`
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`_____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_____________
`
`HTC CORPORATION, HTC AMERICA, INC.,
`and APPLE INC.,
`Petitioners,
`
`v.
`
`INVT SPE LLC,
`Patent Owner.
`_____________
`
`Case Nos. IPR2018-01555 and IPR2018-01581
`U.S. Patent No. 7,848,439
`_____________
`
`
`PETITIONERS’ CONSOLIDATED REPLY TO
`PATENT OWNER’S CONSOLIDATED RESPONSE
`
`
`
`
`
`
`
`

`

`
`
`
`
`
`
`
`LIST OF PREVIOUSLY FILED EXHIBITS
`
`EXHIBIT
`
`1001
`
`DESCRIPTION
`
`DATE
`
`U.S. Patent No. 7,848,439 to She et
`al. (“the ’439 Patent”)
`
`December 7, 2010
`
`1002
`
`File History of the ’439 Patent
`
`n/a
`
`1003
`
`1004
`
`1005
`
`1006
`
`U.S. Patent No. 6,904,283 to Li et al.
`(“Li”)
`
`June 7, 2005
`
`U.S. Patent No. 7,221,680 to Vijayan
`et al. (“Vijayan”)
`
`May 22, 2007
`
`U.S. Patent No. 6,721,569 to Hashem
`et al. (“Hashem”)
`
`April 13, 2004
`
`U.S. Patent No. 5,596,604 to Cioffi et
`al. (“Cioffi”)
`
`January 21, 1997
`
`1007
`
`Declaration of Zhi Ding, Ph.D.
`
`August 22, 2018
`
`1008
`
`Curriculum Vitae of Zhi Ding, Ph.D.
`
`n/a
`
`1009
`
`1010
`
`1011
`
`1012
`
`U.S. Patent No. 6,167,031 to
`Olofsson et al.
`
`December 26, 2000
`
`Complaint, Inventergy, Inc. v. HTC
`Corporation, and HTC America, Inc.,
`C.A. No.: 17-cv-200-VAC-CJB (D.
`Del. 2017)
`
`February 27, 2017
`
`Inventergy’s Voluntary Dismissal
`Without Prejudice
`
`May 25, 2017
`
`Complaint, INVT SPE LLC v. HTC
`Corporation, and HTC America, Inc.,
`2:17-cv-03740-JMV-JBC (D.N.J.
`2017)
`
`May 25, 2017
`
`-ii-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`EXHIBIT
`
`DESCRIPTION
`
`DATE
`
`1013
`
`1014
`
`1015
`
`2001
`
`2002
`
`2101
`
`HTC Corp. and HTC America, Inc.’s
`Motion To Transfer
`
`March 9, 2018
`
`INVT’S Opposition to HTC Corp.
`and HTC America, Inc.’s Motion to
`Transfer
`
`March 23, 2018
`
`HTC Corporation and HTC America,
`Inc.’s Reply Brief In Support Of
`Their Motion To Transfer
`
`April 2, 2018
`
`Expert Declaration of Dr. Branimir
`Vojcic
`
`December 11, 2018
`
`Curriculum Vitae of Dr. Branimir
`Vojcic
`
`n/a
`
`Expert Declaration of Dr. Branimir
`Vojcic in Support of Patent Owner’s
`Response
`
`June 26, 2019
`
`2102
`
`Curriculum Vitae of Dr. Branimir
`Vojcic
`
`n/a
`
`
`
`-iii-
`
`
`
`
`
`
`
`
`
`
`
`
`

`

`EXHIBITS FILED WITH PETITIONERS’ CONSOLIDATED REPLY
`
`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`EXHIBIT
`
`1016
`
`1017
`
`1018
`
`1019
`
`1020
`
`1021
`
`1022
`
`DESCRIPTION
`
`DATE
`
`Declaration of Zhi Ding, Ph.D. in
`Support of Petitioners’ Consolidated
`Reply to Patent Owner’s
`Consolidated Response
`
`Transcript of Deposition of Dr.
`Branimir Vojcic in this consolidated
`inter partes review proceeding
`
`Jacky S. Chow, Jerry C. Tu, and John
`M. Cioffi, A Discrete Multitone
`Transceiver System for HDSL
`Applications, IEEE JOURNAL ON
`SELECTED AREAS IN
`COMMUNICATIONS, Vol. 9, No. 6,
`Pgs. 895-908
`
`N.M. Maslin, High data rate
`transmissions over h.f. links, Radio
`and Electronic Engineer, Vol. 52, No.
`2, Pgs. 75-87
`
`Sections 6.3.10.1, 6.3.17.4 and
`8.1.4.1.2.5 from the IEEE 802.16-
`2004 Standard
`
`October 30, 2019
`
`October 3, 2019
`
`August 1991
`
`February 1982
`
`October 1, 2004
`
`Section 3.7 of Erik Dahlman, Stefan
`Parkvall, and Johan Sköld, 4G
`LTE/LTE-ADVANCED FOR MOBILE
`BROADBAND
`
`Peter S. Chow, John M. Cioffi, and
`John A.C. Bingham, A Practical
`Discrete Multitone Transceiver
`Loading Algorithm for Data
`Transmission over Spectrally Shaped
`
`2011
`
`1995
`
`
`
`
`
`
`-iv-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`EXHIBIT
`
`DESCRIPTION
`
`DATE
`
`1023
`
`1024
`
`1025
`
`1026
`
`Channels, IEEE TRANSACTIONS ON
`COMMUNICATIONS, Vol. 43, No.
`2/3/4, Pg. 773
`
`Andreas Czylwik, Adaptive OFDM
`for wideband radio channels,
`PROCEEDINGS OF 1996 IEEE GLOBAL
`TELECOMMUNICATIONS CONFERENCE,
`Vol. 1, pp. 713-718
`
`1996
`
`March 21-25, 2004
`
`September 2010
`
`2007-2008
`
`Alexander M. Wyglinski, Fabrice
`Labeau, and Peter Kabal, An Efficient
`Bit Allocation Algorithm for
`Multicarrier Modulation, IEEE
`WIRELESS COMMUNICATIONS AND
`NETWORKING CONFERENCE, Pg.
`1194, 1195
`
`Liastos Christos, Koutitas George,
`Virtual Labs #2: ‘Fading Process-
`Channel Characterization’:
`Documentation, International Helenic
`University
`
`Ranjeet Mohapatra, Sameer Ranjan
`Behera, Application of Visual
`Simulation in Communication
`Systems, National Institute of
`Technology
`
`
`
`-v-
`
`
`
`
`
`
`
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`TABLE OF CONTENTS
`
`
`
`I.
`
`II.
`
`INTRODUCTION ........................................................................................... 1
`
`A POSITA WOULD HAVE HAD EXPERIENCE WITH
`MULTICARRIER NETWORKS, BOTH WIRED AND WIRELESS .......... 2
`
`III. CLAIM CONSTRUCTION ............................................................................ 5
`
`A.
`
`B.
`
`“Subband,” “Pattern Storage Section,” and “Modulation
`Parameters with a Highest Classification” Do Not Need an
`Express Construction for Petitioners to Prevail .................................... 5
`
`INVT Improperly Limits the Phrase “Patterns For Selecting
`Subbands” (Claims 1 and 8) .................................................................. 5
`
`1.
`
`2.
`
`“Patterns for selecting subbands” need not be stored in
`isolation from the subband groups they define. .......................... 5
`
`INVT fails to explain why “patterns for selecting
`subbands” should be limited to “fixed rules.” ............................ 9
`
`IV. CLAIM 1 OF THE ’439 PATENT IS OBVIOUS OVER LI IN VIEW
`OF VIJAYAN AND HASHEM ...................................................................... 9
`
`A.
`
`Li in View of Vijayan and Hashem Teaches the Claimed
`“Parameter Deciding Section” .............................................................. 9
`
`1.
`
`2.
`
`INVT improperly addresses Li, Vijayan, and Hashem
`individually, rather than in combination. .................................... 9
`
`In combination, Li in view of Vijayan and Hashem
`teaches the claimed “parameter deciding section.” .................. 10
`
`B.
`
`Li in View of Vijayan Teaches the Claimed “Pattern Storage
`Section” ............................................................................................... 11
`
`1.
`
`2.
`
`3.
`
`Li’s “group index” means that patterns for selecting
`subbands are necessarily stored in advance. ............................. 11
`
`Li’s Figure 6 teaches “patterns for selecting subbands”
`under the Board’s or INVT’s construction. .............................. 14
`
`Even if Li’s subscriber measures the subbands in real
`time, the selection of subbands is still done within the
`context of pre-defined, indexed subband groups. ..................... 16
`
`
`
`
`
`
`-vi-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`C.
`
`INVT Does Not Dispute That Li in View of Vijayan and
`Hashem Teaches the Remaining Limitations in Claim 1 .................... 19
`
`V.
`
`CLAIMS 2-7 ARE OBVIOUS OVER LI IN VIEW OF VIJAYAN
`AND HASHEM ............................................................................................. 19
`
`VI. A POSITA WOULD HAVE BEEN MOTIVATED TO COMBINE LI
`WITH VIJAYAN ........................................................................................... 19
`
`A. A POSITA Would Have Been Motivated to Combine Li with
`Vijayan to Reduce Overhead ............................................................... 20
`
`B.
`
`C.
`
`D.
`
`E.
`
`Combining Li with Vijayan Would Have Furthered Li’s Goal of
`Reducing Overhead ............................................................................. 22
`
`Li Does Not Teach Away from Using Vijayan’s Joint
`Modulation and Coding Parameters .................................................... 23
`
`INVT Improperly Relies upon a Bodily Incorporation of
`Vijayan into Li ..................................................................................... 27
`
`Petitioners’ Expert Gives Ample Support for His Opinion That
`a POSITA Would Have Been Motivated to Combine Li with
`Vijayan to Reduce Overhead ............................................................... 29
`
`VII. A POSITA WOULD HAVE BEEN MOTIVATED TO COMBINE LI
`AND VIJAYAN WITH HASHEM ............................................................... 30
`
`A. Having the Subscriber Decide the Modulation and Coding
`Parameters (per Hashem) in the Combined System Would Have
`Further Reduced Overhead (as Further Taught by Hashem) .............. 31
`
`B.
`
`INVT Improperly Relies upon a Bodily Incorporation of
`Hashem into Li .................................................................................... 32
`
`VIII. THE MOTIVATION TO COMBINE DOES NOT RELY ON
`HINDSIGHT .................................................................................................. 33
`
`IX. CLAIM 8 IS OBVIOUS OVER LI IN VIEW OF VIJAYAN,
`HASHEM, AND CIOFFI .............................................................................. 34
`
`A.
`
`B.
`
`Li in View of Vijayan and Hashem Teaches the Claimed
`“Parameter Deciding Section” ............................................................ 34
`
`Li in View of Vijayan and Cioffi Teaches “Assigning a Weight
`per Subband Group to a Sum of Information Bits That Are Able
`to Be Assigned to All of the Subbands Within the Subband
`Group” ................................................................................................. 34
`
`
`
`
`
`
`-vii-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`C.
`
`INVT Does Not Dispute That Li in View of Vijayan, Hashem,
`and Cioffi Teaches the Remaining Limitations in Claim 8 ................ 37
`
`X. A POSITA WOULD HAVE BEEN MOTIVATED TO COMBINE
`LI, VIJAYAN, AND HASHEM WITH CIOFFI .......................................... 37
`
`A.
`
`B.
`
`C.
`
`Cioffi Is Analogous Art ....................................................................... 37
`
`A POSITA Would Have Recognized That Cioffi’s Weighting
`Would Be Used to Address Other Sources of Interference in the
`Combination ........................................................................................ 39
`
`Petitioners’ Expert Gives Ample Support for His Opinion on
`Motivation to Combine Li, Vijayan, Hashem, and Cioffi .................. 40
`
`XI. CONCLUSION .............................................................................................. 40
`
`
`
`
`
`
`
`-viii-
`
`
`
`
`
`

`

`
`
`
`
`
`Cases
`
`TABLE OF AUTHORITIES
`
`Page(s)
`
`Allied Erecting & Dismantling Co. v. Genesis Attachments, LLC
`825 F.3d 1373 (Fed. Cir. 2016) ........................................................ 27, 28, 32, 33
`
`Bradium Techs. LLC v. Iancu
`923 F.3d 1032 (Fed. Cir. 2019) ............................................................................ 9
`
`Commonwealth Scientific & Indus. Research Org. v. Buffalo Tech.
`(USA), Inc.
`542 F.3d 1363 (Fed. Cir. 2008) .......................................................................... 38
`
`Daiichi Sankyo Co. v. Apotex, Inc.
`501 F.3d 1254 (Fed. Cir. 2007) ............................................................................ 3
`
`Environmental Designs, Ltd. v. Union Oil Co.
`713 F.2d 693 (Fed. Cir. 1983) .............................................................................. 3
`
`Harmonic, Inc. v. Avid Tech., Inc.
`IPR2013-00252, paper 27 ................................................................................... 23
`
`In re Keller
`642 F.2d 413 (C.C.P.A. 1981) .............................................................................. 9
`
`Medichem, S.A. v. Rolabo, S.L.
`437 F.3d 1157 (Fed. Cir. 2006) .......................................................................... 28
`
`Meiresonne v. Google, Inc.
`849 F.3d 1379 (Fed. Cir. 2017) .......................................................................... 23
`
`In re Merck & Co.
`800 F.2d 1091 (Fed. Cir. 1986) ............................................................................ 9
`
`In re Mouttet
`686 F.3d 1322 (Fed. Cir. 2012) ............................................................................ 9
`
`Princeton Biochemicals, Inc. v. Beckman Coulter, Inc.
`411 F.3d 1332 (Fed. Cir. 2005) .......................................................................... 38
`
`
`
`
`
`
`-ix-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`HTC Corporation, HTC America, Inc., and Apple Inc. (collectively,
`
`“Petitioners”) submit this Consolidated Reply to Patent Owner’s Consolidated
`
`Response (“Response”). The arguments in this Reply are confirmed and supported
`
`by the Declaration of Zhi Ding, Ph.D. filed herewith (Ex. 1016), along with the
`
`Declaration of Zhi Ding, Ph.D. filed with the Petitions (Ex. 1007).
`
`I.
`
`INTRODUCTION
`
`The Response misapplies the law, argues limitations not in the claims,
`
`excludes embodiments, and mischaracterizes both the references and the ’439
`
`Patent, in an attempt to avoid obviousness by Li, Vijayan and Hashem of claim 1,
`
`and by Li, Vijayan, Hashem, and Cioffi of claim 8. For claims 1-7 in particular,
`
`INVT misapplies the law by addressing the references individually rather than in
`
`combination. This is a veiled attempt to avoid the combinations. For claim 8,
`
`INVT fails to make a plausible argument that wireless and ADSL are not
`
`analogous art. INVT’s construction of “patterns for selecting subbands”
`
`improperly relies on limitations not in claims 1 or 8.
`
`Additionally, INVT’s attack on motivation to combine relies on a contention
`
`that Li and Vijayan, and Li and Hashem, teach away from each other while
`
`ignoring how Petitioners are using these references. Last, INVT contends that Dr.
`
`Ding’s declaration is conclusory, when in fact it is well supported by facts or data,
`
`plus his expertise in the field.
`
`
`
`
`
`
`-1-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`II. A POSITA WOULD HAVE HAD EXPERIENCE WITH
`MULTICARRIER NETWORKS, BOTH WIRED AND WIRELESS
`
` A
`
` person of ordinary skill in the art (“POSITA”) would have had a
`
`bachelor’s degree in electrical engineering, computer science, or an equivalent
`
`field, plus at least three years of experience working in wireless communication
`
`systems, communication networks, and signal processing. Ex. 1016 ¶61.
`
`Particularly, a POSITA would have had experience with wireless and wired
`
`multicarrier communication networks, including wired systems like the ADSL
`
`(“asymmetric digital subscriber line”) system taught in Cioffi (cited against claim
`
`8). Ex. 1006; Ex. 1016 ¶62. There is not strict separation between wired and
`
`wireless communication engineering. Id. For example, both wired and wireless
`
`systems must transmit/receive data over distortive channels that can degrade
`
`performance. Id. Both systems typically require encoding/decoding,
`
`modulation/demodulation, and multiplexing/demultiplexing, among other steps.
`
`Id.
`
`INVT claims that a POSITA would not necessarily have had experience with
`
`general “communication networks” or “signal processing” outside of wireless
`
`systems. Response at 15. INVT is wrong and cites no support for its assertion,
`
`other than a statement in the ’439 Patent that the technical field relates to a wireless
`
`communication OFDM (“orthogonal frequency-division multiplexing”) system.
`
`
`
`
`
`
`-2-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`But the OFDM principles applicable to wireless networks like WiFi and
`
`cellular are the same as those applicable to wired networks like ADSL. Ex. 1016
`
`¶64. OFDM is at the heart of major wireless and wired standards. Id. The concept
`
`of multicarrier modulations used by OFDM was applied in both wired applications
`
`such as ADSL1 and wireless applications2 before the ’439 Patent. Id. Frequency-
`
`selective attenuation is a problem in both ADSL and wireless communications, and
`
`a POSITA would have been familiar with how OFDM handles this problem in both
`
`implementations. Id.
`
`Also, a POSITA’s level of skill is not limited to the exact technical field as
`
`stated in the patent. Daiichi Sankyo Co. v. Apotex, Inc., 501 F.3d 1254, 1256 (Fed.
`
`Cir. 2007) (considering “(1) the educational level of the inventor; (2) type of
`
`problems encountered in the art; (3) prior art solutions to those problems; (4)
`
`rapidity with which innovations are made; (5) sophistication of the technology; and
`
`(6) educational level of active workers in the field” to determine the level of skill)
`
`(citing Environmental Designs, Ltd. v. Union Oil Co., 713 F.2d 693, 696 (Fed. Cir.
`
`1983)).
`
`A typical curriculum for a bachelor’s degree in electrical engineering,
`
`
`1 Ex. 1018.
`
`2 Ex. 1019.
`
`
`
`
`
`
`-3-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`computer science, or an equivalent field would have included courses/experiences
`
`pertaining to communication networks and signal processing for both wired and
`
`wireless communications. Ex. 1016 ¶66. The curriculum would have used
`
`materials discussing both wired and wireless systems. Id. People working with
`
`OFDM systems would have completed such a curriculum. Id.
`
`Additionally, understanding prior-art solutions to the types of frequency-
`
`selective attenuation encountered in OFDM systems would have required a
`
`POSITA to have experience with communication networking protocols in wired
`
`systems like ADSL, as well as signal processing concepts such as channel
`
`estimation in both wired and wireless communications. Ex. 1016 ¶67.
`
`Moreover, the field of OFDM systems was rapidly changing before the
`
`alleged invention. Ex. 1016 ¶69. OFDM was used in wired ADSL subsequent to
`
`1995. Id. By 1999, OFDM was being used for WiFi. Id. By October of 2004,
`
`OFDM was being used in the wireless standard WiMAX. Id. In light of the
`
`rapidity and progression of innovation in OFDM systems, a POSITA would have
`
`had experience with both wired and wireless OFDM. Id.
`
`For these reasons, a POSITA would have had experience with both wired
`
`and wireless communications, including wired systems like the ADSL system
`
`taught by Cioffi. Ex. 1016 ¶70.
`
`
`
`
`
`
`-4-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`III. CLAIM CONSTRUCTION
`
`A.
`
`“Subband,” “Pattern Storage Section,” and “Modulation
`Parameters with a Highest Classification” Do Not Need an
`Express Construction for Petitioners to Prevail
`
`In instituting these proceedings, the Board held that “pattern storage section”
`
`and “modulation parameters with a highest classification” do not need an express
`
`construction. Decision Instituting Inter Partes Review of IPR2018-01555
`
`(“IPR2018-01555 Decision”), 13-14. The Board also held it was unnecessary to
`
`determine whether “subband” is limited to a group of subcarriers “in neighboring
`
`positions on the frequency domain.” IPR2018-01555 Decision at 12; Decision
`
`Instituting Inter Partes Review of IPR2018-01581 (“IPR2018-01581 Decision”),
`
`12. Petitioners agree for purposes of these proceedings and, for the reasons in the
`
`Board’s decisions, that these terms do not need an express construction for
`
`Petitioners to prevail.
`
`B.
`
`INVT Improperly Limits the Phrase “Patterns For Selecting
`Subbands” (Claims 1 and 8)
`
`1.
`
`“Patterns for selecting subbands” need not be stored in
`isolation from the subband groups they define.
`
`At Institution, the Board acknowledged that Petitioners proposed, in part,
`
`“patterns for selecting subbands” means “configurations or arrangements of
`
`subbands,” and Patent Owner proposed, in part, “patterns for selecting subbands”
`
`means “fixed rules for choosing subbands.” IPR2018-01555 Decision at 14-15.
`
`
`
`
`
`
`-5-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`The Board characterized the dispute as follows:
`
`[W]hether [1] the requisite patterns must be used to assign subbands
`to a subband group (that is, create the subband group) or [2] the
`requisite patterns encompass identifying which subbands are in a
`particular subband group that has been defined previously.
`
`IPR2018-01555 Decision at 16. The Board noted there is merit in the first
`
`interpretation and invited Petitioners’ response. Id.
`
`For purposes of this proceeding, Petitioners agree that the claimed patterns
`
`are used to assign subbands to groups. But Petitioners disagree that, under the
`
`broadest reasonable interpretation (“BRI”), these patterns must be stored in
`
`isolation from the groups they define (or an identification of the subbands that
`
`make up those groups). Such a narrow interpretation is inconsistent with the
`
`intrinsic record, which depicts patterns that concretely define which subbands
`
`constitute which subband groups. Thus, under the BRI, there is no material
`
`distinction between representing a pattern in the abstract and representing a pattern
`
`by identifying which subbands constitute the groups created using the pattern.
`
`In each of the ’439 Patent’s examples, the pattern is described both in the
`
`abstract and also concretely by identifying which subbands will constitute the
`
`groups created using the pattern. The first pattern, depicted in Figure 8, creates
`
`groups of every four adjacent subbands:
`
`
`
`
`
`
`-6-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`
`
`While this “pattern” can be expressed in the abstract (e.g., every four adjacent
`
`subbands), the “pattern” also concretely defines which subbands belong to which
`
`groups. Namely, subbands 1-4 constitute a first group, subbands 5-8 constitute a
`
`second group, etc. Because the abstract “pattern” also concretely defines the
`
`makeup of all groups created using the pattern, under the BRI there is no material
`
`distinction between storing the abstract pattern definition and storing an
`
`identification of the subbands that make up the groups created with the abstract
`
`pattern.
`
`
`
`Figures 9-10 depict similar relationships between patterns and the groups
`
`defined by the patterns.
`
`
`
`
`
`
`-7-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`
`
`
`
`Figure 9 depicts the abstract “pattern” that a group consists of every fourth
`
`subband. This “pattern” concretely defines a first group as subbands 1, 5, 9, etc., a
`
`second group as subbands 2, 6, 10, etc., and so on. Similarly, Figure 10 depicts the
`
`abstract “pattern” that a single group includes every subband. This also concretely
`
`defines which specific subbands belong to a group.
`
`Thus, every exemplary “pattern” in the ’439 Patent can be described either
`
`in the abstract or concretely by identifying the makeup of groups created using the
`
`pattern (e.g., Group 1 = subbands 1, 5, 9, etc.). In either case, the same
`
`information is conveyed because the pattern defines the group makeup and the
`
`group makeup defines the pattern.
`
`In sum, Petitioners agree that the claimed patterns are used to assign
`
`subbands to subband groups, but under the BRI these patterns need not be stored in
`
`isolation from the groups that they define or the subband makeup of those groups.
`
`
`
`
`
`
`-8-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`INVT fails to explain why “patterns for selecting subbands”
`should be limited to “fixed rules.”
`
`2.
`
`As noted above, INVT contends that “patterns for selecting subbands”
`
`should be construed in part as “fixed rules for choosing subbands based on
`
`frequency.” Response at 17. INVT, however, fails to explain what it means for
`
`rules to be “fixed” or for how long and by what entity they are “fixed.” All that
`
`claims 1 and 8 require is that the patterns be stored “in advance.” Ex. 1016 ¶86.
`
`IV. CLAIM 1 OF THE ’439 PATENT IS OBVIOUS OVER LI IN VIEW
`OF VIJAYAN AND HASHEM
`
`A. Li in View of Vijayan and Hashem Teaches the Claimed
`“Parameter Deciding Section”
`
`1.
`
`INVT improperly addresses Li, Vijayan, and Hashem
`individually, rather than in combination.
`
`The test for obviousness is what the combined teachings of the references
`
`would have suggested. See In re Mouttet, 686 F.3d 1322, 1333 (Fed. Cir. 2012)
`
`(citing In re Keller, 642 F.2d 413, 425 (C.C.P.A. 1981)). Thus, “[a] finding of
`
`obviousness … cannot be overcome ‘by attacking references individually where
`
`the rejection is based upon the teachings of a combination of references.’”
`
`Bradium Techs. LLC v. Iancu, 923 F.3d 1032, 1050 (Fed. Cir. 2019) (quoting In re
`
`Merck & Co., 800 F.2d 1091, 1097 (Fed. Cir. 1986)) (emphasis added).
`
`The Board already noted that “Patent Owner addresses each of the
`
`references individually, without addressing what Petitioner’s combination would
`
`
`
`
`
`
`-9-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`have suggested to a” POSITA. IPR2018-01555 Decision at 39 (emphasis added);
`
`see also IPR2018-01581 Decision at 35. This criticism applies equally to INVT’s
`
`Response, which largely fails to consider the references in combination.
`
`For example, INVT spends pages arguing that Li does not teach “modulation
`
`parameters and coding parameters per subband group.” Response at 33-37.
`
`Petitioners, however, rely upon Vijayan for this limitation, not Li. See, e.g.,
`
`IPR2018-01555 Petition, 33-39; Ex. 1016 ¶¶102-104.
`
`Likewise, INVT argues that Vijayan does not teach modulation and coding
`
`parameters “based on a result of the channel estimation per subband.” Response at
`
`37-39. Petitioners, however, rely upon Li to teach this limitation, not Vijayan.
`
`See, e.g., IPR2018-01555 Petition, 30-33; Ex. 1016 ¶¶108-113.
`
`Finally, INVT argues that Hashem does not teach “joint modulation and
`
`coding parameters on a ‘per subband group’ basis based on a result of channel
`
`estimation per subband.” Response at 39-40. Petitioners, however, rely upon Li
`
`and Vijayan to teach this limitation, not Hashem. Ex. 1016 ¶¶105-107.
`
`Thus, INVT improperly addresses Li, Vijayan, and Hashem individually,
`
`rather than in combination.
`
`2.
`
`In combination, Li in view of Vijayan and Hashem teaches
`the claimed “parameter deciding section.”
`
`As discussed in the Petitions, Li discloses a subscriber with a parameter
`
`
`
`
`
`
`-10-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`deciding section (block 303 or block 405) that decides modulation and coding
`
`parameters based on a channel estimation per subband (cluster). See, e.g.,
`
`IPR2018-01555 Petition, 30-33; Ex. 1016 ¶¶108-113. INVT does not appear to
`
`dispute this. Ex. 1017, 14:6-17, 16:18-17:3, 17:17-18:7.
`
`INVT does not appear to dispute that, as established in the Petitions, Vijayan
`
`teaches deciding a joint modulation and coding parameter for a claimed subband
`
`group. Ex. 1017, 19:18-27:8, particularly 27:4-8.
`
`As discussed in the Petitions, Hashem teaches reducing overhead by having
`
`the subscriber calculate the modulation and coding parameters (Link Mode) based
`
`on measured SINR and transmit an index for the Link Mode to the base station
`
`(“BS”). Ex. 1005, 2:52-54, 6:50-53, 7:1-7; see, e.g., IPR2018-01555 Petition, 39-
`
`41; Ex. 1016 ¶¶105-107. INVT does not appear to dispute this. Ex. 1017, 89:12-
`
`17.
`
`Thus, the combination of Li, Vijayan, and Hashem teaches the claimed
`
`parameter deciding section. Ex. 1016 ¶114.
`
`
`
`B.
`
`Li in View of Vijayan Teaches the Claimed “Pattern Storage
`Section”
`
`1.
`
`Li’s “group index” means that patterns for selecting
`subbands are necessarily stored in advance.
`
`With respect to the “pattern storage section,” INVT asserts that Li and
`
`Vijayan merely disclose generic storage elements. Response at 42. INVT’s
`
`
`
`
`
`
`-11-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`assertion is incorrect. INVT fatally overlooks Li’s “group index.” Ex. 1003, 12:9-
`
`23 & Fig. 7. Using a group index to identify patterns for selecting subbands means
`
`that the patterns are necessarily stored in advance at the subscriber and BS. Ex.
`
`1017, 74:12-17; 75:19-23; 96:23-97:2; Ex. 1016 ¶127.
`
`A group index works only if the subscriber and the BS store a table of the
`
`group indices and their corresponding subband patterns in advance; otherwise, the
`
`group index is meaningless. Ex. 1017, 72:21-73:6; Ex. 1016 ¶127. In Li, the
`
`subscriber selects the subband (cluster) group having the best overall performance
`
`and then feeds the subband (cluster) SINRs in that group back to the BS. Ex. 1003,
`
`12:9-11; Ex. 1017, 74:3-6; Ex. 1016 ¶127. Because the subscriber and the BS
`
`know in advance which subbands constitute which groups, “only the group
`
`index…needs to be transmitted.” Ex. 1003, 12:14-16; see also id., 2:64-3:4 (the
`
`base station merely “receiv[es] an indication of a selection by the subscriber of one
`
`or more [cluster] groups,” indicating the pre-defined patterns for the cluster groups
`
`are known in advance); Ex. 1016 ¶127. INVT’s expert admitted as much. Ex.
`
`1017, 74:12-17.
`
`Because the subscriber no longer needs to send feedback indicating which
`
`subbands constitute which subband groups, the index “reduc[es] the data bits for
`
`cluster indexing, thereby reducing the bandwidth requirements…for cluster
`
`allocation.” Ex. 1003, 11:62-66; Ex. 1017, 7:9-19; Ex. 1016 ¶127.
`
`
`
`
`
`
`-12-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`If only the index is transmitted (and not data indicating which subbands
`
`constitute which subband groups), then the patterns must necessarily be stored in
`
`advance at both the subscriber and the BS. Ex. 1016 ¶128. INVT wholly
`
`overlooks this point in its Response. INVT’s expert even admitted that a POSITA
`
`would have understood that in at least some cases Li’s patterns have to be stored in
`
`advance. Ex. 1017, 77:25-78:10.
`
`The above should end the inquiry about whether Li teaches the claimed
`
`pattern storage section. Nevertheless, INVT spends pages arguing that the
`
`“generic” storage elements in Li and Vijayan do not establish the pattern storage
`
`section. Response at 41-45. INVT ignores the obvious—a “storage medium” was
`
`a standard part of communication devices (including subscribers and BSs) before
`
`the ’439 Patent and would have been used to perform the operations of the devices,
`
`such as Li’s subband group selection. Ex. 1016 ¶133.
`
`INVT overlooks that Li’s storage medium would have stored the group
`
`indices and their corresponding patterns. Ex. 1016 ¶134. Otherwise, the indices
`
`would have been meaningless. Ex. 1003, 12:9-23, Fig. 7; Ex. 1016 ¶134. INVT
`
`questions “why or how” Li’s storage medium stores patterns for selecting subbands
`
`in advance, Response at 42, but ignores the above facts. For the BS to allocate
`
`clusters to the subscriber for communication based upon a group index and cluster
`
`SINRs from the subscriber, both the BS and the subscriber would need to store the
`
`
`
`
`
`
`-13-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`patterns in advance so that they both know which clusters belong to which group.
`
`Ex. 1016 ¶134. The obvious place to store the patterns would have been Li’s
`
`storage medium at both the subscriber and the BS. Id.
`
`Finally, INVT complains that certain paragraphs in Dr. Ding’s initial
`
`declaration (Ex. 1007) are nearly identical to statements in the Petitions. Response
`
`at 45. This complaint is legally deficient. The Board already noted that the issue
`
`is not whether the declaration is the same as what is asserted in the Petition, but
`
`whether Dr. Ding’s opinions are well-reasoned and supported. IPR2018-01555
`
`Decision at 38. Dr. Ding’s opinions are well reasoned and supported, and nothing
`
`in the Response suggests otherwise.
`
`2.
`
`Li’s Figure 6 teaches “patterns for selecting subbands”
`under the Board’s or INVT’s construction.
`
`Under the Board’s construction of “patterns for selecting subbands,” Figure
`
`6 of Li plainly teaches stored patterns for assigning subbands to groups. Ex. 1016
`
`¶120. The section above addresses the “stored” aspect. Figure 6 of Li confirms
`
`the “patterns” aspect—every fourth subband is assigned to a cluster/subband group
`
`(Groups 1-4). Ex. 1003, Fig. 6; Ex. 1016 ¶120.
`
`
`
`
`
`
`-14-
`
`
`
`
`
`

`

`Petitioners’ Consolidated Reply
`(IPR2018-01555 and IPR2018-01581)
`
`
`
`
`Even under INVT’s construction, Figure 6 teaches “fixed rules for choosing
`
`subbands based on frequency”—every fourth subband is chosen for a group. Ex.
`
`1003, Fig. 6; Ex. 1016 ¶121. As discussed in the Petitions, Figure 6 shows that the
`
`cluster (su