Phoenix, AZ
`February 12, 2001
`
`3GPP2-C50-20010212- 011
`
`Title: Per-User Reverse Rate Control for Shared Packet Data Channel in 1xEV-DV
`
`Abstract: A per-user forward rate-control channel (F-RCCH) is proposed to allow a base
`station to control the reverse link data rate of each active Mobile Station (MS) sharing a
`packet data channel. By assigning data rates to active MS’s based on their forward link data
`rate controls (DRC’s), overall reverse link sector throughput can be improved, potentially
`very significantly. Basic simulation results are provided to illustrate these potential gains.
`
`Source:
`
`Sae-Young Chung, Dae-Young Kim, Vedat Eyuboglu
`Airvana, Inc.
`
`Contact:
`
`Dae-Young Kim
`dykim@airvananet.com, 978-250-2623
`
`Date:
`
`Feb. 12, 2001
`
`Recommendation: Discuss and Adopt in WG-5
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`Notice
`
`©2001 Airvana, Inc. All rights reserved.
`The information contained in this contribution is provided for the sole purpose of promoting discussion within the 3GPP2
`and its Organization Partners and is not binding on the contributor. The contributor reserves the right to add to, amend, or
`withdraw the statements contained herein.
`The contributor grants a free, irrevocable license to 3GPP2 and its Organization Partners to incorporate text or other
`copyrightable material contained in the contribution and any modifications thereof in the creation of TIA or 3GPP2
`publications; to copyright and sell in Organizational Partner’s name any Organizational Partner’s standards publication even
`though it may include portions of the contribution; and at the Organization Partner’s sole discretion to permit others to
`reproduce in whole or in part such contributions or the resulting Organizational Partner’s standards publication.
`
`1
`
`Ex. 1004 - Sierra Wireless, Inc.
`Sierra Wireless, Inc., et al. v. Sisvel S.P.A., IPR2021-01141
`Page 1 of 9
`
`

`

`Phoenix, AZ
`February 12, 2001
`
`
`
`3GPP2-C50-20010212-xxx011
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`29
`
`30
`
`31
`
`32
`
`33
`
`34
`
`35
`
`36
`
`37
`
`38
`
`39
`
`1 INTRODUCTION
`
`1xEV-DV requirements [1] have set very challenging goals for reverse link packet data
`performance:
`
`
` Peak data rate of at least 1.25 Mbit/s in an outdoor, vehicular environment;
` Peak data rate of 2 Mbit/s in a stationary, indoor environment;
` System-wide average data rates in a fully loaded system of at least 600 kbit/s in
`an outdoor, vehicular environment.
`
`
`However much of the focus of 1xEV-DV framework proposals to date have been in
`increasing forward link throughput by adopting many of the concepts previously developed
`for 1xEV-DO. Specifically, these proposals take advantage of the fact that in packet data
`applications it is not necessary to provide any rate guarantees to individual users. By instead
`serving these users at variable rates based on their channel characteristics (described by a
`parameter fSIR, defined later), forward link sector throughput (sum of rates) can be increased
`significantly.
`
`Even though reverse link is operationally quite different from the forward link, it is possible
`to similarly increase reverse link sector throughput by supporting variable rate packet data
`transmission and adjusting user’s data rates based on their channel characteristics (now
`described by a parameter rSIR, also defined later). Such an approach also better aligns the
`forward and reverse link rates of a packet data user.
`
`To effectively support such variable-rate reverse link operation, a per-user rate control
`mechanism needs to be included in 1xEV-DV. In this contribution, we describe how the
`reverse link sector throughput can be improved using variable-rate transmission with per-user
`rate control and discuss alternative ways of adding a per-user rate control channel to the
`forward link without introducing significant overhead.
`
`
`2 INCREASING REVERSE LINK SECTOR THROUGHPUT
`
`
`Consider a set of N active Mobile Stations (MS’s) who are sharing a packet data channel
`within in a sector. Let the reverse link transmission rate of the i’th user be Ri, i = 1, 2,…N,
`where Ri is chosen from a finite set. As an example, in 1xEV-DO Ri's are chosen from the
`set {9.6, 19.2, 38.4, 76.8, 153.6 kbit/s}. We For simplicity in the analysis that follows, we
`assume that active MS’s always have data to transmit.
`
`
`
`2
`
`Ex. 1004 - Sierra Wireless, Inc.
`Sierra Wireless, Inc., et al. v. Sisvel S.P.A., IPR2021-01141
`Page 2 of 9
`
`

`

`Phoenix, AZ
`February 12, 2001
`
`
`
`3GPP2-C50-20010212- 011
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`29
`
`30
`
`31
`
`32
`
`33
`
`34
`
`35
`
`36
`
`37
`
`38
`
`39
`
`40
`
`41
`
`42
`
`43
`
`
`A. INCREASING THROUGHPUT IN THE SINGLE-SECTOR CASE
`
`First, consider the case of a single-sector operation, where system performance is examined
`in only one sector, paying no attention toessentially ignoring interference from/to other
`sectors. The Base Station (BS) is power controlling the MS’s in such a way that their signals
`arrive at the receiver with just enough power Si to achieve a certain Eb/I0. Generalizing the
`results given in [2] to the case of variable-rate operation considereded here, it can shown that
`the data rate Ri of the i’th user is approximately proportional to
`
`
` Ri  Si /(Sintra(i) + Noise),
`
`where Sintra (i) = ji Sj is the total signal power received from all other MS’s in the sector,
`and “Noise” represents the total receiver noise power including the effects of receiver noise
`figure. Next, we write the signal power Si received from the i’th MS as
`
`
`Si = Pi Ai,
`
`where Pi is the transmit power of the i’th MS and Ai is the squared reverse link channel gain
`between the i’th MS and the serving sector.
`
`Suppose that initially all MS’s are transmitting at the same data rate Ri = R0 producing the
`same received power Si = S0 at the BS receiver. Now, suppose we modify the rate allocation
`such that the data rate Ri of the i’th user is set to be proportional to its channel gain Ai. At
`the same time, we modify the transmit powers Pi to achieve the desired Eb/I0, while we
`increase the transmit power Pi of the MS with the highest value of G0(i) and at the same time
`reduce the transmit power of the MS with the smallest value of G0(i) by the same amount,
`keeping the total transmit power [i.e., i Pi ] fixed. It can be shown that as long as the
`channel gains Ai are not all identical and the system is fully loaded (such that the “Noise”
`term can be ignored), this new power and rate allocation will always increase the sector
`throughput. if the “Noise” term can be ignored, which would be the case when the sector
`becomes fully loaded..
`
`This example illustrates that by allocating data rates based on the individual reverse link
`channel characteristics (represented by Ai), reverse link sector throughput can be increased.
`Even higher sector throughputs can be achieved realized if we were to allow only 1 MS with
`the best channel condition to transmit, essentially thus operating with no interference from
`other MS’s in the same sector. In fact it is known [3] that the optimum throughput-
`maximizing strategy on a multi-user fading channel is to use a TDMA system approach
`where only the user with the best channel condition gets to transmit. Of course, such an
`approach can be unfair to certain users who remain in poor channel conditions for a long
`time. Therefore, in normal real operation, one would seek to maximize the sector throughput
`while satisfying a certain fairness criterion.1
`
`
`1 This is similar to fairness issues found on the forward link.
`
`
`
`3
`
`Ex. 1004 - Sierra Wireless, Inc.
`Sierra Wireless, Inc., et al. v. Sisvel S.P.A., IPR2021-01141
`Page 3 of 9
`
`

`

`Phoenix, AZ
`February 12, 2001
`
`
`
`3GPP2-C50-20010212-xxx011
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`29
`
`30
`
`31
`
`32
`
`33
`
`34
`
`35
`
`36
`
`37
`
`38
`
`39
`
`40
`
`41
`
`42
`
`43
`
`44
`
`45
`
`46
`
`
`
`
`
`B. REDUCING INTERFERENCE TO OTHER SECTORS
`
`In the previous subsection, we saw how that if we ignore inter-sector interference we can
`increase the sector throughput by letting MS’s with good channel characteristics (large Ai)
`transmit at a higher data rate relative to MS’s with worse channel characteristics. In multi-
`sector operation, however, we are also concernedIn this subsection, we'll consider inter-
`sector about interference created to other sectors. Next, we and illustrate how rate allocation
`on the reverse link can also reduce such interference.
`
`First, we write the total interference from the i’th MS to all other sectors as PiBi, where Bi is
`the sum of the squared reverse link channel gains between the i’th MS and all other
`neighboring sectors. Here, Pi is again the transmit power of the i’th MS.
`
`As in the previous section, suppose that initially all MS’s are transmitting at the same bit data
`rate Ri = R0 producing the same received power Si = S0. Now, suppose we increase the
`transmit power Pi of an MS with a small value of Bi and at the same time reduce by the same
`amount the transmit power of the an MS with a larger value of Bi by the same amount,
`keeping the total transmit power [i.e., i Pi ] fixed. It is straightforward to see that this simple
`modification will always reduce the total interference created to other sectors.
`
`This example illustrates that by allocating data rates to MS’s based on their individual
`interference characteristics (represented by Bi(i)), ), total interference to other sectors can be
`significantly reduced. As in the previous example, interference can be reduced, possibly
`dramatically, if we allow only 1 MS with the best interference condition to transmit. Again,
`in normal operation one would seek to minimize interference while satisfying a certain
`fairness criterion.
`
`C. INCREASING THROUGHPUT IN MULTI-SECTOR OPERATION
`
`In multi-sector operation we try to increase sector throughput while taking into account both
`intraintra-sector and inter-sector interference. In this case, the data rate Ri of an MS is
`approximately proportional to
`
`
` Ri  Si /(Sintra(i)+ Sinter + Noise),
`
`where Sinter is represents the interference from adjacent sectors.
`
`Combining the earlier resultsBased on the results of the previous two subsections, one can
`show that the reverse link sector throughput can be increasedwe now introduce the following
`if we allocate the data rates based on the following reverse link SIR parameter:
`
`
`rSIR(i) = Ai/Bi.
`
`4
`
`Ex. 1004 - Sierra Wireless, Inc.
`Sierra Wireless, Inc., et al. v. Sisvel S.P.A., IPR2021-01141
`Page 4 of 9
`
`

`

`Phoenix, AZ
`February 12, 2001
`
`
`
`3GPP2-C50-20010212- 011
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`29
`
`30
`
`31
`
`32
`
`33
`
`34
`
`35
`
`36
`
`37
`
`38
`
`39
`
`40
`
`41
`
`42
`
`43
`
`
`This captures the two components of the channel which we used for rate allocationallocation
`discussed earlier: Ai and Bi. By assigning higher data rates to users based on the channel
`parameter rSIR, one can increase the sector throughputwe would take into account the effect
`of Ai in increasing single-sector throughput gain as well as the effect of Bi in reducing inter-
`sector interference by controlling both intra-sector and inter-sector interference.
`
`Of course, the values of rSIR(i) are generally unknown at the serving sector, particularly the
`component Bi, which would have to be measured at other BS's and then somehow
`communicated to the serving sector. Luckily, various strategies can be employed to estimate
`approximate rSIR(i) from a knowledge of fSIR(i) which is the forward link SIR seen by the
`i’th MS. Specifically, we define fSIR(i) according to
`
`
`fSIR(i) = Ci/Di (i),
`
`where Ci is the squared forward link channel gain between the i’th MS and the serving sector
`and Di(i) is the sum of the squared forward link channel gains between the i’th MS and all
`other neighboring sectors.
`
`It should be noted that most 1xEV-DV framework proposals include a data rate request
`capability, similar to the Data Rate Control (DRC) feature found in 1xEV-DO, where the MS
`reports to the sector its achievable data rate based on its measurement of fSIR(i), or an
`approximation to it. As a result, the BS already has knowledge of fSIR(i). Now, if the
`channel gains in the forward and reverse directions were identical (i.e., Ai = Ci, Bi (i) = Di
`(i)), we would have rSIR(i) = fSIR(i), and the serving sector would also know the values of
`rSIR(i).
`
`In reality, both rSIR(i) and fSIR(i) are random quantities that typically depend on path loss,
`shadow fading and Raleigh fading. Path loss and shadow fading tend to be highly correlated
`in the two directions of transmission, but Raleigh fading is almost completely uncorrelated.
`As a result, rSIR(i) and fSIR(i) are rarely the same. However, if we average rSIR(i) and
`fSIR(i) over a sufficiently long period, they will become strongly correlated. Therefore, we
`believe using an appropriatelyan averaged version of the DRC values received from the MS,
`the serving sector can allocate reverse link rates essentially in proportion to the forward link
`throughput allocated given to each MS. Simulation results provided in Section 4 show that
`very significant gains can be achieved using this approach.
`
`
`3 POSSIBLE WAYS OF ADDING A PER-USER RATE CONTROL CHANNEL
`
`1xEV-DO uses RateLimit messages and RAB (Reverse Activity Bit) to control the reverse
`link rate. However, RAB is a probabilistic global control mechanism that cannot be used for
`per-user rate control. Alternatively, the BS can use RateLimit messages to control the
`maximum allowed reverse link rate of each MS. However, these messages carry significant
`
`5
`
`Ex. 1004 - Sierra Wireless, Inc.
`Sierra Wireless, Inc., et al. v. Sisvel S.P.A., IPR2021-01141
`Page 5 of 9
`
`

`

`Phoenix, AZ
`February 12, 2001
`
`
`
`3GPP2-C50-20010212-xxx011
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`29
`
`30
`
`31
`
`32
`
`33
`
`34
`
`35
`
`36
`
`37
`
`38
`
`overhead and is can consume a lot of forward link radio resources if used in a rapid rate
`control scheme and thereby lower the overall forward link sector throughput.
`
`In this contribution, we propose a per-user Forward Rate Control CHannel (F-RCCH) for
`each active MS sharing a packet data channel. This channel can be used to control each MS’s
`reverse link rate as a function of its reverse link channel condition. As described above, a BS
`may simply choose the reverse link rate to be proportional to average forward link rate
`throughput based on an averaged version of the DRC values received from the MS and the
`number of active MS's being served.
`
`The F-RCCH can be implemented in different ways. Since rapid rate changes are not
`necessarydesirable it may be sufficient to use a 2-bit indicator to send “rate up”/”rate
`down”/”no change” information once every L frames of 20 ms. The value of L can be
`determined by the configurableaccess network. One can then either define a new code
`channel for F-RCCH or simply puncture the power control channel bits to make room for
`rate control bits.
`
`
`4 SIMULATION RESULTS
`
`This section shows the results of a basic simulation to illustrate the potential reverse link
`sector throughput improvement when the individual MS data rates are chosen to be
`proportional to forward link ratethroughput. The following assumptions are made in this
`simulation:
`
`
`1. Number of hexagonal cells: 7 (center cell used for measurement)
`2. Number of sectors per cell: 3
`3. Path loss exponent: 3.5
`4. Standard deviation for shadow fading: 8 dB.
`5. Reverse link soft handoff: Yes, with three nearest sector’s.
`6. Forward link soft handoff: Sector selection
`7. Number of Active MS’s in each sector: 6 or 12
`8. Reference System: Randomly placed mobiles operating at a fixed data rate of 38.4
`kbit/s (6 MS’s per sector) or 19.2 kbit/s (12 MS’s per sector).
`9. New System Data Rates: 4.8, 9.6, 19.2, 38.4, 76.8, 153.6, 307.2, 614.4, and 1,228.8
`kbit/s
`10. Power allocated to data channel relative to pilot power: 0, 3, 6, 9, 12, 15, 18, 21, and
`24 dB for data rates 4.8, 9.6, 19.2, 38.4, 76.8, 153.6, 307.2, 614.4, and 1,228.8 kbit/s,
`respectively.
`
`6
`
`Formatted: Bullets and Numbering
`
`Ex. 1004 - Sierra Wireless, Inc.
`Sierra Wireless, Inc., et al. v. Sisvel S.P.A., IPR2021-01141
`Page 6 of 9
`
`

`

`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`29
`
`30
`
`31
`
`32
`
`33
`
`34
`
`35
`
`36
`
`37
`
`38
`
`39
`
`40
`
`Phoenix, AZ
`February 12, 2001
`
`
`
`3GPP2-C50-20010212- 011
`
`
`
`
`
`
`In the simulations we used identical path loss and shadow fading for the forward link and
`reverse link of an MS. Two scenarios are considered in simulating Raleigh fading. In the
`first scenario (Scenario A), no Raleigh fading is assumed, and only shadow fading and path
`loss are considered. This may correspond to the case of a very fast moving MS or a
`sufficiently long DRC averaging interval. In the second scenario (Scenario B), we added
`Raleigh fading that was independent on the reverse and forward links. We believe the actual
`system performance with DRC averaging will lie somewhere inbetween these two cases.
`
`In each test simulation, the positions of the MS’s were chosen randomly and independently
`in each sector. Each MS was power-controlled iteratively until the frame error rate (FER)2
`of eachthe MS converged to 1%. When a MS is in soft handoff, its FER is calculated by
`multiplying eachthe FER fromof all serving sectors. Further, in each test the position of the
`MS and its shadow fading value was were chosen randomly. In each the center sectorcell,
`sectorSIR throughputs were measured by repeating each test 100 times and averaging
`contributions from MS’s in the center cell.
`
`We assigned reverse link rates based on each MS’s forward-link SIR (or equivalently
`supportable forward-link rate), i.e., we assign reverse link rate
`revR , of i-th MS proportional i
`
`fwdR , as shown in the following i
`to its supportable forward-link rate
`
`
`
`aR
`
`
`
`fwd
`
`,
`
`i
`
`b
`
`,
`
`
`
`R
`
`rev
`
`,
`
`i
`
`
`
`
`where a and b are coefficients that are defined per sector and depend on the number of MS’s
`in theper sector, the maximum allowed per-sector reverse throughput Rmax, and a fairness
`factor, and  x is the largest possible allowed reverse rate not bigger than x. We keep the
`total transmission power of all MS’s the same as in the reference system for a fair
`comparison.
`
`We used the following fairness condition to guarantee a minimum certain fairness relative to
`its the maximum maximum achievable forward-link throughput assuming each MS gets 1/Nth
`of its the (Vedat, instantaneous forward-link rate assigned by the BS may be smaller
`than DRC. Perhaps we can say requested rate, DRC, or supportable rate.) forward-link
`bandwidth:
`
`
`Formatted
`Formatted
`
`min 
`
`R
`
`R
`
`rev
`i
`,
`/,i
`
`
`
`,
`
`N
`
`fwd
`
`
`min is the minimum relative fairness value.
`where N is the number of MS’s per sector and
`For example, a fairness value of 1/5 implies that the reverse link rate of a user will never
`exceed be smaller than the one-fifth of the actual forward link throughput of that user. We
`
`
`
`2 We employedused the FER curve of a 1000- bit long Turbo Code.
`
`7
`
`Ex. 1004 - Sierra Wireless, Inc.
`Sierra Wireless, Inc., et al. v. Sisvel S.P.A., IPR2021-01141
`Page 7 of 9
`
`

`

`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`Phoenix, AZ
`February 12, 2001
`
`
`
`3GPP2-C50-20010212-xxx011
`
`
`
`
`
`finddetermined the values of a and b sosuch that the above fairness condition is satisfied for
`all MS’s while also maintaining a certain maximum per-sector throughput Rmax. The value of
`Rmax is iteratively determined to make the total transmission power of MS’s equal to that of
`the reference system.
`
`The following table summarizes the improvement in performance found in the simulations.
`All improvements are relative to the reference system where MS's transmit at a fixed rate
`with a sector throughput of 230.4 kbit/s:
`
`Formatted
`Formatted
`Formatted
`Formatted
`
`
`
`
`
`# of MS Scenario Fairness Gain
`
`Throughput
`
`6
`
`12
`
`A
`
`B
`
`A
`
`B
`
` 1/5
`
`54%
`
`354 kbpskbit/s
`
` 1/10
`
`88%
`
`434 kbit/skbps
`
` 1/20
`
`161% 603 kbit/skbps
`
` 1/80
`
`225% 749 kbit/skbps
`
` 1/5
`
`20%
`
`275 kbit/skbps
`
` 1/10
`
`33%
`
`307 kbit/skbps
`
` 1/20
`
`57%
`
`362 kbit/skbps
`
` 1/80
`
`213% 721 kbit/skbps
`
` 1/5
`
`43%
`
`330 kbit/skbps
`
` 1/10
`
`66%
`
`382 kbit/skbps
`
` 1/20
`
`97%
`
`453 kbit/skbps
`
` 1/40
`
`172% 627 kbit/skbps
`
` 1/5
`
`17%
`
`270 kbit/skbps
`
` 1/10
`
`31%
`
`302 kbit/skbps
`
` 1/20
`
`50%
`
`344 kbit/skbps
`
` 1/40
`
`130% 531 kbit/skbps
`
`Fairness is the minimum relative fairness value
`
`min . It can be seen that cCompared to the
`reference system, where the sector throughput is about 230 kbps, reverse-link throughput is
`
`improved as much asby 130- 225% when the fairness factor is low. For a reasonable
`
`moderate value of 1/10 for the fairness factor (quite typical for web-based applications), we
`
`8
`
`Ex. 1004 - Sierra Wireless, Inc.
`Sierra Wireless, Inc., et al. v. Sisvel S.P.A., IPR2021-01141
`Page 8 of 9
`
`

`

`Phoenix, AZ
`February 12, 2001
`
`
`
`3GPP2-C50-20010212- 011
`
`
`
`
`
`get about 31- ~ 88% gain. OtherInter-sector interference factor , the ratio of inter-sector
`
`interference to intra-sector interference, is also reduced from about 0.89 to 0.4
`
`.
`
`
`
`5 CONCLUSIONS
`
`
`We described how the reverse link sector throughput can be improved by controlling the data
`rate of individual users based on the forward link data rate control (DRC) signals received
`from the MS’s. Making the reverse link rates proportional to the forward link data rates also
`appears to be a logical strategy for most web-based applications.
`
`We proposed that a new forward link rate control channel (F-RCCH) be added to 1xEV-DV
`to support such rate control.
`
`We described basic system simulation results which suggest potential significant
`improvements in reverse link sector throughput in a fully loaded data-only system.
`
`When used in a mixed voice-data mode, the proposed scheme will also indirectly help
`increase the number of voice users that can be mixed with data traffic on the same carrier.
`
`Significant potential throughput improvements shown in this contribution suggests that more
`attention needs to be paid to the design of the reverse link in 1xEV-DV.
`
`
`
`6 REFERENCE
`
`
`[1] C50-20001023-009 S.R0026_1xEV-DV_Stage1_v1.0, “High-Speed Data Enhancements for
`cdma2000 1x – Integrated Data and Voice Stage 1 Requirements”
`
`[2] K.I. Kim, Handbook of CDMA System Design, Engineering, and Optimization, Prentice Hall PTR,
`1999.
`
`[3] R. Knopp and P.A. Humblet, "Information capacity and power control in single-cell
`multi-user
`communications,"
`Proceedings
`IEEE
`International Conference
`on
`Communications ICC 95, Seattle, WA, USA, June 1995.
`Knopp/Humblet reference.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`29
`
`30
`
`31
`
`32
`
`33
`
`34
`
`35
`
`36
`
`9
`
`Ex. 1004 - Sierra Wireless, Inc.
`Sierra Wireless, Inc., et al. v. Sisvel S.P.A., IPR2021-01141
`Page 9 of 9
`
`