`U.S. Patent 7,206,587
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`
`APPLE INC.,
`ZTE(USA) INC.
`Petitioners
`
`v.
`
`INVT SPE LLC
`Patent Owner
`____________
`
`
`Case No. 2018-01474
`U.S. Patent No. 7,206,587
`____________
`
`
`
`
`
`
`PETITION FOR INTER PARTES REVIEW
`OF U.S. PATENT NO. 7,206,587
`
`
`
`
`
`
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`
`
`
`TABLE OF CONTENTS
`
`INTRODUCTION .......................................................................................... 1
`I.
`II. SUMMARY OF THE ’587 PATENT ........................................................... 1
`A. DESCRIPTION OF THE ALLEGED INVENTION OF THE ’587 PATENT ..................... 1
`B. SUMMARY OF THE PROSECUTION HISTORY OF THE ’587 PATENT ..................... 3
`C. LEVEL OF SKILL OF A PERSON HAVING ORDINARY SKILL IN THE ART ................ 6
`III. REQUIREMENTS FOR INTER PARTES REVIEW UNDER 37 C.F.R.
`§ 42.104 .................................................................................................................... 6
`A. GROUNDS FOR STANDING UNDER 37 C.F.R. § 42.104(A) ................................. 6
`B. IDENTIFICATION OF CHALLENGE UNDER 37 C.F.R. § 42.104(B) AND RELIEF
`REQUESTED ............................................................................................................ 6
`C. CLAIM CONSTRUCTION UNDER 37 C.F.R. § 42.104(B)(3) ................................ 8
`IV. THERE IS A REASONABLE LIKELIHOOD THAT THE
`CHALLENGED CLAIMS OF THE ’587 PATENT ARE UNPATENTABLE 8
`A. GROUND 1: BENDER IN VIEW OF JARVINEN RENDERS CLAIM 3 OBVIOUS ............ 8
`B. GROUND 2: BENDER IN VIEW OF PIRET RENDERS CLAIM 4 OBVIOUS .............. 23
`V. CONCLUSION ............................................................................................. 31
`VI. MANDATORY NOTICES UNDER 37 C.F.R. § 42.8(A)(1) ..................... 32
`A. REAL PARTY-IN-INTEREST ............................................................................ 32
`B. RELATED MATTERS ....................................................................................... 32
`C. LEAD AND BACK-UP COUNSEL ..................................................................... 33
`
`
`
`
`
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`
`I.
`
`INTRODUCTION
`
`Petitioners Apple Inc. and ZTE (USA) Inc. (collectively “Petitioners”)
`
`request an Inter Partes Review (“IPR”) of claims 3 and 4 (collectively, the
`
`“Challenged Claims”) of U.S. Patent No. 7,206,587 (“the ’587 Patent”). ’587
`
`Patent (Ex. 1001).
`
`II.
`
`SUMMARY OF THE ’587 PATENT
`A. Description of the alleged invention of the ’587 Patent
`
`The ’587 Patent generally describes a “cellular communication system” in
`
`which channel/reception quality between a base station and handset is measured by
`
`the handset and reported back to the base station as a means of selecting an
`
`efficient “transmission rate . . . according to the downlink channel quality.” ’587
`
`Patent (Ex. 1001) at 1:9-26. The handset “estimates the downlink channel quality”
`
`as a CIR (desired carrier to interference ratio) value and “transmits the result to the
`
`base station.” Id. at 1:30-41.
`
`The ’587 Patent notes that feedback signals can be “represented by numbers
`
`from 1 to N, with a higher number indicating a proportionally better downlink
`
`channel quality.” Id. at 1:53-56. But there is a risk that communications will slow
`
`or cease if the communication mode selected by the handset is not properly
`
`interpreted by the base station. Id. at 2:14-22 (“[I]f the communication mode
`
`determined by a communication terminal is received erroneously by the base
`
`
`
` 1
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`station due to deterioration of the channel conditions . . ., the base station will
`
`transmit data using that erroneous mode” such that “the communication terminal
`
`cannot demodulate or decode the data.”).
`
`To prevent such communication breakdowns, the ’587 Patent discloses
`
`methods for providing channel quality feedback that is less susceptible to errors. In
`
`an embodiment relevant to the Challenged Claims, the ’587 Patent teaches that the
`
`CIR value can be represented by a plurality of digits/bits including an upper digit
`
`and a lower digit, e.g., 8.7 dB—where “8” is the upper digit and “7” is the lower
`
`digit. Id. at 19:34-58. A key premise of the ’587 Patent’s proposed solution is that
`
`it is more important to accurately convey the upper digit than the lower digit. For
`
`example, misinterpreting 8.7 as 8.6 is a much less impactful than misinterpreting
`
`8.7 as 9.7. Singer Decl. (Ex. 1003) at ¶¶ 56-58.
`
`Accordingly, in one embodiment, the ’587 Patent proposes encoding the
`
`upper digit with more bits than the lower digit to create a longer code length for the
`
`upper digit comparatively to the lower digit. Id. at 20:33-67 (“The 6-bit coding
`
`section 1203 converts the value output from the upper digit information generation
`
`section 1201 (here, ‘8’) to a 6-bit code word” and the “4-bit coding section 1204
`
`converts the value output from the lower digit information generation section 1202
`
`(here, ‘7’) to a 4-bit code word.”). As the ’587 Patent explains, “a code word
`
`represented by 6 bits is less susceptible to being mistaken for another code word
`
`
`
` 2
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`than a code word represented by 4 bits. That is to say, in this embodiment, the
`
`value of the upper digit of a CIR value is less susceptible to errors.” Id. at 21:15-
`
`20.
`
`B.
`
`Summary of the prosecution history of the ’587 Patent
`
`The application that resulted in the ‘587 Patent was filed on Dec. 18, 2002 as
`
`U.S. Patent Application No. 10/321,623 (“the ’623 Application”). ’587 Patent (Ex.
`
`1001). The ‘623 Application is a continuation of U.S. Patent Application No.
`
`10/089,605 (“the ’605 Application”), which issued as U.S. Patent No. 6,760,590
`
`(“the ’590 Patent”). Id. The PCT application that resulted in the ’590 Patent was
`
`filed on April 2, 2001 as PCT/JP01/06654. ’590 Patent (Ex. 1001). The ’590
`
`Patent and ’587 Patent claim priority to two separate Japanese priority filings—
`
`JP2000-234420 (“the ’430 Application”), dated Aug. 2, 2000, and JP2000-285405
`
`(“the ’405 Application”), dated Sep. 20, 2000. Id. All prior art references relied
`
`on by Petitioners in this Petition are prior art regardless of which Japanese priority
`
`filing Patent Owner is entitled, if either. To the extent Patent Owner attempts to
`
`swear behind any prior art reference on the basis of one or more priority filings,
`
`Petitioners may demonstrate the priority filings fail to support the Challenged
`
`Claims.
`
`After a number of Office Actions and a Restriction Requirement, only four
`
`claims ultimately issued in the ’587 Patent. The relevant portion of the prosecution
`
`
`
` 3
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`history pertinent to these surviving claims is summarized below.
`
`A first Office Action issued on April 22, 2004, which rejected then-pending
`
`Claims 1-3, 8-10, 14, 16 and 18-20 and objected to the remaining claims as being
`
`dependent upon a rejected base claim, but otherwise allowable. ’587 Patent File
`
`History (Ex. 1002) at 244-49. On October 22, 2004, the Applicant cancelled all
`
`original Claims 1-20, proposed new claims 21-29, and alleged that the newly
`
`proposed claims incorporated subject matter the Examiner previously deemed
`
`allowable. Id. at 386-93.
`
`Applicant’s characterization was not accurate—proposed claims 25 and 28,
`
`which ultimately issued as the Challenged Claims, did not recite subject matter
`
`deemed allowable by the Examiner. For example, the Applicant alleged, “Claim 25
`
`recites features of [rejected original] claim 1 and allowable [original] claim 11.” Id.
`
`at 392. But newly proposed claim 25 recites “a code word whose length is
`
`proportional the digit’s degree of significance,” which is different than original
`
`claim 11’s recitation of “a code word whose length is proportional to a value of an
`
`upper digit.” Id. at 221, 389. Similarly, the Applicant alleged, “Claim 28 recites
`
`features of [rejected original] claim 10 and features generic to the allowable
`
`subject matter of [original] claims 11-13.” It is unclear what Applicant might have
`
`meant by this representation, but a review of original claims 11-13 reveals that the
`
`subject matter of newly proposed claim 28 was not previously deemed patentable.
`
`
`
` 4
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`
`Id. at 221-22, 390.
`
`On March 16, 2005, the Examiner issued a restriction requirement to which
`
`the Applicant responded on April 29, 2005, selecting to pursue then-pending
`
`claims 23-26 and 28. Id. at 468, 472.
`
`The Examiner issued another Office Action on October 25, 2005, rejecting
`
`proposed claims 25 and 28 as obvious over the Applicant’s own ’590 Patent in
`
`view of U.S. Patent Publication No. 2001/0050926 to Kumar. On January 25,
`
`2006, Applicant filed an Office Action Response, which argued that Kumar failed
`
`to teach “code words having variable lengths” or “coding a most significant bit of
`
`information so that it is less susceptible to error,” and concluded that proposed
`
`claims 25 and 28 were thus patentable over Kumar. Id. at 502-04. With the Office
`
`Action Response, Applicant also filed a Terminal Disclaimer, forfeiting any term
`
`beyond that of the ’590 Patent. Id. at 509.
`
`On March 1, 2006, the Examiner issued a new Office Action, rejecting
`
`proposed claims 25 and 28 as obvious over U.S. Patent No. 6,452,914 to Niemelia
`
`(“Niemelia”) in view of Kumar, which relied on Kumar for the same teachings as
`
`the previous Office Action. Id. at 525. The Applicants responded on April 26,
`
`2006, arguing that Niemelia was not prior art in light of the Japanese patent filings
`
`to which the ‘587 Patent claims priority. Id. at 536-38.
`
`On August 1, 2006, the Examiner issued a new Office Action, rejecting
`
`
`
` 5
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`proposed claims 25 and 28 as obvious over U.S. Patent No. 6,603,980 to Kitagawa
`
`(“Kitagawa”). Id. at 724-26. Applicant responded on November 1, 2006,
`
`distinguishing proposed claims 25 and 28 from Kitagawa’s teaching that a TPC bit
`
`is sent with a variable amplitude based on the measured channel quality. Id. at 739-
`
`42.
`
`A notice of allowance issued on December 24, 2003. ’587 Patent File
`
`History (Ex. 1002) at 751.
`
`C.
`
`Level of skill of a person having ordinary skill in the art
`
`A person having ordinary skill in the art (POSITA) at the time of the ’587
`
`Patent would have been a person having a Bachelor’s degree in electrical
`
`engineering or the equivalent plus three years of experience working with wireless
`
`communication systems or a Master’s degree in electrical engineering with an
`
`emphasis on communication systems or the equivalent plus one year of experience
`
`working with digital communication systems or in network engineering. Singer
`
`Decl. (Ex. 1003) at ¶ 39.
`
`III. REQUIREMENTS FOR INTER PARTES REVIEW UNDER 37 C.F.R.
`§ 42.104
`A. Grounds for standing under 37 C.F.R. § 42.104(a)
`
`Petitioners certify that the ’587 Patent is available for IPR and that the
`
`Petitioners are not barred or estopped from requesting IPR challenging the claims
`
`of the ’587 Patent. Specifically, Petitioners state: (1) each Petitioner is not the
`
`
`
` 6
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`owner of the ’587 Patent, (2) each Petitioner has not filed a civil action challenging
`
`the validity of any claim of the ’587 Patent, and (3) this Petition is filed less than
`
`one year after any of the Petitioners was served with a complaint alleging
`
`infringement of the ’587 Patent.
`
`B.
`
`Identification of challenge under 37 C.F.R. § 42.104(b) and relief
`requested
`
`In view of the prior art and evidence presented, claims 3 and 4 of the ’587
`
`Patent are unpatentable and should be cancelled. 37 C.F.R. § 42.104(b)(1). Further,
`
`based on the prior art references identified below, IPR of the Challenged Claims
`
`should be granted. 37 C.F.R. § 42.104(b)(2).
`
`Proposed Grounds of Unpatentability
`Ground 1: Claim 3 is obvious under § 103(a) over CDMA/HDR: A
`Bandwidth-Efficient High-Speed Wireless Data Service
`for
`Nomadic Users,” Bender, et al. (“Bender”) in view of U.S. Patent
`No. 6,470,470 to Jarvinen et al. (“Jarvinen”)
`Ground 2: Claim 4 is obvious under § 103(a) over Bender in view
`of U.S. Patent No. 4,747,104 to Piret (“Piret”)
`
`Exhibits
`
`1004, 1006
`
`1004, 1005
`
`
`
`Section IV identifies where each element of the Challenged Claims is found
`
`in the prior art. 37 C.F.R. § 42.104(b)(4). The exhibit numbers of the supporting
`
`evidence relied upon to support the challenges are provided below and the
`
`relevance of the evidence to the challenges raised are provided in Section IV. 37
`
`C.F.R. § 42.104(b)(5). Exhibits 1001–1029 are also attached.
`
`
`
` 7
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`C. Claim construction under 37 C.F.R. § 42.104(b)(3)
`
`In this proceeding, claim terms of an unexpired patent should be given their
`
`“broadest reasonable construction in light of the specification.” 37 C.F.R.
`
`§ 42.100(b); Cuozzo Speed Techs., LLC v. Lee, 136 S. Ct. 2131, 2144-46 (2016)
`
`(referred to herein as the BRI standard). Petitioners understand that the Patent Trial
`
`and Appeal Board (“PTAB”) may soon apply the standard applied by Article III
`
`courts (i.e., the Phillips standard). Petitioners have applied the plain and ordinary
`
`meaning of all claim terms below1 and do not believe any claim terms require
`
`express construction to resolve the proposed grounds of rejection presented herein.
`
`Petitioners do not, however, waive any argument in any litigation that claim terms
`
`in the ’587 Patent are indefinite or otherwise invalid nor do Petitioners waive the
`
`right to raise specific issues of claim construction in any litigation.
`
`IV. THERE
`IS A REASONABLE LIKELIHOOD THAT THE
`CHALLENGED CLAIMS OF THE
`’587 PATENT ARE
`UNPATENTABLE
`A. Ground 1: Bender in view of Jarvinen renders claim 3 obvious
`Bender was published in the July 2000 edition of IEEE Communications
`
`Magazine. Bender (Ex. 1004) at 1. The determination of whether a given
`
`reference qualifies as a prior art “printed publication” involves a case-by-case
`
`
`1 Petitioners believe the plain and ordinary meanings applied herein are consistent with
`
`both the BRI and Phillips standards.
`
`
`
` 8
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`inquiry into the facts and circumstances surrounding the reference’s disclosure to
`
`members of the public, and the key inquiry is whether the reference was made
`
`sufficiently publicly accessible before the priority date.
`
` EMC Corp. v.
`
`PersonalWeb Techs., Inc. et al., IPR2013-00087, 2014 WL 2090666, at *16
`
`(PTAB May 15, 2014) (citing In re Klopfenstein, 380 F.3d 1345, 1350 (Fed. Cir.
`
`2004) and In re Cronyn, 890 F.2d 1158, 1160 (Fed. Cir. 1989)).
`
`Petitioners provide a declaration from James L. Mullins, a librarian with a
`
`Doctor of Philosophy with a concentration in academic library administration and
`
`more than forty years of experience as an academic librarian. Mullins Decl. (Ex.
`
`1007) at ¶5, App. A. Dr. Mullins obtained Bender from the UW-Madison, College
`
`of Engineering. Id. at ¶¶ 24-26. In addition, Dr. Mullins obtained a copy of
`
`Bender from the IEEE Explore database, and concluded that Bender was publicly
`
`available no later than July 19, 2000. Mullins Decl. (Ex. 1007) at ¶¶ 27-37.
`
`For these reasons, Bender qualifies as a printed publication and is prior art
`
`with regard to the ’587 Patent under 35 U.S.C. § 102(a) (pre-AIA). Bender (Ex.
`
`1004). Bender discloses a CDMA system in which pilot signals are transmitted
`
`from the base station to user handsets, which use the pilot signals to measure
`
`channel quality. Id. at 71. Bender teaches that the measured channel quality is
`
`reported back to the base station on the reverse uplink:
`
`
`
` 9
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`[O]n the basis of the received common pilot from each access point
`(or base station), each access terminal (subscriber terminal) can
`measure
`the received signal-to-noise-plus-interference ratio
`(SNR). The data rate which can be supported to each user is
`proportional to its received SNR. This may change continuously,
`especially for mobile users. Thus, over each user’s reverse (uplink)
`channel, the SNR or equivalently the supportable data rate value is
`transmitted to the base station.
`Id. (emphasis added).
`
`As illustrated in the following annotated Table, the SNR2 values in Bender
`
`are composed of an upper digit reflecting the whole number component of the SNR
`
`(highlighted in blue) and a lower digit reflecting the fractional component of the
`
`SNR (highlighted in green):
`
`
`2 Bender explains that the calculated SNR is the ratio of received energy to interference
`
`and noise—EC/Nt. Id. at 73 (“EC represents the received signal energy density and Nt
`
`represents the total nonorthogonal single sided noise density. Nt comprises intercell
`
`interference, thermal noise, and possibly nonorthogonal intracell interference.”).
`
`
`
` 10
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`
`
`
`
`
`
`
`Thus, in both Bender and the ’587 Patent, channel quality measurements are
`
`communicated from a cellular handset to a base station.’587 Patent (Ex. 1001) at
`
`1:9-41, 7:31-35. Additionally, in both Bender and the ‘587 Patent, the channel
`
`quality measurement (SNR shown in the right-hand column of the table excerpted
`
`above) is composed of an upper (more significant) digit (highlighted in blue in the
`
`table excerpted above) and a lower (less significant) digit (highlighted in green in
`
`the table excerpted above). Id. at 19:34-58. Because both Bender and the ’587
`
`Patent teach cellular systems with channel quality feedback mechanisms based on
`
`a two-digit quality measurement, Bender is in the same field of endeavor and is
`
`
`
` 11
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`reasonably pertinent to the claims in the ’587 Patent. Singer Decl. (Ex. 1003) at ¶¶
`
`41-42. Therefore, Bender is also analogous to the claimed invention in the ’587
`
`Patent.
`
`Bender was not cited or considering during prosecution of the ’587 Patent
`
`and, as shown in this Petition, raises new invalidity issues relating to the
`
`challenged claims that have not previously been before the Patent Office.
`
`In connection with the teaching in Bender that its SNR values are
`
`transmitted to the base station, it was well known at the time of filing of the ’587
`
`Patent to employ various established methods for coding those SNR values prior to
`
`transmission, including, for example, Unequal Error Protection (“UEP”) when
`
`coding data of varying importance for transmission. One example is Jarvinen,
`
`which was filed February 6, 1998 and therefore qualifies as prior art with regard to
`
`the ’587 Patent under 35 U.S.C. § 102(e) (pre-AIA). Jarvinen (Ex. 1006). Jarvinen
`
`describes a cellular communication system that divides data into two categories
`
`based on its relative importance and performs error correction encoding to the
`
`more important data to increase the important data’s resiliency to communication
`
`errors. Id. at 1:7-40, 2:64-3:60, 7:44-65. Jarvinen further teaches that the system
`
`monitors the quality of the communication channel and adjusts the ratio of
`
`important and less important data accordingly. Id. The ’587 Patent similarly
`
`teaches that changes to communication modes (e.g., data rates) should be made
`
`
`
` 12
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`based on the measured channel quality. ’587 Patent (Ex. 1006) at 1:42-56. Because
`
`Jarvinen, like the ’587 Patent, discloses a cellular communication system directed
`
`to monitoring transmission channel quality and adjusting transmission parameters
`
`accordingly, it is in the same field of endeavor and is reasonably pertinent to the
`
`claims of the ’587 Patent. Singer Decl. (Ex. 1003) at ¶¶ 43-44. Therefore, Jarvinen
`
`is also analogous to the claimed invention in the ’587 Patent.
`
`Jarvinen was not cited or considered during prosecution of the ’587 Patent
`
`and, as shown in this Petition, raises new invalidity issues relating to the
`
`challenged claims that have not previously been before the Patent Office.
`
`i. Claim 3
`
`3. A communication terminal apparatus comprising:
`
`To the extent the preamble is deemed limiting, Bender discloses a
`
`communication terminal apparatus. Namely, Bender describes a cellular handset
`
`(subscriber terminal) that measures and reports downlink channel quality based on
`
`a received pilot signal from a cellular base station:
`
`[O]n the basis of the received common pilot from each access point
`(or base station), each access terminal (subscriber terminal) can
`measure
`the received signal-to-noise-plus-interference ratio
`(SNR). The data rate which can be supported to each user is
`proportional to its received SNR. This may change continuously,
`especially for mobile users. Thus, over each user’s reverse (uplink)
`
`
`
` 13
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`channel, the SNR or equivalently the supportable data rate value is
`transmitted to the base station.
`Bender (Ex. 1004) at 71 (emphasis added).
`
`[3(a)] a measuring device that measures downlink channel quality; and
`
`
`As described above, Bender describes a cellular handset (subscriber
`
`terminal) that measures and reports the received signal-to-noise-plus-interference
`
`ratio (“SNR”) based on a received pilot signal from a cellular base station:
`
`[O]n the basis of the received common pilot from each access point
`(or base station), each access terminal (subscriber terminal) can
`measure
`the received signal-to-noise-plus-interference ratio
`(SNR). The data rate which can be supported to each user is
`proportional to its received SNR. This may change continuously,
`especially for mobile users. Thus, over each user’s reverse (uplink)
`channel, the SNR or equivalently the supportable data rate value is
`transmitted to the base station.
`Bender (Ex. 1004) at 71 (emphasis added). Bender explains that the measured SNR
`
`is the ratio of received energy to interference and noise—EC/Nt. Id. at 73 (“EC
`
`represents the received signal energy density and Nt represents the total
`
`nonorthogonal single sided noise density. Nt comprises intercell interference,
`
`thermal noise, and possibly nonorthogonal intracell interference.”). The measured
`
`SNR values are illustrated in the following annotated Table 2:
`
`
`
` 14
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`
`
`
`
`
`
`
`Put simply, SNR compares the power of a signal to the power of the
`
`background noise and interference. Singer Decl. (Ex. 1003) at ¶ 32. It was well
`
`known in the art that a signal-to-noise-plus-interference ratio is a measurement of
`
`channel quality. Id. at ¶¶ 32, 41. If an SNR value drops below an acceptable
`
`threshold, for example due to atmospheric conditions such as rain or interference
`
`from other transmitters, the bit error rate on the link becomes unacceptable for
`
`reliable communications integrity, resulting in a low downlink channel quality
`
`value. Id. at ¶ 32.
`
`
`
` 15
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`[3(b)] a transmitter that transmits a notification signal to notify a base station
`apparatus of information generated based on said measured downlink channel
`quality, wherein:
`
`
`As discussed above regarding limitation [3(a)], Bender teaches that the
`
`cellular handset transmits the SNR value (i.e. a notification signal) to the base
`
`station. For an SNR value to be received at the base station, this value must be
`
`digitized (i.e. converted to a code word) before transmission, as claimed in
`
`limitation [3(b)(i)] below. Singer Decl. (Ex. 1003) at ¶¶ 46-48. To determine how
`
`the SNR values in Bender should be digitized prior to transmission, a POSITA
`
`would have looked to related disclosures describing data coding systems. Id.
`
`Jarvinen teaches just such a data coding system that is specifically directed to
`
`improving robustness to errors by applying error correction coding to only the
`
`more important information. Jarvinen (Ex. 1006) at Abstract, 5:40-54 (Jarvinen’s
`
`encoding process is described in detail below). Thus, the proposed combination of
`
`Bender and Jarvinen digitizes the SNR values disclosed in Bender.
`
`A POSITA would have recognized that the digitized information of the
`
`proposed combination would necessarily be transmitted to the base station by “a
`
`transmitter.” Singer Decl. (Ex. 1003) at ¶¶ 28, 30. In support, as illustrated in the
`
`following Jarvinen figures, digitized information in a wireless system is
`
`transmitted by a “TX/RX,” which is a transceiver comprising a transmitter and
`
`receiver:
`
`
`
` 16
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`
`
`
`Jarvinen (Ex. 1006) at Fig. 4, 13:51-52 (“The signal is transmitted to the
`
`information transfer connection using transceiver unit 240.”); see also Singer Decl.
`
`(Ex. 1003) at ¶ 30, 45
`
`(confirming
`
`that “TX/RX”
`
`is shorthand
`
`for
`
`transmitter/receiver—the primary components of a transceiver).
`
`[3(b)(i)] each of a plurality of digits representing the information of the
`notification signal is converted, prior to its transmission, to a code word whose
`code length is proportional to the digit's degree of significance.
`
`
`Bender discloses a plurality of digits representing information of the
`
`notification signal, for notifying the base station of the measured SNR value.
`
`Bender (Ex. 1004) at Table 2. These plurality of digits are comprised of an upper
`
`digit (highlighted blue) and a lower digit (highlighted green), as illustrated in the
`
`following annotated Table 2:
`
`
`
` 17
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`
`
`
`
`
`
`
`In connection with the teachings in Bender, it was well established at the
`
`time of filing of the ’587 patent that the SNR value would be coded in some
`
`manner prior to transmission. Singer Decl. (Ex. 1003) at ¶¶ 46-47. CDMA cellular
`
`systems, like all digital wireless systems, communicate information that has been
`
`converted to binary code words, i.e., series of ones and zeroes. The conversion
`
`process is generally utilized by employing an analog to digital converter, where
`
`analog values are afforded specific binary codewords to represent the analog data
`
`in a manner that a computer can recognize. Id. Accordingly, the Bender SNR
`
`
`
` 18
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`values must necessarily be converted to binary code words prior to being
`
`transmitted over the wireless infrastructure. Id.
`
`To determine how the SNR values in Bender should be represented as binary
`
`code words prior to transmission, a POSITA would have looked to related
`
`disclosures describing data coding systems, such as Jarvinen as described above in
`
`limitation [3(b)]. Singer Decl. (Ex. 1003) at ¶¶ 41-49. Jarvinen is directed to
`
`improving robustness to errors by applying error correction coding to only the
`
`more
`
`important
`
`information. Jarvinen (Ex. 1006) at Abstract, 5:40-54.
`
`Specifically, Jarvinen teaches that encoded data should be divided into most
`
`important information and less important information:
`
`The speech parameter bits are in FIG. 1C presented in the speech
`encoding importance order in such a way that the subjectively most
`important bits are presented topmost, and the less important bits
`at the bottom.
`Id. at 8:36-39. Jarvinen explains that “the most significant bits . . . are more
`
`important and they must be protected carefully,” but that “[t]he least significant
`
`bits . . . can be left unprotected” because the impact of their erroneous receipt “is
`
`small.” Id. at 3:55-60.
`
`Jarvinen further teaches that a convolutional error correcting coder should
`
`be used to increase the number of bits (i.e. code length) used to represent the most
`
`significant information. In the following example, the most important 182 bits are
`
`
`
` 19
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`represented by 378 coded bits before being combined with the least important 78
`
`bits to create 456 bit frames:
`
`Error detection- and error correction parameters are formed for
`the most important 182 bits (Class I). At first 3-bit CRC-error
`detection parameters are calculated in block 105 for the 50 most
`important bits, after which the generated bit stream (182+3 bits) is
`directed
`to
`convolution
`encoder 106.
`Convolution
`encoder 106 calculates for the bits a ½-rate convolution code with
`four tail bits. The result is 378 bits (2*182+2*4+2*3) of convolution
`coded data 107. Convolution coded data 107 is directed further to
`multiplexer 109, in which it is combined with the least important
`78 bits (Class II, ref. 108). In all channel encoder 104 produces to the
`output (ref. 110) 456 bits for each 20 ms speech frame
`Id. at 7:52-63 (emphasis added); see also id. at 10:4-7 (“a sufficiently effective
`
`error correction is obtained by focusing the error correction code on only the most
`
`important of the 260 speech parameter bits.”). Therefore, Jarvinen teaches that a
`
`code word’s code length is proportional to each digit’s degree of significance,
`
`where more significant information is represented by more bits and therefore a
`
`longer code length comparatively to less significant information. Singer Decl. (Ex.
`
`1003) at ¶¶ 36, 43, 57. For example, utilizing the combination of Bender and
`
`Jarvinen, an SNR reading of “7.2” (see Bender (Ex. 1004) at Table 2) would be
`
`digitized into unequal length code words, where the “7” would receive more bits
`
`and therefore a longer code length when compared to the “2”. Id. The “7” is a more
`
`
`
` 20
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`significant digit than the “2” due to the difference in magnitude between the two
`
`digits and
`
`therefore requires a greater degree of error protection. Id.
`
`Misinterpreting a “7.2” SNR as an “8.2” SNR has a much larger resultant error
`
`than mistaking a “7.2” SNR as a “7.3” SNR. Id. Therefore, the resulting length of
`
`code words representing digitized SNR values would be proportional to the digit’s
`
`degree of significance. Id.
`
`It would have been obvious to encode the SNR values of Bender into code
`
`words whose “code length is proportional to the digit’s degree of significance” in
`
`accordance with the teachings of Jarvinen. The motivations to make such a
`
`modification of the Bender system are many, including express motivations from
`
`the references themselves. First, as discussed above, Bender teaches that its SNR
`
`values are composed of two components—an upper digit reflecting the whole
`
`number component of the SNR and a lower digit reflecting the fractional
`
`component of the SNR where correctly relaying the upper digit is far more
`
`important than correctly relaying the lower digit. Singer Decl. (Ex. 1003) at ¶ 57.
`
`A POSITA would have further been well versed in Unequal Error Protection
`
`(“UEP”) techniques, which were prevalent in the prior art and which included
`
`many approaches to data transmissions where more significant data requires more
`
`protection to potential transmission errors. Id. at ¶¶ 48-49. A POSITA with
`
`knowledge of UEP would have understood that the Bender teachings effectively
`
`
`
` 21
`
`
`
`IPR2018-01474
`U.S. Patent No. 7,206,587
`direct a skilled artisan to employ encoding techniques to ensure that the upper digit
`
`of the SNR value is not misinterpreted by the base station. Id. at ¶¶ 48-49, 57.
`
`Second, as noted above, Jarvinen provides express motivation to increase the bit
`
`count proportionally to the significance of the information such that it is protected
`
`from potential transmission and reception errors. Jarvinen (Ex. 1006) at Abstract,
`
`5:40-54, 7:52-63.
`
`A POSITA would have recognized that utilizing a proportional bit count
`
`according to the significance of each digit of Bender’s SNR value would have
`
`provided numerous improvements, increasing the chances of interpreting the SNR
`
`value either correctly or with only insignificant error. Singer Decl. (Ex. 1003) at ¶¶
`
`41-49. Further, the modification to Bender would have been straightforward,
`
`would not have required undue experimentation, and would have produced
`
`predictable results. Id. at ¶ 49. Upon reading the disclosure of Bender, a POSITA
`
`would have recognized that representing the SNR upper digit with a longer code
`
`length than the lower digit would create more robustness to communication errors