United States Patent 55
`6,167,031
`[11] Patent Number:
`[45] Date of Patent:
`Dec.26, 2000
`Olofsson et al.
`
`US006167031A
`
`Matsuoka, H. et al., ‘Adaptive Modulation System with
`Variable Coding Rate Concatenated Code for High Quality
`Multi-Media Communication Systems’, Vehicular Technol-
`ogy Conference, Apr. 1996, IEEE 46th, pp. 487-491.
`
`Pearce, D.AJ. et al., ‘Comparison of Counter—-Measures
`Against Slow Rayleigh Fading for TDMA Systems’,
`Advanced TDMATechniques and Applications (Digest No.
`1996/234), IEE Colloquium, pp. 9/1-9/6, Oct. 1996.
`
`Najjoh, M. et al., ‘ARQ Schemes With Adaptive Modula-
`tion/TDMA/TDD Systems for Wireless Multimedia Com-
`munication Services’, Personal, Indoor and Mobile Radio
`Communications, Sep. 1997, PIMRC’97, 8th IEEE Interna-
`tional Symposium, pp. 709-713.
`
`Biglieri, E. et al., ‘Coding and Modulation Under Power
`Constraints’, IEEE Personal Communications, pp. 32-39,
`Jun. 1998.
`
`European Search Report dated May 12, 1998.
`
`J.E. Kleider et al., “An Adaptive-Rate Digital Communica-
`tion System for Speech”, 1997 IEEE International Confer-
`ence on Acoustics, Speech and Signal Processing, vol. 3,
`Apr. 21-24, 1997, Los Alamitos, CA.
`
`Primary Examiner—Melvin Marcelo
`Attorney, Agent, or Firm—Burns, Doane, Swecker &
`Mathis, L.L.P.
`
`[57]
`
`ABSTRACT
`
`[54] METHOD FOR SELECTING A
`COMBINATION OF MODULATION AND
`CHANNEL CODING SCHEMESIN A
`DIGITAL COMMUNICATION SYSTEM
`
`[75]
`
`Inventors: Hakan Gunnar Olofsson, Stockholm,
`Sweden; Jérn Thielecke, Erlangen,
`Germany
`
`[73] Assignee: Telefonaktiebolaget LM Ericsson
`(publ), Stockholm, Sweden
`
`[21] Appl. No.: 08/921,321
`
`[22]
`
`Filed:
`
`Aug. 29, 1997
`
`Tint, C07 occ ccccecseeeeeecccceessneeeeeccsessnneneeees H04Q 7/34
`[SD]
`[52] U.S. Ch oe 370/252; 370/345; 455/67.3
`[58] Field of Search oe 370/252, 345;
`455/63, 67.3, 226.3, 228, 522
`
`[56]
`
`References Cited
`U.S. PATENT DOCUMENTS
`
`6/1999 Frodigh et al. cee 375/302
`5,909,469
`FOREIGN PATENT DOCUMENTS
`
`OTHER PUBLICATIONS
`
`Sampei,S. et al, ‘Adaptive Modulation/TDMA Schemefor
`Personal Multi-Media Communication Systems’, Global
`Telecommunications Conference, 1994, GLOBECOM?’94,
`Communications: The Global Bridge, IEEE, pp. 989-993,
`Nov. 1994.
`
`0 137695 A2
`......... HO4N 1/00
`4/1985 European Pat. Off.
`0 154565 A2
`9/1995 European Pat. Off.
`........ HO4L 27/00
`0 731573 Al
`9/1996 European Pat. Off.
`........ HO4B 17/00
`
`WO 93/00751
`.. HO4B 1/10
`1/1993 WIPO...
`WO 97/13388—4/1997 WIPO. ooseeesssscsessssseseseseeees H04Q 7/38
`Acommunication system that supports multiple modulation
`and channel coding schemesselects an optimum RF link by
`measuring link quality parameters, such as C/I ratio. All of
`the available RF links are characterized based on the mea-
`sured link quality parameters by calculating mean values
`and variances of the parameters. Based on the characteriza-
`tion of the RF link, user quality values, such as user data
`throughput and speech quality values, are estimated. The
`communication system selects the RF link that provides the
`best user quality value.
`
`UE, T. et al., ‘Symbol Rate and Modulation Level Con-
`trolled Adaptive Modulation/TDMA/TDD for Personal
`Communication Systems’, Vehicular Technology Confer-
`ence, 1995 IEEE 45th, pp. 306-310, Jul. 1995.
`
`41 Claims, 10 Drawing Sheets
`
`IPR2019-00958
`Apple Inc. EX1009 Page 1
`
`
`
`MEASURE LINK
`QUALITY PABAMETERS
`
`STATISTICALLY
`DETERMINE MEAN
`AWD VARIANCE OF LINK
`QUALITY PARAMETERS
`
`REPORT CHABHEL
`CHARACTERISTIC
`
`MEASURES TO LINK
`QUBLITY ESTIMATOR
`
`
`
`
`
`PERFORM
`
`MAPPING FURCTION
`
`
`
`
`CALCULATE
`USER QUALITY
`
`SELECT OPTIMUM
`AE OE
`
`IPR2019-00958
`Apple Inc. EX1009 Page 1
`
`

`

`U.S. Patent
`
`Dec. 26, 2000
`
`Sheet 1 of 10
`
`6,167,031
`
`(119-9)Wey $¢O¢SfO50
`
`fDIA
`
`pA
`
`NVOH-0
`
`giropple
`SNHSd0-0--*62
`~>‘O!
`
`\
`
`
`
`
`
`‘INIDVAHOITIAVYLVTA}‘\
`
`
`
`INIAVITYTLM!LIFAUTd
`
`pO
`
`'£4\
`
`ol
`
`YY
`
`IPR2019-00958
`Apple Inc. EX1009 Page 2
`
`IPR2019-00958
`Apple Inc. EX1009 Page 2
`
`

`

`U.S. Patent
`
`Dec. 26, 2000
`
`Sheet 2 of 10
`
`6,167,031
`
`FIG. 2
`
`'
`
`PSTW
`
`WSC
`
`14
`
`WSC
`
`14
`
`0
`
`GATEWAY
`WSC
`
`|g
`
`{6
`
`16
`¢
`
`BSC
`
`one
`
`BSC
`
`
`
`IPR2019-00958
`Apple Inc. EX1009 Page 3
`
`IPR2019-00958
`Apple Inc. EX1009 Page 3
`
`

`

`U.S. Patent
`
`Dec. 26, 2000
`
`Sheet 3 of 10
`
`6,167,031
`
`27
`
`}——----Tam FRAME—__——|
`
`fea[le
`
`FIG. 3
`
`TRAINING
`TAIL
`DATA
`SEQUENCE
`GUARD
`DATA
`TAI|GUARD
`
`30
`
`3}
`
`32
`
`
`
`
`
`
`
`32
`
`31
`
`30
`
`33
`
`
`
`
`
`FIG. 4
`
`IPR2019-00958
`Apple Inc. EX1009 Page 4
`
`IPR2019-00958
`Apple Inc. EX1009 Page 4
`
`

`

`U.S. Patent
`
`Dec. 26, 2000
`
`Sheet 4 of 10
`
`6,167,031
`
`19
`
`6£
`
`09
`
`29
`
`
`
`GOLVINGOW
`
`“LL
`
`789000
`
`TTNNUKI
`
`YINVTTUTLM
`
`Y7009
`
`8S
`
`1S
`
`9S
`
`YOSSFIONd
`
`
`
`YOLVTHOONTO
`
`GDIA
`
`7YI00I70
`
`TINNUKI
`
`YIAVITYTLNI-T0
`700990
`
`bb
`
`HITS
`
`n>LhdLho
`
`6G
`
`+6£G
`
`IPR2019-00958
`Apple Inc. EX1009 Page 5
`
`IPR2019-00958
`Apple Inc. EX1009 Page 5
`
`
`
`
`
`
`
`
`

`

`U.S. Patent
`
`Dec. 26, 2000
`
`Sheet 5 of 10
`
`6,167,031
`
`beeeRLLeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee4
`
`'''t{ppdd!__&|!tII!i
`
`~-----------4----}---------4TeeTTa74'{III1SsSS!!SaSs!!ScF<;SSSS||.ji)'|'2'™'ti|1'''t1\1l
`
`FIG. 6
`
`“--~
`
`IPR2019-00958
`Apple Inc. EX1009 Page 6
`
`IPR2019-00958
`Apple Inc. EX1009 Page 6
`
`
`

`

`U.S. Patent
`
`Dec. 26, 2000
`
`Sheet 6 of 10
`
`6,167,031
`
`I901p0l80
`
`
`
`
`
`NG
`
`bl
`
`LAA
`
`18
`
`HULETANOI7WHOO:YITAWUS|
`
`1SSY
`
`Le
`
`ZOA
`
`YOLVTNGONI0
`
`YFITTOULNODXL
`
`
`
`TINNVHI
`
`ONVEISVE
`
`YOSSTIONA
`
`IPR2019-00958
`Apple Inc. EX1009 Page 7
`
`IPR2019-00958
`Apple Inc. EX1009 Page 7
`
`
`
`
`
`

`

`U.S. Patent
`
`Dec. 26, 2000
`
`Sheet 7 of 10
`
`6,167,031
`
`FIG. 8
`
`MEASURE LINK
`QUALITY PARAMETERS
`
`STATISTICALLY
`DETERMINE MEAN
`AND VARIANCE OF LINK
`QUALITY PARAMETERS
`
`AF LINK
`
`REPORT CHANNEL
`GHARACTERISTIC
`MEASURES TO LINK
`QUALITY ESTIMATOR
`
`PERFORM
`MAPPING FUNCTION
`
`CALCULATE
`USER QUALITY
`
`SELECT OPTIMUM
`
`IPR2019-00958
`Apple Inc. EX1009 Page 8
`
`IPR2019-00958
`Apple Inc. EX1009 Page 8
`
`

`

`U.S. Patent
`
`Dec. 26, 2000
`
`Sheet 8 of 10
`
`6,167,031
`
`YOLITIIS
`
`(4719-1)Y=I
`
`YIS/
`
`ALIT0N0
`
`YOLYWILST
`
`LLELA
`
`LLL
`
`A510FHL
`
`SIMSTETLIVYVHI
`
`YOLVWHS7
`
`TINNVH
`
`Bt
`
`git
`
`blk
`
`cH
`
`IPR2019-00958
`Apple Inc. EX1009 Page 9
`
`IPR2019-00958
`Apple Inc. EX1009 Page 9
`
`
`
`
`
`
`
`

`

`U.S. Patent
`
`Dec. 26, 2000
`
`Sheet 9 of 10
`
`6,167,031
`
`FIG. 10
`
`MEASURE
`MEAN G/T
`
`OPTIMUM COMBINATION
`
`CALCULATE NEW OPTIMAL POWER FOR
`EACH COMBINATION f
`Poot ti) ®t (CMe (:) ~ MEAN O/T)
`
`TRUWCATE Poyy7;) FOR EACH COMBINATION é
`
`fopt(i) ==minPmax? OX(fyjin, bovnyll
`
`CALCULATE CORRESPONDING MEAN C/I;
`FOR EACH COMBINATION i BASED ON hpf
`MEAN C/I, =MEAN C/T +(Bpt(i) 2)
`
`SELECT AN
`
`103
`
`105
`
`107
`
`SELECTED LINK
`
`USE Poy ON
`
`i
`
`IPR2019-00958
`Apple Inc. EX1009 Page 10
`
`IPR2019-00958
`Apple Inc. EX1009 Page 10
`
`

`

`U.S. Patent
`
`Dec. 26, 2000
`
`Sheet 10 of 10
`
`6,167,031
`
`FIG. [1
`
`100
`
`I0
`
`G0
`
`10
`
`60
`
`30
`
`40
`
`50
`
`20
`
`BLER(%)
`
`10
`
`HERyo=0
`
`od, Nyoss
`
`AyeSo?
`
`G/L (dB)
`
`IPR2019-00958
`Apple Inc. EX1009 Page 11
`
`IPR2019-00958
`Apple Inc. EX1009 Page 11
`
`

`

`6,167,031
`
`1
`METHOD FOR SELECTING A
`COMBINATION OF MODULATION AND
`CHANNEL CODING SCHEMESIN A
`DIGITAL COMMUNICATION SYSTEM
`
`BACKGROUND
`
`This invention generally relates to the field of communi-
`cation systems and, more particularly, to digital communi-
`cation systems that supports multiple modulation and chan-
`nel coding schemes.
`In wireless digital communication systems, standardized
`air interfaces specify most of system parameters, including
`modulation scheme, channel coding scheme, burst format,
`communication protocol, symbol rate, etc. For example,
`European Telecommunication Standard Institute (ETSI) has
`specified a Global System for Mobile Communication
`(GSM)standard that uses time division multiple access
`(TDMA) to communicate control, voice and data informa-
`tion over radio frequency (RF) physical channels or links
`using Gaussian Minimum Shift Keying (GMSK) modula-
`tion scheme at a symbol rate of 271 ksps. In the U.S.,
`Telecommunication Industry Association (TIA) has pub-
`lished a numberof Interim Standards, such as IS-54 and
`IS-136,
`that define various versions of digital advanced
`mobile phone service (D-AMPS), a TDMAsystem that uses
`a Differential QPSK (DQPSK) modulation scheme for com-
`municating data over RF links.
`Digital communication systems use a variety of linear and
`non-linear modulation schemes to communicate voice or
`data information in bursts. These modulation schemes
`include, GMSK, Quadrature Phase Shift Keying (QPSK),
`Quadrature Amplitude Modulation (QAM), etc. GMSK
`modulation scheme is a non-linear low level modulation
`(LLM) scheme with a symbolrate that supports a specified
`user bit rate. In order to increase user bit rate, high-level
`modulation (HLM) schemescan be used. Linear modulation
`schemes, such as QAM scheme, may havedifferent level of
`modulation. For example, 16QAM schemeis used to rep-
`resent the sixteen variations of 4 bits of data. On the other
`hand, a QPSK modulation scheme is used to represent the
`four variations of 2 bits of data.
`
`In addition to various modulation schemes, digital com-
`munication systems can support various channel coding
`schemes, which are used to increase communication reli-
`ability. For example, General Packet Radio Service (GPRS),
`which is a GSM extension for providing packet data service,
`supports four channel coding schemes. A Convolutional
`Half-Rate Code scheme, CS1 coding scheme, whichis the
`“mother” channel coding scheme of GPRS. The CS1 scheme
`is punctured to obtain approximately two-third rate and
`three-fourth rate code schemes, CS2 and CS3 coding
`schemes. GPRSalso supports an uncoded scheme, knownas
`CS4 coding scheme.
`Generally, channel coding schemes code and interleave
`data bits of a burst or a sequence of bursts to prevent their
`loss under degraded RF link conditions, for example, when
`RF links are exposed to fading. The numberof coding bits
`used for channel coding of data bits corresponds to error
`detection accuracy, with higher number of coding bits pro-
`viding higherbit error detection accuracy. For a given gross
`bit rate, a high numberof coding bits, however, reduces user
`bit rate, since coding bits reduce the numberofuser data bits
`that can be transmitted in a burst.
`
`The communication channeltypically introduceserrors in
`sequence. In order to improve coding efficiency, the coded
`bits are interleaved, before transmission. The purpose of
`
`10
`
`15
`
`30
`
`35
`
`45
`
`50
`
`55
`
`60
`
`65
`
`2
`interleaving is to distribute the errors over several code
`words. The term perfect
`interleaving is used when the
`sequenceof the received data bit errors are uncorrelated. The
`more less uncorrelated the received data bits are at the
`receiver, the easierit is to recoverlost data bits. On the other
`hand,if interleavingis not effective, large portions or blocks
`of transmitted data bits may be lost under degraded RF link
`conditions. Consequently, error correction algorithms may
`not be able to recover the lost data.
`
`TDMAsystems subdivide the available frequency band
`into one or several RF channels. The RF channels are
`divided into a numberof physical channels corresponding to
`time slots in TDMAframes. Logical channels are mapped
`onto one or more physical channels, where modulation and
`channel coding schemes are specified. An RF link includes
`one or more physical channels that support
`the logical
`channels. In these systems, the mobile stations communicate
`with a plurality of scattered base stations by transmitting and
`receiving bursts of digital
`information over uplink and
`downlink RF channels.
`
`The growing number of mobile stations in use today has
`generated the need for more voice and data channels within
`cellular telecommunication systems. As a result, base sta-
`tions have become more closely spaced, with an increase in
`interference between mobile stations operating on the same
`frequency in neighboring or closely spaced cells. Although
`digital techniques gain more useful channels from a given
`frequency spectrum, there still remains a need to reduce
`interference, or more specifically to increase the ratio of the
`carrier signal strength to interference, (ie., carrier-to-
`interference (C/1)) ratio. RF links that can handle lower C/I
`ratios are considered to be more robust than those that only
`can handle higher C/I ratios.
`Depending on the modulation and channel coding
`schemes, grade of service deteriorates more rapidly as link
`quality decreases. In other words, the data throughput or
`grade of service of more robust RF links deteriorates less
`rapidly than those of less robust RF links. Higher level
`modulation schemes are more susceptible to link quality
`degradation than lower level modulation schemes. If a HLM
`schemeis used, the data throughput drops very rapidly with
`a drop in link quality. On the other hand, if a LLM scheme
`is used, data throughput and grade of service does not
`deteriorate as rapidly under the same interference condi-
`tions.
`
`Therefore, link adaptation methods, which provide the
`ability to dynamically change modulation scheme, channel
`coding, and/or the number of used time slots, based on
`channel conditions, are used to balance the user bit rate
`against link quality. Generally, these methods dynamically
`adapt
`a system’s combination of channel coding,
`modulation, and numberof assignable time slots to achieve
`optimum performance overa broad range of C/I conditions.
`One evolutionary path for next generation of cellular
`systems is to use high-level modulation (HLM), e.g.,
`16QAM modulation scheme, to provide increased user bit
`rates compared to the existing standards. These cellular
`systems include enhanced GSM systems with GPRS
`extension, enhanced D-AMPS systems,
`International
`Mobile Telecommunication 2000 (IMT-2000), etc. A high
`level
`linear modulation, such as 16QAM modulation
`scheme,has the potential to be more spectrum efficient than,
`for example, GMSK, which is a low-level modulation
`(LLM) scheme. Because higher level modulation schemes
`require a higher minimum C/I
`ratio for acceptable
`performance, their availability in the system becomes lim-
`
`IPR2019-00958
`Apple Inc. EX1009 Page 12
`
`IPR2019-00958
`Apple Inc. EX1009 Page 12
`
`

`

`6,167,031
`
`3
`ited to certain coverage areas of the system or certain parts
`of the cells, where more robust links can be maintained.
`In order to provide various communication services, a
`corresponding minimum userbit rate is required. In voice
`and/or data services, user bit rate corresponds to voice
`quality and/or data throughput, with a higher user bit rate
`producing better voice quality and/or higher data through-
`put. The total user bit rate is determined by a selected
`combination of techniques for speech coding, channel
`coding, modulation scheme, and for a TDMAsystem, the
`number of assignable timeslots per call.
`Data services include transparent services and non-
`transparent services. Transparent services, which have a
`minimum link quality requirement, provide target user bit
`rates. A system that provides transparent communication
`services varies the gross bit rate to maintain a constant user
`bit
`rate with the required quality. Conversely,
`in non-
`transparent services, for example, GPRS, the user bit rate
`may vary, because erroneously received data bits are retrans-
`mitted. Unlike non-transparentservices, transparent services
`do not retransmit erroneously received data bits. Therefore,
`transparent services have a constant point-to-point transmis-
`sion delay, and non-transparent services have a non-constant
`point-to-point transmission delay.
`A communication system may provide a data service
`through a numberof RF links supporting different combi-
`nations of channel coding, speech coding, and/or modulation
`schemes. For example, the system may provide a multime-
`dia service using two or more separate RF links that sepa-
`rately provide audio and video signals. Underthis scenario,
`one of the two RF links may use HLM schemeandthe other
`link may use LLM scheme. In order to provide a constant
`user bit rate in a TDMAsystem, lower level modulation
`schemes may use a higher numberof time slots than higher
`level modulation schemes.
`
`Moreover, digital communication systems must also
`select a suitable combination of channel coding and modu-
`lation schemesbased on link quality. For example, for a high
`quality link, higher level modulation or less channel coding
`results in higher user bit rate, which may be used advanta-
`geously by different communication services. For example,
`in a non-transparent data service, user data throughput is
`increased. For a speech service, the increased user bit rate
`may be used for deploying an alternative speech coder with
`higher quality. Therefore, a system that supports multiple
`modulation and channel coding schemes should provide
`sufficient flexibility for selecting an optimum combination
`of modulation and channel coding schemes.
`Conventional method for selecting an optimum combina-
`tion of modulation and channel coding schemes assumethat
`the link quality parameters are perfectly known at a given
`instant. Usually,
`these methods determine link quality
`parameters by measuring, at predefined instances, one or
`more of received signal strength (RSS) or bit error rate
`(BER), etc. Using these instantaneous measurements, these
`methods also assume that user quality as a function of link
`quality parameters is perfectly known for all combinations
`of modulation and channel coding schemes.
`the mean
`Because these parameters vary continuously,
`measurement of link quality parameters do not give an
`accurate indication of user quality, especially after a link
`with a different combination of modulation and channel
`
`coding schemesis selected. One method dynamically adapts
`user bit rate of a TDMAsystem to achieve optimum voice
`quality over a broad range of channel conditions. This
`system continuously monitors link quality by making instan-
`
`10
`
`15
`
`20
`
`25
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`4
`taneous measurements of a RF link’s C/I ratio. The system
`dynamically adapts its combination of modulation and chan-
`nel coding schemes and the numberof assignable time slots
`to optimize voice quality for the measured conditions. In
`addition, the system determines cost functions to derive at a
`cost of using RF links with different modulation and coding
`schemes to improve voice quality.
`User quality, however, varies considerably with variations
`in link quality parameters. FIG. 1 showslink performance of
`two modulation schemes, i.e., QPSK and 16QAM schemes,
`which are exposed to three channel conditions: an Average
`White Gaussian Noise (AWGN) channel condition, a fast
`Rayleigh Fading channel condition, and a slow Rayleigh
`fading channel condition. In FIG. 1, link performance is
`expressed in terms of BER. For a given C/I ratio, the AWGN
`channel provides the best performance, due to the lack of
`fading dips. In fast Rayleigh fading channel, where fading
`varies fast enough to makeeffective use of interleaving, link
`performance is degraded compared to the AWGN channel.
`In slow Rayleigh fading channel, where fading varies slowly
`such that interleavingis not effective, the worst link perfor-
`manceis obtained. Conventional methods use mean C/I ratio
`to determine the channel condition. As shown in FIG. 1,
`however, mean C/I ratio for different channel conditions
`may be the same, when link performance under different
`combination of modulation and channel coding schemes
`may be quiet different. Therefore, more information is
`needed to accurately estimate link performance, if different
`combinations of modulation and channel coding is used.
`An additional factor affecting user quality is time disper-
`sion. Receiver equalizers can not effectively handle large
`time dispersions. As a result, link performance degrades,
`even when C/I
`ratio distribution remains the same.
`Accordingly, mean measurements of C/I ratio, BER or time
`dispersion alone are not sufficient for estimating perfor-
`mance of a selected link. Therefore, there exists a need for
`an effective link selection method in systems that support
`various modulation and channel coding schemes.
`
`SUMMARY
`
`The present invention that addresses this need is exem-
`plified in a selection method that statistically characterizes
`combinations of available modulation and channel coding
`schemes using measured link quality parameters to deter-
`mine which combination provides the best user quality. The
`method of the invention measures at least one link quality
`parameter of at least one RF link, for example, C/I ratio,
`BER,received signal strength, or time dispersion. Then,at
`least one channel characteristic measure is calculated based
`on the measured link quality parameter by computing both
`its mean value and variance. By introducing the variance of
`for example C/I ratio, it is possible to estimate the type of
`channel conditions a transmission is susceptible to.
`Consequently, it is possible to estimate how a change of
`modulation and/or channel coding scheme would effect the
`link quality.
`In an exemplary embodiment,
`the channel
`characteristic measure may be calculated for each one of
`available combinations of modulation and channel coding
`schemes of an RF link. Thereafter, a user quality estimator
`estimates user quality values,
`for example, user data
`throughput or speech quality values, based on the calculated
`channel characteristic measure. Finally, the present inven-
`tion selects a combination of modulation and channel coding
`schemes on an RFlink that provides the best user quality.
`According to some of its more detailed features,
`the
`present invention mapsthe calculated channel characteristic
`
`IPR2019-00958
`Apple Inc. EX1009 Page 13
`
`IPR2019-00958
`Apple Inc. EX1009 Page 13
`
`

`

`6,167,031
`
`5
`measure with estimated user quality values of the supported
`combinations of modulation and channel coding schemes.
`The mapping function may use simulation results, labora-
`tory results, or results derived during normal operation of a
`communication system.
`According to another aspect of the invention, the selection
`method determines an optimal
`transmit power for each
`combination of modulation and channel coding schemes
`based on the measured link quality parameter. Thereafter,
`the user quality values are estimated based on the optimal
`transmit power. Also, data bursts are transmitted on the
`selected RF link at the optimal transmit power.
`Other features and advantages of the present invention
`will become apparent from the following description of the
`preferred embodiment, taken in conjunction with the accom-
`panying drawings, whichillustrate, by way of example, the
`principles of the invention.
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`FIG. 1 is a diagram of the performance of two variously
`modulated RFlinks underthree different channel conditions.
`
`FIG. 2 is a block diagram of a communication system
`which advantageously uses the present invention.
`FIG. 3 is a diagram of a subdivided RF channelthat is
`used in the communication system of FIG. 2.
`FIG. 4 is a diagram of a normal
`transmission burst
`transmitted on the RF channel of FIG. 3.
`
`FIG. 5 is a block diagram of a mobile unit used in the
`communication system of FIG. 2.
`FIG. 6 is a block diagram of a radio base station used in
`the communication system of FIG. 2.
`FIG. 7 is a block diagram ofa radio transceiver used in the
`base station of FIG. 6.
`
`FIG. 8 is a flow chart of a link selection method according
`to an exemplary embodiment of the invention.
`FIG. 9. is a block diagram ofthe selection method of FIG.
`
`8.
`
`FIG. 10 is a flow chart of a power selection scheme
`according to another aspect of the invention.
`FIG. 11 is a graph of link performances of two combina-
`tions of channel coding and modulation schemes.
`DETAILED DESCRIPTION
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`Referring to FIG. 2, a communication system 10 accord-
`ing to an exemplary embodiment of the present invention
`supports multiple modulation schemes. In an exemplary
`embodiment of the invention, the system 10 supports three
`modulation schemes: a first LLM (LLM1) scheme, a second
`LLM (LLM2) scheme, and a HLM scheme. LLM1 scheme
`is a non-linear modulation scheme, such as GMSK modu-
`lation scheme used in GSM systems. LLM2 schemeis a
`linear modulation scheme, such as QPSK. Finally, HLM
`scheme is a higher level linear modulation schemes, for
`example, 16QAM scheme, that could be supported by the
`second generation of enhanced GSM systems, which as of
`yet are not standardized.
`The communication system 10 also supports the channel
`coding schemes of GSM’s GPRSextension. The system 10,
`For example, for providing a non-transparent service, the
`therefore, supports CS1, CS2, CS3, and CS4 channel coding
`schemes. The system 10 supports various combinations of
`system 10 estimates user quality values of available com-
`
`modulation and channel coding schemesonaplurality of RF binations of modulation and channel coding schemeson the
`65
`links. Although, the system 10 is described with reference to
`one or more RF links in terms of data throughput S. For a
`the above specified exemplary modulation and channel
`predefined time period, the system 10 continuously mea-
`coding schemes, it should be noted that a wide range of
`sures link quality parameters and calculates their mean
`
`6
`modulation and coding schemes may be used to implement
`the present invention.
`The mode of operation of GSM communication systems
`is described in European Telecommunication Standard Insti-
`tute (ETSI) documents ETS 300 573, ETS 300 574 and ETS
`300 578, which are hereby incorporated by reference.
`Therefore, the operation of the GSM system is described to
`the extent necessary for understanding of the present inven-
`tion. Although, the present invention is described as embod-
`ied in a GSM system, those skilled in the art would appre-
`ciate that the present invention could be used in a wide
`variety of other digital communication systems, such as
`those based on PDC or D-AMPSstandards and enhance-
`
`ments thereof. The present invention may also be used in
`CDMAora hybrid of CDMA and TDMA communication
`systems.
`The communication system 10 covers a geographical area
`that is subdivided into communication cells, which together
`provide communication coverage to a service area,
`for
`example, an entire city. Preferably, the communication cells
`are patterned accordingto a cell pattern that allows some of
`the spaced apart cells to use the same uplink and downlink
`RF channels. In this way, the cell pattern of the system 10
`reduces the number of RF channels needed to cover the
`service area. The system 10 may also employ frequency
`hopping techniques, for example, to avoid “deadspots.”
`Initial selection of modulation scheme would preferably
`depend on either measured or predicted link quality param-
`eters of a new RF link. Alternatively, the initial selection
`may be based on a predefined cell parameter. Due to a
`possible difference in link robustness for LLM1, LLM2, and
`HLM schemes, a mobile station 12 continues to use LLM1
`scheme until the channel characteristic allows the use of
`
`other schemes, in which case a link adaptation procedure is
`initiated to switch modulation scheme from LLM1 scheme
`to LLM2, or HLM scheme.
`When noinformation is transferred to or from a mobile
`station 12, for example, during idle states or wait states of
`GPRS, the mobile station 12 preferably measures link qual-
`ity parameters of different RF links. For instance, the mobile
`station 12 measures the interference on RF links that are
`candidates for use in the future as well as the received signal
`strength ofits current link. The measurementresults are used
`to determine a distribution of channel characteristic mea-
`sures. These measurements serve as the basis for deciding
`which combination of modulation and channel coding
`schemes to use subsequently.
`According to the present invention, during an ongoing
`communication, user quality values are estimated based on
`channel characteristics, which are expressed in terms of
`variations and mean values of link quality parameters. The
`channel characteristics are derived based on measurements
`
`of link quality parameters over a predefined period.In this
`way, the system 10 estimates user quality values provided by
`available combinations of modulation and channel coding
`schemes of one or more RF links. By comparing the
`estimated user quality values of these combinations, the
`present invention selects a modulation and channel coding
`combination on an RFlinkthat providesthe best user quality
`value.
`
`IPR2019-00958
`Apple Inc. EX1009 Page 14
`
`IPR2019-00958
`Apple Inc. EX1009 Page 14
`
`

`

`6,167,031
`
`7
`values and variances. The present inventionrelies on statis-
`tical measures to characterize an RF link. Although the
`exemplary embodiment uses mean values and variances,
`other statistical measures may also be used, for example,
`standard deviation, median, etc. The system 10 calculates
`the mean values of such link quality parameters as C/I ratio
`or BER values that are obtained over the predefined time
`period. Based on measured link quality parameters over the
`predefined time period, the system 10 also determines the
`variances of one or more of the link quality parameters.
`Based on the variances, the system 10 estimates the data
`throughputs S for all combinations of modulation and chan-
`nel coding schemes over one or more RF links. The system
`then selects a new combination of modulation and channel
`
`if switching to the new
`coding schemes on a RF link,
`combination on that RF link providesa higher data through-
`put S than that provided by a current combination.
`For a speech service, the system 10 may use a different
`user quality value measure than the data throughput S used
`for a non-transparent data service. Preferably,
`the user
`quality value in speech service is expressed in terms of a
`voice quality value Q, which may be based on estimated
`frame erasure rate (FER) and/orresidual user bit error rate
`(RBER) originated from the use of various speech coding
`schemes. Under this arrangement,
`the present
`invention
`estimates voice quality values Q for different combinations
`of modulation and channel coding schemes. Then, the sys-
`tem 10 selects a combination that provides the best esti-
`mated voice quality value.
`The system 10 is designed as a hierarchical network with
`multiple levels for managing calls. Using an allocated set of
`uplink and downlink RF links a number of mobile stations
`12 operating within the system 10 participate in calls using
`allocated time slots. At a high hierarchical level, a group of
`Mobile Service Switching Centers (MSCs) 14 are respon-
`sible for the routing of calls from an originator to a desti-
`nation. In particular, they are responsible for setup, control
`and termination of calls. One of the MSCs 14, knownas the
`gateway MSC, handles communication with a Public
`Switched Telephone Network (PSTN) 18, or other public
`and private networks.
`Different operators support different communication stan-
`dards with different modulation and channel coding
`schemes. The same operator may also support different
`modulation and channel coding schemes in different cells.
`For example, one operator may support LLM1 modulation
`scheme and CS4 channel coding scheme only, whereas,
`another operator may support all of the modulation and
`channel coding schemes. The communication system 10
`uses the present invention to select a combination of modu-
`lation and channel coding schemesthat provide the best user
`quality value.
`Ata lowerhierarchical level, each one of the MSCs 14 are
`connected to a group of base station controllers (BSCs) 16.
`The primary function of a BSC 16 is radio resource man-
`agement. For example, based on reported received signal
`strength at the mobile stations 12, the BSC 16 determines
`whetherto initiate a hand over. Under the GSM standard, the
`BSC 16 communicates with a MSC 14 under a standard
`interface known as the A-interface, which is based on the
`Mobile Application Part of CCITT Signaling System No. 7.
`Ata still lower hierarchical level each one of the BSCs 16
`controls, a group of base transceiver stations (BTSs) 20.
`Each BTS20 includes a number of TRXsthat use the uplink
`and downlink RF channels to serve a particular common
`geographical area. The BTSs 20 primarily provide the RF
`
`8
`links for the transmission and reception of data bursts to and
`from the mobile stations 12 within their designated cell. In
`an exemplary embodiment, a number of BTSs 20 are incor-
`porated into a radio base station (RES) 22. The RBS 22 may
`be configured according to a family of RBS-2000 products,
`which is offered by Ericsson, the assignee of the present
`invention.
`
`With reference to FIG. 3, an RF channel 26 (uplink or
`downlink) is divided into repetitive time frames 27 during
`which information are communicated. Each frame 27 is
`
`10
`
`further divided into time slots 28 that carry packets of
`information. Speech or data is transmitted during timeslots
`designated as traffic channels (TCH,,
`.
`.
`.
`, TCH,). All
`signaling functions pertaining to call management in the
`system, including initiations, hand overs, and terminati