`REALTIME ADAPTIVE
`STREAMING LLC,
`
`
`
`
`
`GOOGLE LLC, and YOUTUBE LLC,
`
`
`
`
`Plaintiff,
`
`vs.
`
` Defendants.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`UNITED STATES DISTRICT COURT
`CENTRAL DISTRICT OF CALIFORNIA, WESTERN DIVISION
`
`
`Case No. 2:18-CV-03629-GW-JC
`
`
`
`
`
`
`
`EXPERT REPORT OF DR. IAIN E. RICHARDSON REGARDING INVALIDITY OF
`ASSERTED PATENTS
`
`
`
`
`
`
`
`
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 1
`
`
`
`(a)
`
`(b)
`
`There is no clear dividing line between “asymmetric” and
`“symmetric” compression algorithms ............................................30
`The nature of the systems on which compression and
`decompression are performed can have a significant impact
`on relative timing ...........................................................................33
`“slow compress encoder” / “fast decompress decoder” .............................35
`2.
`Invalidity in View of Prior Art ...............................................................................37
`1.
`Pauls ...........................................................................................................40
`2.
`Brooks ........................................................................................................42
`3.
`Ishii ............................................................................................................43
`4.
`Dye .............................................................................................................47
`5.
`Imai ............................................................................................................49
`6.
`Couwenhoven ............................................................................................52
`7.
`Rynderman .................................................................................................56
`8.
`Beyda .........................................................................................................59
`9.
`Darwin System ...........................................................................................61
`10. Microsoft NetMeeting 3.0 ..........................................................................67
`11.
`RealSystem G2 ...........................................................................................69
`The Asserted Patents Are Directed To An Abstract Idea ......................................74
`D.
`CONCLUSION ..................................................................................................................81
`
`C.
`
`IX.
`
`
`
`
`ii
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 2
`
`
`
`1.
`
`I have been engaged by counsel for Google LLC and YouTube, LLC (collectively,
`
`“Google”) to opine on the validity of the claims of U.S. Patent Nos. 7,386,046 (“the ’046
`
`patent”), 8,934,535 (“the ’535 patent”), and 9,769,477 (“the ’477 patent”) (collectively, the
`
`“Asserted Patents”) that are being asserted by Realtime Adaptive Streaming LLC (“Realtime”) in
`
`this litigation.
`
`I.
`
`QUALIFICATIONS
`
`2.
`
`I received a Masters of Engineering (M.Eng.) degree in Electronic and Electrical
`
`Engineering from Heriot-Watt University in Edinburgh, Scotland in 1990. I received a Doctor of
`
`Philosophy (Ph.D.) degree in Video Compression from Robert Gordon University in Aberdeen,
`
`Scotland in 1999. I worked as a Digital Signal Processing (“DSP”) Hardware Designer with
`
`GEC Avionics Ltd. from 1990 to 1993.
`
`3.
`
`In 1993, I assumed a post as a Lecturer, then Reader, and eventually a Full
`
`Professor in the field of image and video compression in the School of Engineering at Robert
`
`Gordon University. In 2009, I was honored as an Honorary Professor with the Robert Gordon
`
`University, a position I maintain to this day.
`
`4.
`
`I am the author of four books and over 70 journal and conference papers on video
`
`and image coding and communication, including two widely cited books on the H.264 / MPEG-4
`
`industry standards for video compression.
`
`5.
`
`At Robert Gordon University, I founded and ran an image communication
`
`technology research laboratory. I carried out original research in the field of data, image, and
`
`video compression, initiated and managed research projects, and supervised research students.
`
`6.
`
`A particular focus of my research has been the analysis and optimization of video
`
`compression computational complexity. For example, I developed novel approaches to adapting
`
`
`
`
`1
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 3
`
`
`
`10.
`
`In 2009, I founded Onecodec Limited and served as CEO, leading the company’s
`
`development of innovative video, image, and data compression and storage software and
`
`systems. Onecodec’s technology and business activities were merged with Vcodex Limited in
`
`2015.
`
`11.
`
`I have previously prepared expert reports and testified in a number of cases
`
`concerning video and image compression, graphics compression, general-purpose data
`
`compression, and video and image communications.
`
`12. A copy of my curriculum vitae, including all publications authored in the previous
`
`10 years and all cases in which I have testified at trial or by deposition during the previous four
`
`years, is attached as Exhibit 1 to this Report. I am being compensated for my work at a rate of
`
`$650 per hour.
`
`II.
`
`SUMMARY OF OPINIONS
`
`13.
`
`In my opinion, all of the asserted claims of the Asserted Patents are invalid for
`
`multiple reasons.
`
`14. Each of the asserted claims is invalid because it fails to satisfy the written
`
`description and/or enablement requirements.
`
`15. Each of the asserted claims of the ’535 and ’477 patents is invalid because it
`
`includes an indefinite limitation.
`
`16. None of the asserted claims is entitled to claim priority to the provisional patent
`
`application filed on February 13, 2001. Thus the earliest potential priority date for any of the
`
`asserted claims is the date of filing of the original non-provisional patent application on February
`
`13, 2002.
`
`17. Selecting encoders or methods of compression based on the factors described in
`
`
`
`
`3
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 4
`
`
`
`the asserted claims was well known in the art long before the priority date of the Asserted
`
`Patents.
`
`18. Each of the asserted claims of the ’046 and ’477 patents is invalid because the
`
`claimed method or system is anticipated by one or more prior art references. To the extent that
`
`any asserted claim is found not to be anticipated, that claim is rendered obvious by a
`
`combination of prior art references with the knowledge of a person of ordinary skill in the art at
`
`the alleged time of invention and/or with other prior art references.
`
`19. Each of the asserted claims of the ’535 patent is invalid because the claimed
`
`method or system is rendered obvious by a combination of prior art references with the
`
`knowledge of a person of ordinary skill in the art at the alleged time of invention and/or with
`
`other prior art references.
`
`20. Each of the asserted claims is directed to an abstract idea implemented using
`
`routine and conventional steps or apparatus.
`
`III. MATERIALS RELIED ON
`
`21. My opinions are based on my experience as well as my review of relevant
`
`materials. In arriving at my opinions I have read and relied on the documents that are cited in
`
`this report as well as the prosecution histories of the Asserted Patents, Realtime’s Infringement
`
`Contentions, dated November 21, 2018, and Google’s Invalidity Contentions, dated January 18,
`
`2019 (as subsequently supplemented by charts addressing RealNetworks RealSystem).
`
`IV.
`
`LEGAL PRINCIPLES
`
`22.
`
`I understand that a patent is presumed to be valid, and that this presumption can be
`
`overcome only by clear and convincing evidence of invalidity.
`
`A.
`
`23.
`
`
`
`
`Anticipation and Obviousness
`
`I further understand that a patent claim is anticipated if the invention was known or
`4
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 5
`
`
`
`encoder” and “at least one fast decompress decoder” were replaced by “at least one decompress
`
`decoder.” Interpreting the final clause of claim 14 as a definition of “slow compress encoder”
`
`and “fast decompress decoder” thus deprives these terms of any separate and independent
`
`meaning.
`
`96. Moreover, even if “slow compress encoder” and “fast decompress decoder” were
`
`defined by the final clause of claim 14, that would still not resolve the uncertainty in the scope of
`
`the claim. Even in that case, a person of ordinary skill in the art reading the claim in light of the
`
`specification would be unable to determine its scope with reasonable certainty. For example, if
`
`“fast” and “slow” are simply intended to refer to the relative times for compression and
`
`decompression, then a compression method that took more time to compress data than to
`
`decompress data—however minimal the difference—would satisfy the claim limitation. But that
`
`would be inconsistent with the specification. As I previously discussed, the specification
`
`describes an asymmetric compression method as one where the difference in time between the
`
`compression and decompression routines is “substantial” (again, without defining what
`
`“substantial” means).
`
`97.
`
`In light of the above uncertainties and ambiguities, a person of ordinary skill in the
`
`art would be unable to determine, with reasonable certainty, the scope of claim 14 of the ’535
`
`patent, rendering the claim indefinite.
`
`C.
`
`Invalidity in View of Prior Art
`
`98. The purportedly novel ideas underlying the inventions claimed in the Asserted
`
`Patents are, in fact, very old and long predate the effective filing date of those patents.
`
`99. The asserted claims of the ’046 patent are directed to tracking the throughput of a
`
`data compression system by considering the number of pending requests for data transmission
`
`
`
`
`37
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 6
`
`
`
`Pauls, Fig. 5, col. 7:9-21. For example, for video data, the chart shows choosing H.263 encoding
`
`to reduce the bit rate of input MPEG or MPEG2 video streams from 1.5 Mbps or 2.0 Mbps to 8-
`
`
`
`24 Kbps.
`
`110. In my opinion, as shown in detail in Exhibit 2, Pauls discloses all of the elements
`
`of the asserted claims of the ’477 patent and hence anticipates those claims.
`
`2.
`
`Brooks
`
`111. U.S. Patent No. 7,143,432 (“Brooks”) issued from an application filed on February
`
`10, 2000, well before the effective filing date of the Asserted Patents. Brooks is particularly
`
`relevant to the asserted claims of the ’477 patent.
`
`112. Brooks discloses a system for real time video data transmission that determines
`
`which methods of compression to use based on certain relevant factors, including the bandwidth
`
`of the communication channel used to transmit the video.
`
`113. For example, over a DSL connection where “the bandwidth is relatively large,”
`
`
`
`
`42
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 7
`
`
`
`Brooks indicates that the video could be transmitted so as “to display up to a 640×480 pixel color
`
`image at 10 frames per second (fps) using an MPEG1 format.” Brooks, col. 6:32-41. On the
`
`other hand, over a modem connection, “the bandwidth is relatively small” and Brooks suggests
`
`“display[ing] up to a 160×120 color image at 5 fps using an MPEG4 format.” Id. at 6:42-51. A
`
`person of ordinary skill in the art would have recognized MPEG1 and MPEG4 as video data
`
`compression encoders that use asymmetric compression.
`
`114.
`
` Brooks also discloses other asymmetric methods of compressing video data based
`
`on properties of the input video and the desired output. For example, if the input resolution of
`
`the video is greater than the desired output resolution, Brooks discloses using “subsampling” so
`
`that one would “use every Xth pixel in every Xth line for the output image,” where X would
`
`depend on the degree of resolution reduction required. Id. at col. 17:64 – 18:7. Brooks describes
`
`similar methods for reducing the frame rate, the number of bits used to represent color (“color
`
`depth”), and the bitrate if the input frame rate, color depth, or bitrate exceeds the desired output
`
`frame rate, color depth, or bitrate. Id. at col. 17:23 – col. 19:16. In each of these cases, the video
`
`data is compressed because the output uses fewer bits than the input to represent the video.
`
`These compression methods are also asymmetric because additional processing is required by the
`
`encoder to determine what information to remove and then to remove it, while no corresponding
`
`processing needs to be done by the decoder.
`
`115. In my opinion, as shown in detail in Exhibit 3, Brooks discloses all of the elements
`
`of the asserted claims of the ’477 patent and hence anticipates those claims.
`
`3.
`
`Ishii
`
`116. U.S. Patent No. 5,675,789 (“Ishii”) issued on October 7, 1997, well over a year
`
`before the effective filing date of the Asserted Patents. Ishii is particularly relevant to the
`
`
`
`
`43
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 8
`
`
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 9
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`EXHIBIT 3
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 10
`
`
`
`Exhibit 3
`to Expert Report of Dr. Iain E. Richardson Regarding Invalidity
`
`Invalidity Chart for U.S. Patent No. 9,769,477
`U.S. Patent No. 7,143,432 (“Brooks”)
`
`Asserted Claims
`
`Corresponding Disclosure in the Prior Art Reference(s)
`
`[Claim 1 – Preamble] A
`system, comprising:
`
`[1a] 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, and
`
`Brooks discloses a system as set forth below.
`
`
`
`Brooks discloses a plurality of different asymmetric data compression encoders, including MPEG-1,
`MPEG-2, and MPEG-4. See, e.g., col. 6:24 – col. 7:17, col. 9:66 – col. 10:10, col. 14:43-48, Fig. 1. A
`person of ordinary skill in the art would have recognized that MPEG-1, MPEG-2, and MPEG-4 are each
`configured to utilize asymmetric data compression algorithms, for example to achieve different levels of
`compression.
`Brooks discloses a number of other asymmetric data compression algorithms utilized by the
`encoders, see Figs. 4, 6A, 6B, including the following:
`• Reducing the “color depth,” that is approximating the colors in a video frame using fewer
`bits than in the input frame. See, e.g., col. 17:23-42.
`• Reducing the resolution by “subsampling,” for example by using every Xth pixel in every
`Xth line, where X depends on the amount of resolution reduction desired. See, e.g., col.
`17:56 – col. 18:7.
`• Reducing the frame rate by, for example, dropping certain frames. See, e.g., col. 18:19-
`41.
`Each of these is a compression algorithm because fewer bits are used to represent the output video
`data than were used for the input video data. Each would be expected to be asymmetric by a person
`of ordinary skill in the art because, in each case, the algorithm require more significant computation
`by the encoder, for example to approximate the input data using fewer bits or to identify and remove
`certain data. No corresponding work such as adaptively determining a frame dropping rate needs to
`be done by the decoder, and no corresponding work is disclosed by Brooks. Accordingly, a person of
`
`
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 11
`
`
`
`Asserted Claims
`
`[1b] wherein a first asymmetric
`data compression encoder of
`the plurality of different
`asymmetric data compression
`encoders is configured to
`compress data blocks
`containing video or image data
`at a higher data compression
`rate than a second asymmetric
`data compression encoder of
`the plurality of different
`asymmetric data compression
`encoders; and
`
`Invalidity Chart for the ’477 Patent
`Exhibit 3 (Brooks)
`
`
`Corresponding Disclosure in the Prior Art Reference(s)
`ordinary skill in the art would expect that the processing time needed for encoding to be greater than
`the processing time needed for decoding.
`
`Brooks discloses asymmetric data compression encoders that are configured to compress data blocks
`containing video or image data at different compression rates, where a higher data compression rate
`corresponds to faster compression.
`The parties have agreed that “data block” means “a single unit of data, which may range in size from
`individual bits through complete files or collection of multiple files.” Under this definition, the video
`frames in Brooks constitute data blocks that contain video data. See, e.g., col. 11:5-17 (data may be stored
`in frame buffer as frames of video data); col. 14:35-36 (In FIG. 4, encoding block 560 next receives the
`bit-rate adjusted frames of data.”); 14:43-48 (data to be compressed can be I-frame, P-frame, or B-frame
`video data).
`
`Brooks discloses asymmetric data compression encoders with different compression rates. As discussed
`above, Brooks discloses various compression algorithms that are used “to adapt input streams of video
`data to meet desired parameters for output streams of video data.” Col. 1:16-19. Depending on which
`algorithms are used and in what combination, the data will be compressed to a greater or lesser amount.
`As a person of ordinary skill in the art would have understood, a higher compression ratio typically
`requires more computation and results in lower compression rate; thus the compression rate will vary
`depending on the level of compression. A person of ordinary skill in the art would have recognized that
`the standard encoders disclosed in Brooks, such as MPEG-1, MPEG-2, and MPEG-4 allow for different
`compression levels corresponding to different rates. Indeed, as Brooks discloses, video data consists of
`three different types of frames – I frames, P frames, and B frames – but only certain types may be
`compressed. Col. 14:43-48. Typically, the more types of frames that are compressed, the higher the
`compression ratio, but the lower the compression rate. For example, B-frames or B-pictures can provide
`a higher compression ratio but at the expense of more complex bidirectional motion estimation (at the
`encoder) than P-frames or P-pictures.
`Brooks provides specific examples of different asymmetric data compression encoder compressing data
`blocks at different compression rates. For example, Brooks notes that a DSL connection “is capable of
`providing computing system 140 [i.e. the receiving computer] with enough video data to display up to a
`
`
`
`2
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 12
`
`
`
`Asserted Claims
`
`[1c] one or more processors
`configured to: determine one or
`more data parameters, at least
`one of the determined one or
`more data parameters relating
`to a throughput of a
`communications channel
`measured in bits per second;
`and
`
`Invalidity Chart for the ’477 Patent
`Exhibit 3 (Brooks)
`
`
`Corresponding Disclosure in the Prior Art Reference(s)
`640×480 pixel color image at 10 frames per second (fps) using an MPEG1 format.” Col. 6:32-41. On
`the other hand, for a 56kbps modem, “because the bandwidth is relatively small, network connection 210
`is capable of providing computing system 120 with enough video data to display up to a 160×120 color
`image at 5 fps using an MPEG4 format.” Col. 6:42-51. In these examples, the MPEG-1 Video encoder is
`compressing more data (because it is receiving higher resolution frames at a higher frame rate) and providing
`more data per frame at a faster frame rate to be transmitted than the MPEG-4 Video encoder. Thus, the
`MPEG-1 Video encoder is operating at a significantly higher compression rate than the MPEG-4 Video
`encoder.
`Brooks discloses numerous other examples of encoders operating at different compression rates depending on
`the available bandwidth. See col. 6:24 – col. 7:19.
`
`
`
`Brooks discloses a processor that is configured to perform the operations specified in the claim. For
`example, Brooks explains that the “gateway computer” that receives and processes input video and
`provides output video includes a processor. Col. 7:4-11, 51-52; Fig. 2. In particular, as discussed further
`below, control block 450, which “is used to control the processing of data within the processing unit,”
`determines the parameters relating to the encoding of the input video data and selects the appropriate
`encoder. Col. 9:27 – col. 10:62. Brooks notes that “[i]n the present embodiment, control block 450 is
`embodied as firmware running on a microprocessor (uP) including program memory, data memory, and
`the like.” Col. 10:63-65.
`Control block 450 in Brooks determines a number of data parameters relating to the video data to be
`compressed, including parameters of the input video, such as resolution, frames per second, and so forth,
`and parameters required for the output video, including “bandwidth parameters”:
`See, e.g. col. 9:34 - 10:57:
`In one example, control block 450 receives information associated with the stream of input video
`data. Such information typically includes bandwidth parameters such as the spatial resolution of
`the input video contained within the stream of input video data, the color bandwidth, or color bit-
`depth, of the input video, the number of frames per second of the input video, and the like. In this
`embodiment, the information also includes the video format, i.e. how the input stream of data is
`encoded, such as MPEG format, Windows Media format, H.263, QuickTime format, Real Video
`
`
`
`3
`
`Realtime Adaptive Streaming LLC
`Exhibit 2005
`IPR2019-01035
`Page 13
`
`