signal is spread one more time by a short code running at 1.2288 Mcps Short code is also a Pseudonoise (PN) code, and is 15 bits in length All base stations use the same short code, but with different offsets There exist 512 different offsets, thus this scheme can uniquely identify 512 different cdmaOne BSs A mobile can easily distinguish transmissions from two different base stations by their shortcode offsets The resulting signal is transmitted over the wireless medium via Quadrature Phase Shift Keying (QPSK) modulation 120 Wireless Networks 4.3.3.2 Uplink Channels There are two types of uplink channels, access and traffic There can be up to 32 access channels on the uplink, each of which operates at 4800 bps These channels are used by MSs to initiate calls and respond to paging messages An access channel contains information that the BS needs to properly log the mobile into service There can be up to 62 traffic channels on the uplink These are used to carry user data The payload of a traffic channel comes from a variable rate vocoder with four possible output rates: 9600, 4800, 2400 and 1200 bps The data from the vocoder is convolutionally encoded by a 1/3 rate encoder, which adds two redundancy bits to each data bit, thus multiplying the data rate by three, resulting in a binary stream of rate 28.8 ksps The encoded data is interleaved randomly before entering the block encoder, which examines the content of the input data stream in a 6-bit segment and replaces the 6-bit segment with the corresponding 64-bit Walsh function After leaving the block encoder, the data stream is spread by the long code and short codes, respectively The resulting spread data stream has a rate of 1.2288 Mcps and is transmitted over the wireless medium via Offset Quadrature Phase Shift Keying (OQPSK) modulation OQPSK provides more Forward Error Correction (FEC) than QPSK since MSs cannot coordinate their transmissions as efficiently as BSs 4.3.4 Network Operations 4.3.4.1 Handoff There are four handoff categories in cdmaOne, soft, softer, hard and idle handoff A handoff occurs when a MS detects a pilot channel of higher quality than that of the BS currently serving the MS In soft handoff, a link is set up to the new BSs before the release of the old link This ensures reliability, as the new BS may be too crowded to support the roaming mobile terminal or the link to the new BS may degrade shortly after establishment However, the mobile terminal should be able to communicate with two different BSs at the same time Thus, soft handoff causes increased complexity at the mobile terminals since it demands the capability of supporting two links with different BSs at the same time When a soft handoff takes place between sectors inside the same cell, it is also known as softer handoff Hard handoff is relatively simpler than soft handoff since the link to the old BS is released before establishment of the link to the BS of the new cell However, it is somewhat less reliable than soft handoff Finally, the cdmaOne specification defines the idle handoff The main difference of idle handoff with the previous handoff types is that in the previous types the MS being handed off is involved in an active call However, in an idle handoff the MS is in idle mode 4.3.4.2 Power Control Power control is critical in cdmaOne due to the fact that the use of CDMA imposes the need for all MS transmissions to reach the BS with strength difference of no more than dB If the signal received from a near user is stronger than that from a far user, the former signal will be swamped out by the latter This is known as the ‘near-far’ problem Another reason for implementing power control is to increase capacity Power control is implemented on both the uplink and downlink On the uplink, both open-loop and closed-loop power control is used (the principle of ... and multiframe Second Generation (2G) Cellular Systems 117 4.3 cdmaOne (IS-95) In 1993 cdmaOne, a 2G system also known as IS-95, has been standardized and the first commercial systems were deployed... channel rate (24.3 kbaud), the equalizer will be unnecessary in many situations Second Generation (2G) Cellular Systems Figure 4.1 115 Structure of IS-54 slot and frame † The traffic (data) parts... system’s chip rate The various codes assigned to nodes are orthogonal to one another, Second Generation (2G) Cellular Systems 113 meaning that the normalized inner product of the vector representations