`
`––––––––––
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`––––––––––
`
`HULU, LLC,
`AMAZON.COM, INC., and
`NETFLIX, INC.,
`Petitioners,
`
`v.
`
`REALTIME ADAPTIVE STREAMING LLC,
`Patent Owner.
`
`––––––––––
`
`
`Case No. IPR2018-01187
`
`Patent 9,769,477
`
`––––––––––
`
`
`
`PETITION FOR INTER PARTES REVIEW
`
`OF U.S. PATENT NO. 9,769,477
`
`
`
`
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`
`
`
`
`
`
`
`
`
`
`
`TABLE OF CONTENTS
`Introduction ...................................................................................................... 1
`Grounds for Standing ....................................................................................... 1
`Identification of Challenge .............................................................................. 2
`Priority ................................................................................................... 2
`
`Relied Upon Prior Art ........................................................................... 2
`
`Statutory Grounds.................................................................................. 3
`
`’477 Patent ....................................................................................................... 3
`Claim Construction .......................................................................................... 6
`Level of Ordinary Skill in the Art ......................................................... 6
`
`Claim Terms .......................................................................................... 7
`
` “asymmetric data compression encoder[s]” ..................................... 7
` “data blocks” .................................................................................... 8
` Prior Art ......................................................................................................... 10
` Overview of Imai (Ex. 1005) .............................................................. 10
`Overview of Pauls (Ex. 1007) ............................................................. 13
`
`Overview of Chao (Ex. 1016) ............................................................. 15
`
` Challenged Claims ......................................................................................... 15
` Ground 1: Claims 1, 3–5, and 12–14 are Obvious in View of Imai ... 15
` Independent Claim 1 is Obvious .................................................... 15
` Dependent Claims 3 and 4 are Obvious ......................................... 28
` Dependent Claim 5 is Obvious ...................................................... 30
` Dependent Claim 12 is Obvious .................................................... 33
` Dependent Claims 13 and 14 are Obvious ..................................... 34
`Ground 2: Claims 1, 3–6, and 9–14 are Obvious in View of Pauls .... 35
` Independent Claim 1 is Obvious .................................................... 35
` Dependent Claims 3 and 4 are Obvious ......................................... 43
` Dependent Claim 5 is Obvious ...................................................... 45
`
`
`
`
`
`- i -
`
`
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
` Dependent Claim 6 is Obvious ...................................................... 46
` Dependent Claim 9 is Obvious ...................................................... 46
` Dependent Claims 10 and 11 are Obvious ..................................... 48
` Dependent Claim 12 is Obvious .................................................... 50
` Dependent Claims 13 and 14 are Obvious ..................................... 51
`Ground 3: Claims 1, 3–6, 9–14 are Obvious in View of Imai and
`Pauls .................................................................................................... 52
` Motivation to Combine Imai and Pauls ......................................... 52
` Independent Claim 1 and Dependent Claims 6 and 9 are
`Obvious .......................................................................................... 54
` Dependent Claims 3 and 4 are Obvious ......................................... 55
` Dependent Claim 5 is Obvious ...................................................... 56
` Dependent Claim 6 is Obvious ...................................................... 57
` Dependent Claim 9 is Obvious ...................................................... 57
` Dependent Claims 10 and 11 are Obvious ..................................... 58
` Dependent Claim 12 is Obvious .................................................... 58
` Dependent Claims 13 and 14 are Obvious ..................................... 59
` Ground 4: Claims 2, 11, 20–22, and 25–27 of the ’477 Patent Are
`Rendered Obvious by Imai in view of Pauls and Chao ...................... 59
` Arithmetic Coding .......................................................................... 59
` Independent Claim 20 and Dependent Claim 2 ............................. 61
` Dependent Claims 10 and 11 ......................................................... 63
` Dependent Claims 21–22 and 25–27 are Obvious ......................... 64
` Conclusion ..................................................................................................... 66
` Mandatory Notices and Fees ......................................................................... 67
`
`
`
`
`
`- ii -
`
`
`
`EXHIBIT LIST
`
`Exhibit No.
`
`Description
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`
`1001
`1002
`1003
`1004
`
`1005
`1006
`1007
`
`1008
`
`1009
`
`1010
`
`1011
`
`1012
`
`1013
`
`1014
`
`1015
`1016
`
`1017
`
`1018
`
`U.S. Patent No. 9,769,477 to Fallon et al. (“the ’477 Patent”)
`Prosecution File History for the ’477 Patent
`Declaration of Dr. James A. Storer
`Japanese Patent Application Publication No. H11331305 to Imai
`et al. (“Imai”).
`Certified English Translation of Imai
`U.S. Patent No. 6,507,611 to Imai et al. (“Imai ’611”)
`European Patent Application Publication No. EP0905939A2 to
`Pauls et al. (“Pauls”)
`Excerpt from William Pennebaker et al., JPEG Still Image Data
`Compression Standard (Van Nostrand Reinhold, 1993)
`Andreas Spanias et al., Audio Signal Processing and Coding
`(John Wiley & Sons, Inc., 2007)
`Raymond Westwater et al., Real-Time Video Compression
`Techniques and Algorithms (Kluwer Academic Publishers, 1997)
`David Salomon, A Guide to Data Compression Methods
`(Springer-Verlag New York, Inc., 2002)
`Le Gall, MPEG: A Video Compression Standard for Multimedia
`Applications (April 1991)
`Memorandum Opinion and Order, Realtime Data, LLC v.
`Rackspace US, Inc. et al., No. 6:16-CV-00961, Dkt. 183
`(E.D. Tex. June 14, 2017)
`Memorandum Opinion and Order, Realtime Data, LLC v. Actian
`Corp. et al., No. 6:15-CV-00463, Dkt. 362 (E.D. Tex. July 28,
`2016)
`U.S. Patent No. 5,873,065 to Akagiri et al.
`International PCT Patent Application Publication No. WO
`98/40842 to Chao et al (“Chao”)
`Notice of Interested Parties, Realtime Adaptive Streaming, LLC
`v. Hulu LLC, No. 2:17-CV-07611, Dkt. 18 (C.D. Cal. October
`24, 2017)
`Mark Nelson, The Data Compression Book, M&T Books, 1991
`(“Nelson”)
`
`
`
`- iii -
`
`
`
`Exhibit No.
`
`Description
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`
`1019
`
`1020
`
`1021
`
`1022
`1023
`1024
`
`for video content,
`J. Golston, Comparing media codecs
`Embedded Systems Conference, San Francisco, 2004.
`International Telecommunication Union Telecommunication
`Standardization Sector Recommendation H.263 (February 1998)
`(H.263 Standard)
`U.S. Patent No. 6,195,024 to Fallon (incorporated by reference
`into the ’610 Patent)
`International PCT Application Publication WO 00/51243 to Park
`Declaration of Dr. Sylvia Hall-Ellis
`Report and Recommendation of U.S. Magistrate Judge, Realtime
`Data, LLC v. Packeteer, Inc., Case No. 6:08-CV-144, Dkt. 379
`(E.D. Tex. June 23, 2009)
`
`
`
`- iv -
`
`
`
`INTRODUCTION
`Petitioners Amazon.com, Inc., Hulu, LLC and Netflix, Inc. request inter
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`
`
`
`partes review of Claims 1–6, 9–14, 20–22, and 25–27 of U.S. Patent No. 9,769,477
`
`(Ex. 1001). The ’477 Patent claims known compression concepts, such as
`
`selecting data compression encoders based on a throughput of a communications
`
`channel and utilizing asymmetric data compression encoders.
`
`Prior art, including the Imai and Pauls references, taught selecting data
`
`compression encoders based on throughput no later than 1999, well before the ’477
`
`Patent was filed in 2001. Imai’s data transmission system selects asymmetric
`
`compression algorithms based on a determined throughput of the transmission
`
`channel. Pauls’ adaptive communication data formatting system also accounts for
`
`the nature and speed of the network in selecting from a variety of asymmetric
`
`video transcoders. Both references teach selecting asymmetric compression
`
`algorithms for use. Moreover, arithmetic encoders, like those taught by the Chao
`
`reference, were known in the art. The challenged claims are invalid in light of
`
`Imai, Pauls and/or in combination with Chao.
`
` GROUNDS FOR STANDING
`Petitioners certify that the ’477 Patent is eligible for inter partes review.
`
`Petitioners further certify that they are not barred or estopped from requesting this
`
`inter partes review.
`
`
`
`1
`
`
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`
`IDENTIFICATION OF CHALLENGE
`
`Priority
`The application that led to ’477 Patent was filed October 6, 2015 as U.S.
`
`Patent Application No. 14/876,276 and claims priority to U.S. Provisional Patent
`
`Application No. 60/268,394, which was filed February 13, 2001. Petitioners are
`
`not aware of any claim by the Patent Owner that the ’477 Patent is entitled to a
`
`priority date earlier than February 13, 2001.
`
` Relied Upon Prior Art1
`Exhibit 1004 – Japanese Patent Application Publication No. H11331305
`
`(“Imai”) is prior art under at least pre-AIA §§102(a) and (b) because it published
`
`November 30, 1999, which is over one year before the ’477 Patent’s earliest
`
`priority date. See §VI.A. (Exhibit 1005 – certified English translation of Imai);
`
`(Exhibit 1006 – U.S. Patent No. 6,507,611 (“the Imai ’611 Patent”) is the U.S.
`
`sibling of Imai). 2
`
`Exhibit 1007 – U.S. Patent No. 6,920,150 (“Pauls”) is prior art under at least
`
`pre-AIA §102 (e). See §VI.B.
`
`
`1 Because each claim of the ’477 Patent claims priority to an application filed
`before March 16, 2013, pre-AIA 35 U.S.C. §102 applies. MPEP §2159.02.
`2 The Imai ’611 Patent (Ex. 1006) claims priority to Imai and contains
`substantively identical figures and disclosures.
`
`
`
`2
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`Exhibit 1016 – International Patent Application Publication No. WO
`
`98/40842 (“Chao”) is prior art under 35 U.S.C. §§102(a) and (b). See §VI.C.
`
`Imai, Pauls, and Chao were neither cited to nor considered during
`
`prosecution of the ’477 Patent. Ex. 1001; see generally, Ex. 1002.
`
`
`Statutory Grounds
`Petitioners request inter partes review on the following grounds:
`
`Ground
`No.
`1
`2
`
`3
`
`4
`
`
`
`References
`
`Statutory Basis
`
`Claims Challenged
`
`Imai (Ex. 1005)
`Pauls (Ex. 1007)
`Imai (Ex. 1005)
`Pauls (Ex. 1007)
`Imai (Ex. 1005)
`Pauls (Ex. 1007)
`Chao (Ex. 1016)
`
`Obviousness (§103)
`Obviousness (§103)
`
`1, 3–5, 12–14
`1, 3–6, 9–14
`
`Obviousness (§103)
`
`1, 3–6, 9–14
`
`Obviousness (§103)
`
`2, 11, 20–22, 25–27
`
`’477 PATENT
`The ’477 Patent is directed to “compressing and decompressing data based
`
`on the actual or expected throughput (bandwidth) of a system employing data
`
`compression.” Ex. 1001 at 9:27–31. The ’477 Patent states that “dynamic
`
`modification of compression system parameters so as to provide an optimal
`
`balance between execution speed of the algorithm (compression rate) and the
`
`resulting compression ratio, is highly desirable.” Id. at 1:64–67.
`
`
`
`3
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`The ’477 Patent purports to solve “bottlenecks” in the throughput of a
`
`system by switching between different compression algorithms applied to data.
`
`Ex. 1001 at 10:3–8. The ’477 Patent notes that asymmetric algorithms provide “a
`
`high compression ratio (to effectively increase the storage capacity of the hard
`
`disk) and fast data access (to effectively increase the retrieval rate from the hard
`
`disk).” Ex. 1001 at 13:39–45. On the other hand, symmetric routines “compris[e]
`
`a fast compression routine.” Ex. 1001 at 14:40–43. In one embodiment, a
`
`controller “tracks and monitors the throughput . . . of the data compression system
`
`12.” Ex. 1001 at 10:54–57. When the throughput of the system falls below a
`
`predetermined threshold, the system generates control signals to enable/disable
`
`different compression algorithms. Ex. 1001 at 10:55–58.
`
`The ’477 Patent describes different “popular compression techniques” that
`
`were known in the prior art. Ex. 1001 at 5:11. Specifically, the ’477 Patent admits
`
`that arithmetic compression was known and that arithmetic coding was a “popular
`
`compression technique [that] possesses the highest degree of algorithmic
`
`effectiveness.” Ex. 1001 at 5:11–12.
`
`The ’477 Patent also admits that content-dependent compression, as recited
`
`in claims 10 and 11, was well-known in the prior art. The ’477 Patent incorporates
`
`into its specification the disclosure of US Patent No. 6,195,024 (“the ’024
`
`Patent”)—a patent with an application filing date in 1998. Ex. 1001 at 5:45–50;
`
`
`
`4
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`Ex. 1021 Face. The ’024 Patent admits that “there are many conventional content
`
`dependent” compression techniques in the prior art and even presents the below
`
`“diagram of a content dependent high-speed lossless data compression and
`
`decompression system/method according to the prior art.” Ex. 1021 at 5:54–56,
`
`2:41–45.
`
`
`
`Here, the ’024 Patent describes a system that identifies an input data type and
`
`compresses the data in accordance with the identified data type. Thus, the ’024
`
`Patent, as incorporated by reference into the disclosure of the ’477 Patent, admits
`
`that (a) content-dependent compression amounts to merely compressing data with
`
`
`
`5
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`an algorithm based on a data type, and (b) content-dependent compression was
`
`well-known in the prior art.
`
`Petitioners are not aware of any objective evidence of non-obviousness for
`
`the challenged claims of the ’477 Patent.
`
` CLAIM CONSTRUCTION
`To the extent the Applicant has defined a claim term in the specification,
`
`Petitioners have used that definition. For the purpose of deciding the grounds of
`
`invalidity presented by this petition,3 the following terms should be construed:
`
`“asymmetric data compression encoder[s]” and “data blocks.”
`
` Level of Ordinary Skill in the Art
`A person of ordinary skill in the art (“POSITA”) as of February 13, 2001
`
`would have a bachelor’s degree in electrical engineering, computer science, or a
`
`similar field with at least two years of experience in data compression or a person
`
`with a master’s degree in electrical engineering, computer science, or a similar
`
`field with a specialization in data compression. Ex. 1003 at 65. A person with less
`
`education but more relevant practical experience may also meet this standard. Ex.
`
`1003 at 65.
`
`3 None of the claim construction issues that are necessary to resolve the invalidity
`grounds presented by this petition differ based upon the application of broadest
`reasonable
`interpretation versus
`the district court-type claim construction
`standards. Ex. 1003 at 88.
`
`
`
`6
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`
` Claim Terms
`
`“asymmetric data compression encoder[s]”
`A POSITA would have understood that the term “asymmetric data
`
`compression encoder(s)” means “an encoder(s) configured to utilize a compression
`
`algorithm in which the execution time for the compression and decompression
`
`routines differ significantly” in view of the specification. See Ex. 1003 at 89-90.
`
`Although the ’477 Patent does not describe an “asymmetric data compression
`
`encoder,” it provides an express definition for an “asymmetrical data compression
`
`algorithm.” The ’477 Patent states “[a]n asymmetrical data compression algorithm
`
`is referred to herein as one in which the execution time for the compression and
`
`decompression routines differ significantly.” See, e.g., Ex. 1001 at 10:12–23.
`
`Moreover, the specification gives examples of asymmetric and symmetric
`
`algorithms, stating that “dictionary-based compression schemes such as Lempel-
`
`Ziv” are asymmetric, while “table-based compression schemes such as Huffman”
`
`are symmetric. Ex. 1001 at 10:19–20 & 10:24–25.
`
`Accordingly,
`
`the Board should find
`
`that “asymmetric compression
`
`algorithm” means “a compression algorithm in which the execution time for
`
`compression and decompression differ significantly.” Ex. 1003 at 89-90.
`
`
`
`7
`
`
`
`
`“data blocks”
`A POSITA would have understood that “data block,” in the context of the
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`
`specification, means “a unit of data comprising more than one bit.” See Ex. 1003
`
`at 91-94. First, “data block” is used consistently in the claims to refer to a unit of
`
`data that is compressed by a compression algorithm. Ex. 1001 at 21:57–22:66, cls.
`
`1, 9, 19, and 25. The specification further explains that “[d]ata compression is
`
`widely used to reduce the amount of data required to process, transmit, or store a
`
`given quantity of information,” which indicates that a data block must be a unit
`
`large enough for there to be a chance to realize a reduction in size through
`
`compression Ex. 1001 at 2:52–54; Ex. 1003 at 91. The smallest unit of digital data
`
`representation is a bit, and the information contained in a single bit cannot be
`
`represented through compression with fewer bits. Ex. 1003 at 91. Therefore, a
`
`data block must be more than one bit in length so that it can be compressed as
`
`claimed. Ex. 1003 at 91.
`
`The ‘477 patent specification also supports this construction. It describes
`
`“block structured disk compression” as operating on blocks of data that are either
`
`“fixed” or “variable in size.” Ex. 1001 at 7:19–21. The specification states that
`
`data blocks can represent files, and that “[a] single file usually is comprised of
`
`multiple blocks, however, a file may be so small as to fit within a single block.”
`
`Ex. 1001 at 7:21–23. Also, the specification goes on to discuss the pros and cons
`
`
`
`8
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`of smaller and larger data block sizes. Ex. 1001 at 7:25–39. It also contemplates
`
`units of data that comprise more than one bit that are stored in its system. Ex. 1001
`
`at FIG. 4B (stating “2 bits” are reserved for a sector map “type” definition, and “3
`
`bits” are reserved for “c type”). The specification’s discussion of various data
`
`block sizes, including file-sized data blocks and data units as small as 2 bits, are
`
`consistent with Petitioner’s proposed construction.
`
`The ’477 Patent incorporates by reference U.S. Patent No. 6,195,024 (“the
`
`’024 Patent”) which uses the term “data block” in a consistent manner:
`
`It is to be understood that the system processes the input
`data streams in data blocks that may range in size from
`individual bits through complete files or collections of
`multiple files. Additionally, the data block size may be
`fixed or variable. The counter module [] counts the size
`of each input data block (i.e., the data block size is
`counted in bits, bytes, words, any convenient data
`multiple or metric.
`
`Ex. 1021 at 7:9–15. In district court proceedings, the Patent Owner4 has twice
`
`stipulated to a similar construction of this term. Ex. 1013 at 34; Ex. 1014 at 40
`
`(both evidencing Patent Owner’s agreement that “data block” means “a single unit
`
`of data, which may range in size from individual bits through complete files or
`
`4 The entity in those proceeding is Realtime Data, LLC rather than Realtime
`Adaptive Streaming LLC, the Patent Owner here. Ex. 1017.
`
`
`
`9
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`collection of multiple files”). Thus, the Patent Owner’s use of the term in litigation
`
`supports Petitioner’s construction.
`
` PRIOR ART
` Overview of Imai (Ex. 1005)
`Imai is a Japanese Patent Application Publication filed by Sony. Ex. 1005.
`
`Imai is directed to encoding digital data for transmission over a network using real-
`
`time decompression and reproduction at a client by selecting an appropriate
`
`encoder, from a plurality, based on various factors including the detected
`
`throughput of a network. Ex. 1005 at [0001], [0005], [0067]–[0068], [0100]–
`
`[0101], Solution means. After receiving a request for digital data from a client,
`
`Imai’s “frame cutting circuit” cuts the requested digital data into “units of frame”
`
`having a length that is suitable for coding or for transmission on a network. Id. at
`
`[0130], [0066]. Imai’s “units of frame” are units of data bits or digital data blocks
`
`on which Imai’s compression and transmission system operates. Ex. 1003 at 97.
`
`
`
`10
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`
`
`
`Switch 52 supplies each individual digital data “frame” output from the frame
`
`cutting circuit to a selected one of a plurality of available encoders 531 to 53N.
`
`Ex. 1005 at [0066]. Selection instructing unit 55 selects an appropriate “one from
`
`a plurality of coding methods corresponding to the encoders 531 to 53N . . . and
`
`then instructs the encoding selecting circuit 56 to select the decided coding
`
`method.” Id. at [0070]. Imai’s encoders are configured to utilize asymmetric
`
`compression algorithms including the MPEG audio layers 1, 2, and 3. Ex. 1005 at
`
`[0067]; see Ex. 1010 at 7 (stating that MPEG layer 3 is an asymmetric compression
`
`algorithm); Ex. 1003 at 98.
`
`Imai’s “selection instructing unit” analyzes various factors to decide which
`
`compression algorithm to select and apply to each individual data frame. For
`
`example, Imai teaches assessing client processing ability by analyzing the client’s
`
`processing of “dummy data packets” to determine client resources that are
`
`
`
`11
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`“employed for [] other process[es]” and resources that are available. Ex. 1005 at
`
`[0099]–[0100]. Imai’s selection instructing unit also determines characteristics of
`
`the uncompressed data and selects a compression algorithm accordingly. Id. at
`
`[0102]. The selection instructing unit accounts for these variations in selecting a
`
`suitable coding method. Id. Imai additionally describes a detailed process for
`
`deriving a transmission rate of a network communication channel by timing the
`
`transmission and receipt of data packets between the client and server. Id. at
`
`[0149]–[0150].
`
`While some of Imai’s embodiments are directed to audio coding, Imai
`
`explains that its “invention is also applicable to other signals such as video signals,
`
`other types of time series signals . . . .” Ex. 1005 at [0172]. Thus, Imai’s teachings
`
`that are not limited to audio, but apply more generally to selecting and applying
`
`various encoders based on specific data parameters, regardless of the target data set
`
`for each underlying algorithm (e.g., audio or voice). It would have been well-
`
`known to a POSITA at the relevant time that audio, video, and image compression
`
`techniques were related and it would have been common for a POSITA to consult
`
`and utilize teachings from these related data types, even without Imai’s explicit
`
`suggestion to do so. Ex. 1003 at 147.
`
`
`
`12
`
`
`
` Overview of Pauls (Ex. 1007)
`Pauls is a European Patent Application Publication that published in 1999
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`
`and was assigned to Lucent Technologies. Pauls is directed to “improving data
`
`transfer performance over communications networks connecting data networks and
`
`users using adaptive communications formatting.” Ex. 1007 at Abstract. Pauls
`
`explains
`
`that adaptive communications formatting
`
`involves “encoding (or
`
`compressing)” data to “reduce the amount of data being transmitted” using
`
`“transcoding techniques.” Id. at [0003]. Like Imai, Pauls teaches selecting one
`
`from a plurality of encoders based on various data parameters, such as the
`
`throughput of a communication channel. Pauls has extensive teachings on
`
`selecting between different asymmetric video encoders. Id. at [0009]–[0010],
`
`[0012], FIG. 3.
`
`
`
`13
`
`
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`Pauls teaches that the particular transcoders applied are selected based upon
`
`factors such as the “nature of the communications network,” the type of data being
`
`transmitted, and the preferences of the user. See id.at [0003]. For the nature of the
`
`communications network, Pauls teaches a relevant factor is “the available
`
`bandwidth” and the “bit rate” of the network. Id. at [0013].
`
`Pauls teaches that a system may have more than one video/image transcoder.
`
`See FIG. 3. Pauls explains that different transcoders are more effective than others
`
`for particular data types. Id. at 0017. For example, Pauls teaches that the input
`
`video may be in an MPEG or MPEG2 format, and that H.263 may be an effective
`
`transcoding technique. Id. at [0017], [0024], & FIG. 5. Furthermore, H.263 had
`
`the option of forming the compressed video data using an arithmetic compression
`
`algorithm. Ex. 1020 at 69–76.
`
`In one embodiment, Paul teaches that “[t]he communication network 16
`
`connects the user 14 to the access server 20 . . . [and u]pon connecting to the
`
`access server 20, the user 14 can retrieve data from the host 22.” Ex. 1007 at
`
`[0006]. Pauls teaches “[t]he data (or file) is retrieved via a bitstream from the host
`
`22 to the access server 20 to the user 14.” Id. at [0008]. Pauls teaches that “[a]t
`
`the access server 20, the data is formatted using a mixture of transcoding
`
`techniques and error control schemes to facilitate data transmission within
`
`acceptable quality levels.” Id. at [0008].
`
`
`
`14
`
`
`
` Overview of Chao (Ex. 1016)
`Chao is an International Patent Application that published in 1998. Chao
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`
`teaches a compression system for compressing video or image data employed in
`
`“‘video on demand’ systems, such as video servers” for “streaming video” in “real-
`
`time.” Ex. 1016 at 4:29–35. Chao’s video data is compressed using an entropy
`
`coding technique of either “arithmetic, run length, or Huffman encoding.” Id. at
`
`7:5–8. Chao also teaches having a particular entropy compression algorithm
`
`applied where it is “selected at run-time by the user, based on the desired
`
`compression ratio and the amount of time required to get the selected level of
`
`compression.” Id. at 14:19–25.
`
` CHALLENGED CLAIMS
` Ground 1: Claims 1, 3–5, and 12–14 are Obvious in View of Imai
`A POSITA would have found Claims 1, 3–5, and 12–14 are obvious in view
`
`of her own knowledge and the teachings of Imai. Ex. 1003 at 111.
`
`
`Independent Claim 1 is Obvious
`Claim 1 includes a preamble directed to “[a] system.” To the extent the
`
`preamble is a limitation, it is met because Imai teaches “a transmission system” for
`
`compressing data. Ex. 1005 at [0050], [0051]; Ex. 1003 at 112.
`
`
`
`15
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`
`1[a] a plurality of different asymmetric data compression encoders, wherein
`each asymmetric data compression encoder of the plurality of different
`asymmetric data compression encoders is configured to utilize one or
`more data compression algorithms,
`
`Imai teaches this limitation. Imai teaches “a plurality of coding methods
`
`corresponding to the encoders 531 to 53N.” Ex. 1005 at [0070]. Imai’s encoders
`
`531 to 53N employ “different coding methods from each other” and are thus
`
`different encoders. Ex. 1005 at [0067]; Ex. 1003 at 113.
`
`Ex 1005 at [0067]. Imai teaches that the coding methods employed by each
`
`encoder include:
`
`
`
`(Pulse Code Modulation), ADPCM
`linear PCM
`(Adaptive Differential PCM), layers 1, 2, 3 of MPEG
`(Moving Picture Experts Group), ATRAC (Adaptive
`Transform Acoustic Coding), ATRAC 2, and HVXC
`(Harmonic Vector Excitation Coding). Stated otherwise,
`
`
`
`16
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`
`in the embodiment, the encoders 531 to 53N are prepared
`by using encoders which perform encoding of the audio
`signal with various coding methods.
`
`Ex. 1005 at [0067], [0068]–[0071]. At least MPEG layers 1, 2, and 3, ATRAC,
`
`and ATRAC 2 are each data compression algorithms. Ex. 1003 at 114; see Ex.
`
`1015 2:25–34 (describing ATRAC compression); Ex. 1022 at 4:19–21 (describing
`
`audio MPEG layers 1, 2, and 3 compression); Ex. 1005 at [0068] (stating that
`
`MPEG layer 3 and ATRAC 2 provide a high compression rate). A POSITA would
`
`have understood that Imai’s “encoders” are data compression encoders that employ
`
`or use different data compression methods or algorithms because these algorithms
`
`use less bits to represent the input data stream. Ex. 1003 at 115-116.
`
`Imai’s encoders comprise a plurality of different asymmetric data
`
`compression encoders that utilize data compression algorithms. Ex. 1005 at
`
`[0067]; Ex. 1003 at 117-118. Specifically, MPEG layers 1, 2, and 3, and the
`
`ATRAC and ATRAC 2 compression algorithms are each different asymmetric data
`
`compression algorithms that are each used by Imai’s encoders. Ex. 1003 at 117.
`
`The ATRAC family of compression algorithms, including the ATRAC 2
`
`compression algorithm, use an asymmetric architecture that separates the
`
`implementation details of the encoder (such as psychoacoustic analysis and bit-
`
`allocation) from the decoder. Ex. 1003 118; Ex. 1009 at 81. This architecture
`
`
`
`17
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`allows for more sophisticated encoding strategies which are performed with access
`
`to higher performance computing resources (e.g., a computing device or special
`
`purpose recording/encoding device) without modifying the complexity of the
`
`decoding strategies, which are typically performed with less sophisticated
`
`implementations. Ex. 1003 at 118; Ex. 1009 at 81. These aspects of asymmetric
`
`algorithms (including the ATRAC compression algorithm) are advantageous for
`
`and contributes to their use in inexpensive decoder implementations such as the
`
`Sony MiniDisc player. Ex. 1003 at 119; Ex. 1009 at 80–82.
`
`
`
`
`
`ATRAC 2 is also built on this asymmetric architecture and is therefore also
`
`an asymmetric data compression algorithm. Ex. 1003 at 120. The MPEG family
`
`of audio compression algorithms, including MPEG layers 1, 2, and 3, is built on a
`
`similar architecture where the compressor executes a slow, complex algorithm
`
`whereas the decompression algorithm is simple. Ex. 1003 at 120; Ex. 1009 at 81;
`
`Ex. 1011 at 11; see Ex. 1010 at 7. This explains why MPEG layer 3 is “a natural
`
`choice when audio files are compressed into an archive, where they will be
`
`
`
`18
`
`
`
`IPR2018-01187 Petition
`U.S. Patent No. 9,769,477
`
`decompressed and used very often.” Ex. 1003 at 120-121; Ex. 1011 at 11. Thus,
`
`Imai’s disclosure of a configuration of a plurality of different coding methods is a
`
`plurality of different asymmetric data compression algorithms. Ex. 1005 at [0067];
`
`Ex. 1003 at 121. With this disclosure, a POSITA would have understood that each
`
`of the ATRAC and MPEG compression algorithms are asymmetric compression
`
`algorithms. Ex. 1003 at 118-120; Ex. 1009 at 81; Ex. 1010 at 7; Ex. 1011 at 11.
`
`Imai explains that its invention is equally applicable to video. Ex. 1005 at
`
`[0172]. A POSITA would have known to configure Imai’s compression and
`
`transmission teachings with a plurality of different encoders configured to utilize
`
`one or more asymmetric data compression algorithms because video is an
`
`“asymmetric application” that realizes the same benefits from compression with
`
`asymmetric encoders and algorithms as other media, such as audio. Ex. 1003 at
`
`122-123; Ex. 1012 at 5. For instance, the MPEG family of audio compression
`
`algorithms, discussed above, uses a slow, complex algorithm for compression and
`
`a simpler algorithm for decompressi