Cisco Systems, Inc., Exhibit 1107, Page 1
`
`

`
`CDMA Systems
`Engineering Handbook
`
`Jhong Sam Lee
`Leonard E. Miller
`
`ile Communications Library,
`gook.
`
`Artech House
`Boston •London
`
`Cisco Systems, Inc., Exhibit 1107, Page 2
`
`

`
`Library of Congress Cataloging-in-Publication Data
`Lee, Jhong S.
`CDMA systems engineering handbook /Jhong S. Lee, Leonard E. Miller.
`p. cm. — (Artech House mobile communications library)
`Includes bibliographical references and index.
`ISBN 0-89006-990-5 (alk. paper)
`1. Code division multiple access.
`2. l~lobile communication systems.
`I. Miller, Leonard E. II. Title. III. Series.
`TK5103.45.L44 1998
`621.382—dc21
`
`98-33846
`CIP
`
`British Library Cataloguing in Publication Data
`Lee, Jhong S.
`CDMA systems engineering handbook.— (Artech House mobile communications
`library)
`1. Code division multiple access—Handbooks, manuals, etc.
`I. Title II. Miller, Leonard E.
`621.3'84'56
`
`ISBN 0-89006-990-5
`
`Cover design by Lynda Fishbourne
`
`OO 1998 J. S. Lee Associates, Inc.
`
`All rights reserved. Printed and bound in the United States of America. No part of this
`book may be reproduced or utilized in any form or by any means, electronic or me-
`chanical, including photocopying, recording, or by any information storage and re-
`trieval system, without permission in writing from the publisher.
`All terms mentioned in this book that are known to be trademarks or service marks
`have been appropriately capitalized. Artech House cannot attest to the accuracy of this
`information. Use of a term in this book should not be regarded as affecting the validity
`of any trademark or service mark.
`
`International Standard Book Number: 0-89006-990-5
`Library of Congress Catalog Card Number: 98-33846
`
`10987654
`
`Cisco Systems, Inc., Exhibit 1107, Page 3
`
`

`
`nilies during the process of
`ledge the contributions of
`(former employees of J. S.
`alsh functions, interleaving
`and random number gen-
`
`riven to the first author of
`cures in Korea, by Dr. Jung
`s appointed by the Minister
`ce the entire Korea CDMA
`en President of Electronics
`'R~ (now President of the
`~; Dr. Hang Gu Bahk, then
`.t of Hyundai Electronics
`.won, founding President of
`President of STI; Mr. Byung
`ations Systems Division of
`;ommunications, who served
`n in their respective roles did
`s first and most successful
`ch as of this writing brought
`ribers since the beginning of
`leaders and engineers of the
`;DMA apractical reality in
`Korea Telecom Freetel, Inc.;
`Telecom, Ltd.; and Samsung
`
`Jhong Sam Lee
`Leonard E. Miller
`Rockville, Maryland USA
`September, 1998
`
`~
`
`Introduction and Review of Systems Analysis
`Basics
`
`1.1
`
`Introduction
`The purpose of this book is to provide a clear understanding of code-division
`multiple access (CDMA) technology and build a solid understanding of the
`technical details and engineering principles behind the robust new IS-95
`digital cellular system standard. The book is intended to help practicing
`cellular engineers better understand the technical elements associated with
`CDMA systems and how they are applied to the IS-95 standard, which was
`developed in response to the requirement for the design of a second-genera-
`tion cellular telephone system. The CDMA cellular system uses state-of-the-
`art digital communications hardware and techniques and is built on some of
`the more sophisticated aspects of modern statistical communications theory.
`The book is designed to be self-contained in that it includes in this chapter all
`the systems analysis basics and statistical tools that are pertinent to the tech-
`nical discussions in the later chapters.
`The "second-generation" means digital, as opposed to the "first-
`generation" analog system. The current U.S. analog cellular system, known
`as the Advanced Mobile Phone System (AMPS), operates in afull-duplex
`fashion using frequency-division duplexing (FDD), with a 25-MHz bandwidth
`in each direction over the following frequency allocations:
`• From mobile to base station: 824-849 MHz;
`. From base station to mobile: 869-894 MHz.
`The Federal Communications Commission (FCC) further divided the 25-
`MHz bandwidth equally between two service providers, known as the "A"
`(wire) and the "B" (nonwire) carriers, each with 12.5 MHz of spectrum
`allocated for each direction.
`
`Cisco Systems, Inc., Exhibit 1107, Page 4
`
`

`
`,\
`
`2
`
`CDMA Systems Engineering Handbook
`
`~A
`~
`
`'-
`
`~
`
`f
`
`In AMPS, each channel occupies 30 kHz of bandwidth in a frequency-
`division multiple access (FDMA~ system, using analog frequency modulation
`(FM) waveforms. The frequencies that are used in one cell area are reused in
`another cell area at a distance such that mutual interference gives acarrier-to-
`interference power ratio of no less than 18 dB. Given this performance re-
`quirement and the fact that in the mobile radio environment the attenuation
`of carrier power usually is proportional to the fourth power of the distance
`from the emitter to a receiver, the analog cellular system utilizes seven-cell
`clusters, implying a frequency reuse factor of seven. The resulting capacity is
`then just one call per 7 x 30 kHz = 210 kHz of spectrum in each cell, and in
`the total of 12.5 MHz allocated there can be no more than 60 calls per cell.l
`In 1988, the Cellular Telecommunications Industry Association (CTIA)
`released cellular service requirements for the next-generation (second-genera-
`tion) digital cellular system technology, known as a users' performance
`requirements (UPR) document. The key requirements included a tenfold
`increase in call capacity over that of AMPS, a means for call privacy, and
`compatibility with the existing analog system. The compatibility require-
`ment arose from the fact that the FCC did not allocate a separate band for the
`digital system, so the second-generation system must operate in the same band
`as AMPS.
`In 1989, a committee of the Telecommunications Industry Association
`(TIA) formulated an interim standard for asecond-generation cellular system
`that was published in 1992 as IS-54 [1]. In that standard, which became the
`first U.S. digital cellular standard, the committee adopted atime-division
`multiple access (TDMA) technology approach to the common air interface
`(CAI) for the digital radio channel transmissions. The IS-54 TDMA digital
`cellular system employs digital voice produced at 10 kbps (8 kbps plus over-
`head) and transmitted with ~r/4 differentially encoded quadrature phase-shift
`keying (~r/4 DQPSK) modulation. The design envisioned noncoherent
`demodulation, such as by using alimiter-discriminator or a class of differential
`phase detectors. Because the IS-54 system permits 30 kHz/10 kbps = 3 callers
`per 30-kHz channel spacing, the increase of capacity over AMPS is only a
`factor of three (180 calls per cell), and the TDMA digital cellular system so far
`falls short of meeting the capacity objective of the UPR.
`Immediately following the emergence of the IS-54 digital cellular stand-
`ard, Qualcomm, Inc., in 1990 proposed a digital cellular telephone system
`based on CDMA technology, which in July 1993 was adopted as a second
`
`1 As is shown in Section 3.2.5, the actual capacity is lower than 60 calls per cell
`because of the allocation of some channels to signaling traffic.
`
`Cisco Systems, Inc., Exhibit 1107, Page 5
`
`

`
`andbook
`
`Introduction and Review of Systems Analysis Basics
`
`3
`
`z of bandwidth in a frequency-
`.g analog frequency modulation
`ed in one cell area are reused in
`al interference gives acarrier-to-
`B. Given this performance re-
`lio environment the attenuation
`ie fourth power of the distance
`ellular system utilizes seven-cell
`seven. The resulting capacity is
`of spectrum in each cell, and in
`o more than 60 calls per cell.l
`ins Industry Association (CTIA)
`next-generation (second-genera-
`nown as a users' performance
`-equirements included a tenfold
`i, a means for call privacy, and
`:m. The compatibility require-
`~t allocate a separate band for the
`m must operate in the same band
`
`nunications Industry Association
`second-generation cellular system
`that standard, which became the
`nmittee adopted atime-division
`Bch to the common air interface
`~sions. The IS-54 TDMA digital
`ced at 10 kbps (8 kbps plus over-
`y encoded quadrature phase-shift
`design envisioned noncoherent
`riminator or a class of differential
`rmits 30 kHz/ 10 kbps = 3 callers
`~f capacity over AMPS is only a
`AMA digital cellular system so far
`>f the UPR.
`of the IS-54 digital cellular stand-
`digital cellular telephone system
`'~y 1993 was adopted as a second
`
`acity is lower than 60 calls per cell
`ialing traffic.
`
`U.S. digital cellular standard, designated IS-95 [2]. Using spread-spectrum
`signal techniques, the IS-95 system provides a very high capacity, as will be
`convincingly shown in this book, and is designed to provide compatibility
`with the existing AMPS, in compliance with the specifications of the UPR
`document.
`
`1.1.1 Multiple Access Techniques
`
`The first cellular generation's AMPS and the second generation's IS-54 and IS-
`95 are generic examples of the three basic categories of multiple access (MA)
`techniques:
`
`~ FDMA;
`
`• TDMA;
`
`• CDMA.
`
`The three basic techniques can be combined to generate such hybrids as
`combined frequency division and time division (FD/TDMA), combined
`frequency division and code division (FD/CDMA), and others [3]. The IS-95
`system employs FD/CDMA techniques, whereas the IS-54 system uses an
`FD/TDMA method. All these multiple access strategies are competing
`techniques, each aiming at distributing signal energy per access within the
`constrained time-frequency plane resource available.
`In an FDMA system, the time-frequency plane is divided into, say, M
`discrete frequency channels, contiguous along the frequency axis as depicted
`in Figure 1.1. During any particular time, a user transmits signal energy in
`one of these frequency channels with a 100% duty cycle. In a TDMA system,
`the time-frequency plane is divided into M discrete timeslots, contiguous
`along the time axis as shown in Figure 1.2. During any particular time, a user
`transmits signal energy in one of these timeslots with low duty cycle. In a
`CDMA system, the signal energy is continuously distributed throughout the
`entire time-frequency plane. In this scheme, the frequency-time plane is not
`divided among subscribers, as done in the FDMA and TDMA systems,
`Instead, each subscriber employs a wideband coded signaling waveform [3] as
`illustrated in Figure 1.3.
`Having defined the three MA techniques employed by AMPS, IS-54,
`and IS-95 systems, one may wonder why the capacities of these systems differ
`from one another! Is it the inherent property of the MA technique that sets
`
`Cisco Systems, Inc., Exhibit 1107, Page 6
`
`

`
`~~ ~
`'~1
`
`~~ V
`
`4
`
`CDhl1A Systems Engineering Handbook
`
`Base station
`
`fM
`
`W
`
`U;er 1 (U~ r 1 (U3er 1...( U~r
`l ~
`1 ~
`1 ~
`
`f, ~
`
`T ~
`
`I
`I
`~I~I r I r~ ..... I
`~T
`fl f2 fj fM Frequency
`
`'Fixed frequency assignments
`"100% duty cycle
`
`W
`
`~~
`
`Figure 1.1 Frequency-Division Multiple Access.
`
`Base station
`
`~ T
`
`►f
`
`U;er
`
`U2er
`
`U3er .~. User
`
`W
`
`t2 t3 ty Time
`t~
`~--- One frame
`►I
`
`Figure 1.2 Time•Division Multiple Access.
`
`1~ [z
`
`1M
`
`*Fixed timeslot assignments
``Low duty cycle
`
`one apart from another? Before focusing on the reasons for the differences,
`let us consider an ideal, hypothetical situation as follows: Suppose that each
`of the three MA systems has bandwidth W MHz and each user employs an
`uncoded bit rate R6 = 1/T6, where T6 is the bit duration. Let us also assume
`that each MA system employs orthogonal signaling waveforms, as suggested
`in Figures 1.1 to 1.3. Then the maximal number of users is given by
`
`Cisco Systems, Inc., Exhibit 1107, Page 7
`
`

`
`Introduction and Review of Systems Analysis Basics
`
`5
`
`Base station
`
`code CM
`
`code C3 W
`code Cz
`
`u ~ r u2 r U 3 r ~~ uMr code C~
`
`r~~~~~~~~~~
`
`~— W—►~ Frequency ~— T
`
`C,
`
`CZ C3 ••• CM
`
`Codes
`
`Figure 1.3 Code-Division Multiple Access.
`
`*Fixed code assignments
`"100% duty cycle
`
`M =capacity < R = WT6 (1.1a)
`6
`
`Let us now assume that the received power of each user in any MA system is
`S,.. Then the total received power PT is
`
`Pr = MSr (1.1b)
`
`The required signal-to-noise power ratio (SNR) or E6/No (bit energy-to-noise
`spectral density ratio) is assumed to be equal to the actual value, giving
`
`Eb
`
`E6
`
`N~ req
`
`'~~ actual
`
`ST/Rn Pr~M
`NOR6
`NO
`
`from which we obtain
`
`M _ ~Pr~No)
`Rn • ~En~No~req
`
`(1.1c)
`
`(1.1d)
`
`Therefore, in an ideal situation, in principle each MA technique can deliver
`equal capacity, namely
`
`MFDMA = MTDMA = MCDMA = `PrIN~J ~1.1e~
`Rn • ~Eb~No~req
`
`Cisco Systems, Inc., Exhibit 1107, Page 8
`
`

`
`6
`
`1
`
`CDMA Systems Engineering Handbook
`
`Yet, in reality each MA system as used in the cellular telephone industry does
`not deliver the same capacity. In this introduction, we can offer some
`answers as to why a certain system does not deliver the full capacity it could
`be capable of providing, but not all the answers until we penetrate deep into
`the succeeding chapters.
`As stated previously, the FDMA-based AMPS falls far short of meeting
`the orthogonal signaling requirements of the ideal FDMA system, in that the
`signal at a given frequency of one cell cannot tolerate interference from a
`signal at the same frequency in another cell unless a mutual distance
`separation is such that the relative power ratio (carrier-to-interference power
`ratio) is above 18 dB. This is one of the mitigating realities for the low
`capacity of the signal design of the first-generation AMPS. The IS-54 system
`enhances AMPS capacity by about three times, far short of meeting the UPR
`objectives. The three types of MA techniques used in the FDMA, TDMA,
`and CDMA cellular communications systems may be compared with respect
`to their use of time and frequency resources as shown in Figure 1.4.
`
`Power
`
`Time
`
`Frequency
`
`CDMA
`
`Figure 1.4 Multiple access techniques of FDMA (AMPS), TDMA (IS•541, and CDMA (IS•951.
`
`Spread-spec
`frequency 1
`achieve a
`
`z In a me
`power for
`in which
`
`Cisco Systems, Inc., Exhibit 1107, Page 9
`
`

`
`~
`
`yes
`me
`uld
`ito
`
`ing
`the
`1 a
`ice
`ver
`ow
`em
`PR
`~A~
`ect
`
`~
`
`cy
`
`Introduction and Review of Systems Analysis Basics
`
`The question then is: does CDMA meet the UPR's requirement in
`terms of capacity being at least ten times the capacity of AMPS? The authors
`believe strongly that the answer is a positive "yes"! This book is written not
`only to provide explanations of the basic technology involved in the IS-95
`CDMA system design, but also to prove by analysis the reasons why the
`CDMA system can meet such high-capacity requirements. Chapter 11 offers
`ways of meeting optimality requirements of an IS-95 system, in terms of an
`optimal forward link power control scheme that will provide the high
`capacity the CDMA system is capable of delivering.
`The reason for the high capacity of the IS-95 system is not merely the
`near-orthogonality of the signals in any user channels, but also the system's
`exploitation of the fractional duty cycle of human speech voice activity, as
`well as the employment of three or more directional sector antennas that
`increase cell capacity directly through full frequency-time plane reuse, features
`unique to the CDMA system and not in common with either FDMA or
`TDMA systems. In the IS-95 CDMA system, each user is given one out of a
`set of orthogonal codes with which the data is spread, yielding code
`orthogonality. The orthogonality property allows the multiple users to be
`distinguished from one another.2 Although users operate on the same fre-
`quency at the same time, the spreading of the baseband signal spectrum allows
`interference from other users to be suppressed, which increases the capacity of
`the CDMA technique.
`The IS-95 system, conceived and promoted by Qualcomm, Inc., is an
`elegant example of a commercial application of aspread-spectrum system,
`which has opened a new era of spread-spectrum wireless communications in
`nonmilitary applications. It seems appropriate to say that, if one ever wanted
`to see a communications system that is designed and built using most of the
`modern communications and information theoretic disciplines, the IS-95
`CDMA system could be a good example of it [4].
`
`1.1.2 Spread-Spectrum Techniques
`
`Spread-spectrum techniques involve the transmission of a signal in a radio
`frequency bandwidth substantially greater than the information bandwidth to
`achieve a particular operational advantage. Once only of interest to the
`
`2 In a mobile environment, multipath receptions may contribute to the interference
`power for each mobile station. This subject will be dealt with in detail in Chapter 10
`in which CDMA systems engineering issues are discussed.
`
`Cisco Systems, Inc., Exhibit 1107, Page 10
`
`

`
`1
`
`8
`
`CDMA Systems Engineering Handbook
`
`military, spread-spectrum techniques have now been adapted for commercial
`applications, using nearly all the operational advantages that rendered the
`techniques important to the military, as suggested in Table 1.1.
`The manner in which spread-spectrum signals can be processed to yield
`gains for interference rejection can be understood by calculating the jamming
`margin for aspread-spectrum system. Let the following parameters be
`defined:
`
`S =received power for the desired signal in W.
`
`J =received power for undesired signals in W (jamming, other
`multiple access users, multipath, etc.).
`
`R = 1/T6 =Data rate (data signal bandwidth in Hz).
`
`W =spread bandwidth in Hz.
`
`E6 =received energy per bit for the desired signal in W-sec.
`
`No =equivalent noise spectral power density in W/Hz.
`
`Then the ratio of the equivalent "noise" power J to S is
`
`J NoW WT6 W/R
`S Et~7'n
`Eb~No Eb~No
`
`(1.2a)
`
`When the value of E6/No is set to that required for acceptable performance of
`the communications system, then the ratio J/S bears the interpretation of a
`damming margin:
`
`Table 1.1 Applications of spread spectrum systems
`
`Purposes
`Antijamming
`Multiple access
`Low detectability
`Message privacy
`Selective calling
`Identification
`Navigation
`Multipath protection
`Low radiated flux density
`
`Military Commercial
`3
`3
`3
`3
`3
`3
`3
`3
`3
`3
`3
`
`3
`3
`3
`3
`3
`3
`
`Cisco Systems, Inc., Exhibit 1107, Page 11
`
`

`
`ercial
`d the
`
`yield
`iming
`ors be
`
`other
`
`~
`
`~
`
`(1.2a)
`
`mce of
`ra of a
`
`Introduction and Review of Systems Analysis Basics
`
`J =tolerable excess of interference over desired signal power
`S
`W ~R
`
`(1.2b)
`
`~E6~N~~req
`
`or
`
`Margin (dB) = R (dB) — (~o I (dB)
`/ req
`
`(1.2c)
`
`The quantity W~R is called the spread-spectrum processing gain. Note that,
`for a system that is not spread in bandwidth (i.e., W = R), the value of
`E6~No is numerically equal to the signal-to-noise power ratio (SNR).
`
`Example 1.1 Let the information bandwidth be R = 9,600 Hz, corre-
`sponding to digital voice, and let the transmission bandwidth be W = 1.2288
`MHz. If the required SNR is 6 dB, what is the antijam margin?
`Solution: Applying (1.2c~, we have
`
`(1.2288 x 1061 _ 6 = 21.1 — 6 = 15.1 dB
`Margin = 101og10
`9.6 x 103
`
`(1.2d)
`
`The implication of this example is that the communicator can achieve its
`objective (by attaining 6 dB SNR) even in the face of interference (jamming)
`power in excess of 101'5 = 32 times (due to jamming margin the communica-
`tor's SNR requirement, due to the processing gain of W/R = 128. In other
`words, 15.1 dB is a leftover (margin), and the question is how one can use this
`jamming margin. If there is a second communicator in the spread-spectrum
`bandwidth, it can "use up" the entire margin of 15.1 dB if it communicates
`with the base station 15.1 dB closer, relative to the first user. Thus the idea is
`to expend the jamming margin of 15.1 dB by accommodating the maximal
`number of communicators in an MA communications system. In a CDMA
`communications system, the cochannel communicators, occupying the
`frequency-time plane simultaneously, account for the interference (jamming)
`power. This implies that, if every user in the spread-spectrum bandwidth
`supplies the identical amount of signal power to the base station antenna
`through a perfect power control scheme, regardless of location, then
`101.5 = 32 other MA users can be accommodated.
`The example considered above explains the essence of a CDMA system
`such as IS-95. The digital voice data at the rate of 9, 600 bps is coded for
`channel error protection and then spread-spectrum modulated, resulting in a
`
`Cisco Systems, Inc., Exhibit 1107 Page 12
`
`

`
`10
`
`CDMA Systems Engineering Handbook
`
`wideband signal at 1.2288 x 106 chips per sec3 and is transmitted to a where 41
`receiving party in the presence of other users. The basic operation that needs spectral
`to be performed to convert the information data signal to a wideband spread- power ra
`spectrum signal and to retrieve it at the receiver is accomplished simply by a
`multiplication process, as indicated in Figure 1.5. We devote all of Chapter 6
`to the theory and practice of the spreading code sequences and explain their
`use in the IS-95 system.
`
`1.1.3 IS-95 System Capacity Issues
`
`The capacity of a CDMA system is proportional to the processing gain of the
`system, which is the ratio of the spread bandwidth to the data rate. This fact
`may be shown as follows: assuming first that the system in question is isolated
`from all other forms of outside interference (i.e., a single cell), the carrier
`power C - S = E6/Tb = RE6 as in (1.2a). Similarly, analogous to the
`jamming power in (1.2a), the interference power at the base station receiver
`may be defined as
`
`I=W•No
`
`(1.3)
`
`Intertering signal
`Spread signal /~ bother users)
`bandwidth W
`
`~L►
`
`Data signal
`rate R
`
`X
`
`X
`
`Filter
`
`Bandwidth = 2
`
`Recovered
`data signal
`
`Identical
`Spreading signal ~--►Despreading signal
`(code sequence)
`(code sequence)
`
`• Data signal is spread at transmitter (first multiplication).
`
`• Data signal is recovered at receiver (second multiplication).
`
`• Interfering signal is spread at receiver (one multiplication).
`
`Figure 1.5 Basic spread-spectrum operations.
`
`3 The PN sequence code rate is described as the "chip rate," while the
`sequence is described as "bits."
`
`Cisco Systems, Inc., Exhibit 1107, Page 13
`
`

`
`Introduction and Review of Systems Analysis Basics
`
`11
`
`fitted to a
`that needs
`nd spread-
`mply by a
`Chapter 6
`plain their
`
`where W is the transmission bandwidth and No is the interference power
`spectral density. Thus a general expression for the carrier-to-interference
`P°wer ratio for a particular mobile user at the base station is given by
`
`C R • E6 Eb/No
`I W • N W /R
`0
`
`(1.4)
`
`The quantity Eb/No is the bit energy to noise power spectral density ratio,
`and W/R is the processing gain of the system. Let M denote the number
`mobile users. If power control is used to ensure that every mobile has the
`same received power at the base station, then (neglecting thermal noise) the
`interference power caused by the M — 1 interferers is
`
`;ain of the
`This fact
`is isolated
`:he carrier
`us to the Substituting for I, the carrier-to-interference ratio can now be expressed as
`n receiver
`
`j — C • (M — 1~
`
`(1.5~
`
`(1.3)
`
`1
`C
`C
`I C • (M — 1) M — 1
`
`(1.6)
`
`substituting C/I from equation (1.6~ into (1.4), the capacity for a CDMA
`system is found to be
`
`MGM -1=
`R Eb/No
`
`(1.7)
`
`Thus, the capacity of a CDMA system is proportional to the processing gain.
`The capacity of a CDMA system is limited by the interference caused by
`other users simultaneously occupying the same frequency band; this
`interference is reduced by the processing gain of the system. Because tech-
`niques that decrease the amount of interference received by the base station
`translate into equivalent gains in capacity, the CDMA system is capable of
`increasing capacity by virtue of its ability to reduce interference by the
`amount of the spread-spectrum processing gain. This processing gain is based
`on the fact that in the CDMA receiver the interfering users' signals continue
`to have the bandwidth W, while the selected user's signal is despread by the
`removal of the spreading code. The receiver then can perform filtering at the
`selected user's despread (baseband) bandwidth, which admits only a small
`fraction of the energy received from the interfering user signals.
`
`tecovered
`lata signal
`
`iformation
`
`Cisco Systems, Inc., Exhibit 1107, Page 14
`
`

`
`,\
`
`12
`
`CDMA Systems Engineering Handbook
`
`Studies have shown that the average duty cycle of afull-duplex voice
`conversation is approximately 35%. It is therefore possible in a digital system
`to reduce the transmission rate or to use intermittent transmissions during
`these pauses in speech. Because transmission is not eliminated entirely but
`reduced during such pauses, the effective duty cycle of the digital waveform is
`closer to 40% or 50%. If the duty cycle of the speech traffic channels in the
`CDMA system is denoted by the variable a, then the capacity equation
`becomes
`
`i
`
`1
`M N W 1 _ W 1
`aR E6/No R Eb/No a
`
`~1 g~
`
`The capacity equations shown thus far do not consider the directivity, if
`any, of the base station receiving antennas. If the base station employs
`directional antennas that divide the cell into sectors, each antenna will only
`receive a fraction of the interference from within the cell. In practice, the
`receiving antennas have overlapping coverage areas of approximately 15%.
`Standard implementations divide the cell into three sectors, which provides an
`effective capacity increase of G = 3 •0.85 = 2.55, and the corresponding
`equation for capacity is
`
`1 G
`M N W 1
`R E6/No a
`
`(1.9)
`
`Up to this point, the capacity equation assumes a single, isolated CDMA
`cell; the number M in (1.9) therefore is comparable with the number of
`frequencies available to users in a sector of a single-cell AMPS deployment. In
`a multicell system, the interference from signals originating in other cells
`must be taken into account when determining the capacity of a particular
`"home" cell; such interference is, of course, diminished by the attenuation
`incurred by the interferers in propagating to the home cell—for all multiple
`access schemes. In conventional cellular systems using FDMA, the base
`station receiver must have a C/I of at least 18 dB to ensure acceptable analog
`voice quality. Under typical propagation conditions, to achieve this C/I,
`adjacent cells cannot use the same frequency, and the same frequency can be
`used in only one out of seven cells. Thus, the capacity is of an FDMA
`multicell system is reduced by a factor of seven from that of a single-cell
`FDMA system.
`In a CDMA multicell system, the majority of interference emanates
`from mobiles within the same cell (while for FDMA there is no interference
`from mobiles in the same cell). Therefore, the total amount of interference
`
`Cisco Systems, Inc., Exhibit 1107, Page 15
`
`
`
`

`
`p ~
`~ r-
`
`_~-
`
`Introduction and Review of Systems Analysis Basics
`
`21
`
`where c is the speed of light. If OT > T~, where T~ is the PI~1 code ;hip
`duration or T~ = 1/Rcode where Rcode is the PN code sequence rate, then tl~~
`reflected signal is decorrelated (rejected) with the direct path signal, end tl-~i~ ~s
`the mechanism of multipath suppression by processing gain. Therefoa-e, fc~r
`sufficiently high PN code rates, the multipath signal can be effectively' su~-
`pressed and only the good resolvable multipaths are processed in the IS-95
`system. The base station uses a maximum of four multipaths (four "fingers"),
`whereas the mobile station processes a maximum of three fingers in the Rake
`receiver.
`
`x.1.6 Battle of Jamming Power Versus Processing Gain
`
`Spread-spectrum modulations have long been used by the military to combat
`intentional jamming by a hostile transmitter. As indicated in Table 1.1,
`spread-spectrum radio provides antijam (AJ) capabilities through a processing
`gain (PG) that results from using a wideband (large bandwidth) signal. As for
`commercial applications of spread-spectrum systems, the seemingly inefficient
`use of the radio spectrum was thought to be impractical in the past [17]. In
`commercial spread-spectrum systems, however, interference (jamming) comes
`from other similar users in the band, and these interferences, unlike hostile
`jammers, can be controlled, coordinated, and managed for the overall users'
`benefit in a CDMA digital cellular system. Though each commercial user has
`the same PN code, the coordination permits users to be distinguished by code
`phase in the application of the autocorrelation property of PN sequences that
`was described earlier. Not having the luxury of such coordination, each user
`in a military CDMA system generally has a distinct PN code generator to
`ensure a strong correlation with only one signal. Therefore, in addition to
`jamming, a military user is subject to MA inteference that is due to "cross-
`correlation" with different sequences, which is considerably larger than that
`of asingle-PN-generator based commercial system.
`The miltary advantage that the spread-spectrum system offers can also
`be illustrated in terms of a communication range extension capability over a
`conventional non-spread-spectrum communication system. Consider a situa-
`tion where a communicator, who requires a 10-dB SNR, employs a spread-
`spectrum modulation that provides a PG of 30 dB. Also assume that a hostile
`jammer at a 200-unit distance away uses jamming power equal to the
`communicator's transmitter power. Our assumptions are as follows:
`
`Cisco Systems, Inc., Exhibit 1107, Page 16
`
`

`
`pverview of the IS-95 Standard
`
`This book is designed to explain the principles underlying CDMA digital
`cellular systems, and this chapter, by providing an overview of the IS-95
`CDMA standard [1], functions as a "gateway" or point of departure for the
`contents in the rest of the book. The overview is intended for the reader to
`gain an overall familiarity with the system.
`The full title of the IS-95 standard is "Mobile Station-Base Station
`Compatibility Standard for Dual-Mode Wideband Spread Spectrum Cellular
`System," which indicates the fact that the document is a common air interface
`~~1~—it does not specify completely how a system is to be implemented,
`only the characteristics and limitations to be imposed on the signaling
`protocols and data structures. Different manufacturers may use different
`methods and hardware approaches, but they all must produce the waveforms
`and data sequences specified in IS-95.
`Just as little is said in IS-95 about the implementation of a particular
`requirement, the theory motivating and justifying the various requirements is
`not given in any detail. Because in many cases the significance of a particular
`requirement is not obvious, there is a need for the type of background
`information provided in this book. Once the principles behind the require-
`ments are understood, the operational significance of the requirements is
`often enhanced, and the implementation alternatives are apparent.
`The IS-95 standard refers to a "dual-mode" system, one that is capable of
`both analog and digital operation to ease the transition between current
`analog cellular systems and digital systems. Therefore, there are sections of
`IS-95 for both analog and digital cellular systems. Although attention is paid
`in Chapter 3 to analog cellular systems engineering issues, in general this
`book concentrates on the CDMA cellular system, and in this chapter only the
`sections of IS-95 dealing with the digital system are summarized.
`The IS-95 system, like all cellular systems, interfaces with the PSTN
`through an MTSO, as suggested in Figure 4.1. In that figure, the mobile
`stations are shown to communicate with base stations over "forward"
`
`333
`
`40
`angs
`
`50
`
`phone systems attempt to
`holding" or delaying calls
`J lines are all in use. The
`nce that a line will become
`~e of system is not that the
`call is delayed.
`
`Cisco Systems, Inc., Exhibit 1107, Page 17
`
`

`
`334
`
`CDMA Systems Engineering Handbook
`
`~ o.
`~~~
`
`~~!
`
`System
`PSTN ~~ controller and
`switch (MTSO)
`
`Base station
`
`Forward
`link (F)
`
`R
`
`-..`~
`
`s•'
`
`Base station
`
`Racy cta4inn
`
`Reverse link (R)
`
`.~,•
`
`s,.
`
`Figure 4.1 Cellular system architecture.
`
`(base-to-mobile) and "reverse" (mobile-to-base) radio links, also sometimes
`called "downlink" and "uplink," respectively.
`The radio communications over the forward and reverse links of the
`digital communications system that are specified by IS-95 are organized into
`"channels." Figure 4.2 illustrates the different channel types that are desig•
`nated in IS-95: pilot, synchronization, paging, and traffic channels for the
`forward link; and access and traffic channels for the reverse link.
`As will be shown, the modulation techniques and even the multiple
`access techniques are different on the forward and reverse links. Both links of
`the IS-95 system, however, depend on the conformity of all transmissions to
`strict frequency and timing requirements. Therefore, before proceeding to
`describe the forward and reve