`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`___________________________________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`___________________________________________
`
`
`NETFLIX, INC.,
`Petitioner,
`
`v.
`
`DIVX, LLC,
`Patent Owner.
`___________________
`
`IPR2020-00511
`U.S. Patent No. 9,184,920
`___________________
`
`PETITIONER’S REPLY
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`I.
`
`II.
`
`TABLE OF CONTENTS
`DivX’s collateral attack is wrong on the facts and improper under the
`law. ............................................................................................................... 5
`A.
`Chen is not limited to TS packets. ....................................................... 7
`B.
`DivX’s collateral attack is prohibited by the Federal Circuit. ............ 11
`C.
`Frames exist inside the Transport Stream. ......................................... 13
`D. Actual POSITAs combined CA systems with partial frame
`encryption. ........................................................................................ 15
`Ground 1 does not require frame-by-frame transmission................... 21
`E.
`Ground 1 ..................................................................................................... 22
`A. DivX attempts to distinguish the prior art with unclaimed audio
`features. ............................................................................................ 23
`POSITAs knew how to handle alleged signal quality issues.............. 24
`B.
`III. Ground 2 ..................................................................................................... 27
`A.
`Candelore teaches encrypting multiple service (content) keys in
`an EMM. ........................................................................................... 27
`DivX’s arguments contradict its positions in district court. ............... 28
`B.
`IV. Ground 3 ..................................................................................................... 29
`A. Kocher confirms motivation to encrypt EMMs with player
`keys. ................................................................................................. 32
`DivX improperly limits Kocher. ....................................................... 33
`The combination uses Kocher’s keys as intended.............................. 34
`
`B.
`C.
`
`
`
`
`
`
`
`
`
`
`
`-2-
`
`
`
`
`
`
`
`PETITIONER’S UPDATED LIST OF EXHIBITS
`EX Description
`1001 U.S. Patent No. 9,184,920 to Grab et al. (“the ‛920 Patent”)
`1002 Prosecution File History for the ’920 Patent
`1003 Declaration of Dr. Patrick D. McDaniel
`1004 U.S. Patent Application Publication No. 2005/0177741 to Chen et al.
`(“Chen”)
`1005 U.S. Patent Application Publication No. 2004/0081333 (“Grab-333”)
`1006 U.S. Patent Application Publication No. 2005/0063541 to Candelore
`(“Candelore”)
`1007 PCT Publication No. WO 2005/008385 (“Kocher”)
`1008 U.S. Patent No. 6,957,350 to Demos. (“Demos”)
`1009 U.S. Patent No. 7,007,170 to Morten. (“Morten”)
`1010 U.S. Patent Application Publication No. 2004/0022391 to O’Brien
`(“O’Brien”)
`1011 Henry Beker et al., Cipher Systems – The Protection of Communications,
`John Wiley and Sons, 1982. (“Beker”)
`1012 Divx, LLC v. Netflix Inc., No. 2:19-cv-1602-PSG, Am. Compl. (C.D. Cal
`Aug. 21, 2019)
`1013 Declaration of Sylvia Hall-Ellis (served only)
`1014 Block Diagrams
`1015 Nielson et al., Practical Cryptography in Python: Learning Correct
`Cryptography by Example, Apress, 2019
`1016 United Services Automobile Association v. Asghari-Kamrani, IPR2015-
`01842 Ex. 1003
`1017 DivX infringement contention Claim 2
`1018 U.S. Patent Application Publication No. 2004/0181666 (“Candelore-666”)
`1019 U.S. Patent No. 7,747,853 (“Candelore-853”)
`1020 U.S. Patent No. 8,452,010 (“Candelore-010”)
`
`
`
`
`
`
`-3-
`
`
`
`
`
`
`
`EX Description
`1021 Excerpt of Peterson et al., Computer Networks – A Systems Approach,
`Morgan Kaufmann Publishers, 2007
`1022 Recommendation for Block Cipher Modes of Operation, NIST Special
`Publication 800-38A, 2001
`1023 RFC 791, Internet Protocol, DARPA Internet Program Protocol
`Specification, September 1981
`1024 Kerberos Ticket
`1025 Excerpt of Hansen, The Dictionary of Multimedia Terms & Acronyms, 4th
`Ed., Franklin, Beedle & Associates Inc., 2005
`1026 Excerpt of Kaplan, Wiley Electrical and Electronics Engineering
`Dictionary, John Wiley & Sons, Inc., 2004
`[MS-DRM] Digital Rights Management License Protocol Specification,
`Microsoft Corp., 2009
`1028 Data Encryption Standard (DES), FIPS PUB 46-3, October 1999
`1029 Adamson et al., Secure Distributed Virtual Conferencing, 1999
`1030 Deposition Transcript of Dr. Seth Nielson
`1031 RFC 1112, Host Extensions for IP Multicasting, August 1989
`1032 Reply Declaration of Dr. Patrick D. McDaniel
`
`
`1027
`
`
`
`
`
`
`
`
`-4-
`
`
`
`
`
`
`
`DivX does not dispute that Chen teaches the alleged invention of a 3-level
`
`key hierarchy or that Grab-333 teaches partial encryption. Ground 1 combines
`
`these known techniques. Rather than address the references’ express teachings,
`
`DivX attempts to create incompatibilities where there are none, using sideshow
`
`collateral attacks to limit Chen’s teachings to the particular implementations of
`
`other CA references and arguing details, such as networking or audio, that are
`
`mentioned nowhere in the challenged claims.
`
`This is the epitome of bodily incorporation and irrelevant to obviousness.
`
`Nonetheless, if DivX insists on analyzing other CA references, as rebuttal,
`
`Petitioner demonstrates that the prior art includes numerous examples of CA
`
`systems performing partial encryption of video frames, confirming that POSITAs
`
`were motivated and in fact combined the teachings that DivX contends to be
`
`incompatible.
`
`I.
`
`DivX’s collateral attack is wrong on the facts and improper under the
`law.
`As explained in the Petition, Chen teaches scrambling “the video, audio and
`
`data stream components” of a program. Pet., 23; EX1004 ¶[0004]. Chen teaches
`
`scrambling the elementary streams (“ES”) for these components separately.
`
`EX1004 ¶[0004] (“Scrambling may be applied commonly or separately to the
`
`
`
`
`
`
`-5-
`
`
`
`
`
`
`
`different elementary stream components of a program …”),1 ¶[0008] (“A CA
`
`system is also flexible as it may be exercised on an elementary stream-by-stream
`
`basis…”); EX1003 ¶114; EX2016, 14:8-15, 42:9-16. Therefore Chen expressly
`
`teaches scrambling the video ES separate from other components such as audio.
`
`EX1032 ¶5.
`
`It is undisputed that video elementary streams include frames and are not
`
`packetized. See Pet., 24; POR, 10; EX2009 ¶¶68, 75; EX1032 ¶4. Since Chen
`
`teaches encrypting the video ES, it would have been obvious that Chen also
`
`teaches decrypting the video ES because, as a basic networking principle,
`
`functionality applied at one layer in the transmitter is applied in the same layer at
`
`the receiver. EX1032 ¶9.
`
`DivX argues that CA systems cannot encrypt video frames but fails to
`
`address Chen’s express teachings to the contrary. And Dr. Nielson admitted that
`
`video frames may be encrypted even after packetized into a packetized elementary
`
`stream (“PES”). EX1030, 35:14-17, 47:12-48:4, 82:4-6. A PES packet is a video
`
`frame. EX2024 ¶9; EX1030, 33:10-13, 33:22-25; EX2006 §1.1, §1.3. DivX’s
`
`exhibit shows frames being scrambled before they are split into transport stream
`
`(“TS”) packets, explaining that “[m]aximum flexibility in the operation of a
`
`
`1 All emphasis added unless stated otherwise.
`
`
`
`
`
`
`-6-
`
`
`
`
`
`
`
`broadcast infrastructure requires scrambling to be allowed at the PES level.”
`
`EX2008, §§5.3, 4 (Fig. 1):
`
`
`
`EX1032 ¶¶6-8.
`
`A. Chen is not limited to TS packets.
`DivX admits ES and PES both include video frames. It therefore attempts to
`
`limit Chen to scrambling the next layer (TS packets). However, Chen never
`
`mentions “transport” streams or “packets.” Instead, Chen teaches scrambling
`
`elementary streams, which DivX ignores. A POSITA would have known how to
`
`apply Chen’s teachings with or without TS packets—Chen leaves those
`
`networking details to the POSITA. EX1032 ¶¶10-13; EX1030, 32:15-18.
`
`Some CA systems used TS packets; others did not. For example, Candelore-
`
`666 teaches “exemplary” CA embodiments that use the “(MPEG) transport layer”
`
`and other embodiments (Figs. 19F-19G) where “each stream of A/V content is not
`
`configured in accordance with MPEG transport requirements. Rather, each
`
`stream is a program stream of Packetized Elementary Stream (PES) packets.”
`
`EX1018 ¶[0151]. “[N]o MPEG transport is provided. Thus, the A/V content … is
`
`
`
`
`
`
`-7-
`
`
`
`
`
`
`
`sent directly as a collection of PES packets” in IP datagrams. Id. ¶[0161], Figs.
`
`19F-19G:
`
`
`
`Candelore-666 teaches that CA systems may transmit video frames (PES
`
`packets) inside IP datagrams without dividing them into TS packets, using known
`
`networking techniques, and “A/V content associated with the PES packets is
`
`recovered at the tuner/demodulator…” EX1018 ¶[0151]. The IP protocol was well-
`
`known by 2006. A POSITA would have recognized this was appropriate for
`
`delivering video over the Internet using TCP/IP, which provides a reliable,
`
`lossless, sequential connection. See id. ¶¶[0030], [0163]; EX1032 ¶12.
`
`The Candelore references are not obscure examples. They are assigned to
`
`
`
`
`
`
`-8-
`
`
`
`
`
`
`
`Sony, a major company in video equipment and content distribution.2 See EX1019;
`
`EX1006; EX1020; EX1032 ¶13.
`
`DivX argues that Chen must scramble/descramble TS packets because that is
`
`how certain other CA references work, citing Massel. But DivX ignores key
`
`differences. As explained above, Chen expressly teaches scrambling ES. And
`
`Massel shows ECM/EMM passing through the descrambler, whereas Chen’s
`
`ECM/EMM do not, suggesting that, unlike Massel, Chen’s demultiplexing happens
`
`before descrambling. EX1032 ¶¶14-15; compare EX2004, Fig. 9.7 with EX1004
`
`¶[0013], Fig. 1 (text omitted for emphasis):
`
`
`
`Thus Chen suggests an implementation more similar to Candelore-666 Fig.
`
`18, rather than Massel, where a CA set-top box demultiplexes TS packets into
`
`2 Dr. Nielson was unwilling or unable to answer questions on whether Sony played
`
`a major role in digital video or designed CA systems. EX1030, 24:22-26:11.
`
`
`
`
`
`
`-9-
`
`
`
`
`
`
`
`separate audio and video component streams before any decryption/decoding. See
`
`EX1018 ¶[0028], Fig. 18 (text omitted for emphasis):
`
`EX1032 ¶16.
`
`
`
`DivX argues that CA systems can only descramble multiplexed TS packets
`
`where video is indistinguishable from audio. But Candelore-666 disproves this,
`
`teaching a CA system decrypting an isolated video stream (“MAIN PROGRAM
`
`VIDEO”) without audio or any other data. See EX1018, Fig. 18. Chen teaches
`
`encrypting/decrypting elementary stream components “separately.” EX1004
`
`¶[0004]. A POSITA would have known how to implement this—e.g., by
`
`demultiplexing packets before the decryption/decoding process. EX1032 ¶¶17-19.
`
`
`
`
`
`
`-10-
`
`
`
`
`
`
`
`Different CA implementations were known in the art.3 EX1032 ¶¶18, 62;
`
`EX1030, 40:19-41:24, 43:9-18, 43:19-44:3. It would be wrong to limit Chen’s
`
`broad teachings to any single implementation, particularly when it differs from
`
`Chen’s express teachings.
`
`B. DivX’s collateral attack is prohibited by the Federal Circuit.
`In a previous IPR, the Board explained to DivX’s lead counsel that its bodily
`
`incorporation arguments were contrary to Federal Circuit precedent. See Hulu,
`
`LLC v. Sound View Innovations, LLC, Dkt. IPR2018-00366, Paper 41 at 23-24, 31
`
`(PTAB Apr. 9, 2019). Despite this warning, DivX again asks the Board to apply
`
`bodily incorporation here. DivX invites the Board to error.
`
`The Federal Circuit has repeatedly rejected bodily incorporation arguments
`
`that limit prior art teachings to physical/commercial embodiments. See In re
`
`Mouttet, 686 F.3d 1322, 1332 (Fed. Cir. 2012); In re Keller, 642 F.3d 413, 425
`
`(CCPA 1981). DivX argues that Chen should be limited to the particular hardware
`
`
`DivX’s exhibits confirm that CA systems included proprietary
`
`3
`
`implementations. See EX2004, 122 (“[T]he CA system can be (a) a totally
`
`proprietary system or (b) based on a common standard such as the DES or DVB
`
`descramblers.”); EX2014 §4.3 (“It is undesirable for the CA system to be
`
`standardized … a plurality of CA systems may be adopted…”); EX1032 ¶62.
`
`
`
`
`
`
`-11-
`
`
`
`
`
`
`
`descramblers of commercial CA set-top boxes.4 See POR, 17-18; EX1030, 60:24-
`
`61:12 (“[I]t does not get the benefit because of the differences in hardware” in
`
`Chen and Grab-333), 119:6-24 (“the two things don’t fit together”), 59:6-20,
`
`126:16-127:6. That is textbook bodily incorporation. It would be improper to limit
`
`Chen’s teachings to specific hardware, much less hardware mentioned nowhere in
`
`the reference itself.5
`
`Limiting prior art teachings to an embodiment—even a preferred
`
`embodiment—is improper. See In re Applied Materials, Inc., 692 F.3d 1289, 1298
`
`(Fed. Cir. 2012); EWP Corp. v. Reliance Universal Inc., 755 F.2d 898, 907 (Fed.
`
`Cir. 1985); In re Lamberti, 545 F.2d 747, 750 (CCPA 1976). DivX takes bodily
`
`incorporation to the next level, asking the Board to limit Chen’s teachings to the
`
`embodiments of other CA references.
`
`DivX would limit Chen to Massel’s use of TS packets, even though Chen
`
`never mentions packets or TS. Chen is not limited to the networking
`
`Yet Dr. Nielson does not know which companies manufactured CA systems.
`
`4
`
`EX1030, 26:12-27:7.
`
`5
`
`DivX cites an embodiment of Candelore that does not use a processor. But
`
`that is “one embodiment” (EX1006 ¶[0049]), and Candelore teaches that CA
`
`systems may include computers or other devices, not just set top boxes. Id.
`
`¶[0026]. Computers used general-purpose processors. EX1032 ¶¶49-50.
`
`
`
`
`
`
`-12-
`
`
`
`
`
`
`
`implementation of Massel or other references. Networking and packetization were
`
`basic tools, long known in the art (EX1032 ¶18), which a POSITA knew how to
`
`apply to the combination of Chen and Grab-333—this includes but does not
`
`require TS packets. Id. ¶¶10-20, 28-38, 45.
`
`The challenged claims do not recite any networking or packet limitations.
`
`DivX’s attempt to inject networking and packetization into this proceeding is
`
`precisely the type of bodily incorporation argument the Federal Circuit has
`
`rejected. See Packers Plus Energy Servs. Inc. v. Baker Hughes Oilfield Operations,
`
`LLC, 773 F. App’x 1083, 1089 (Fed. Cir. 2019) (nonprecedential) (rejecting
`
`argument that one reference’s cup packers “would be unsuitable for” the
`
`combination, because “the claims do not require a particular type of packer…”);
`
`Smith & Nephew, Inc. v. Rea, 721 F.3d 1371, 1381 (Fed. Cir. 2013) (“[A]n
`
`unclaimed and undisclosed feature … cannot be the basis for finding [a] patent to
`
`be non-obvious over the prior art.”).
`
`Frames exist inside the Transport Stream.
`C.
`Even if Chen were limited to TS packets, DivX’s arguments would still fail
`
`because they rely on the faulty premise that video frames do not exist in the TS.
`
`See POR, 8-10; EX2009 ¶¶72-75; EX1030, 37:14-38:13. The prior art refutes this
`
`notion—a POSITA would have known how to process video frames in TS packets,
`
`which included information in headers and payloads that POSITAs used to identify
`
`
`
`
`
`
`-13-
`
`
`
`
`
`
`
`video frames in the TS. EX1032 ¶¶20-21.
`
`For example, Candelore-666 teaches partial encryption of video frames in
`
`the TS, where “the transport stream packets containing a PES header can be
`
`selected for encryption…” EX1018 ¶¶[0087], [0090]. “For the video and audio,
`
`this can be accomplished by encrypting ‘start of frame’ transport stream packets
`
`containing PES (packetized elementary stream) headers and other headers as part
`
`of the payload…” Id. ¶[0086]. “MPEG-2 streams identify ‘start of frame’ packets
`
`with the ‘Packet Unit Start Indicator’ in the transport header.” Id. “When a
`
`transport stream packet is received at block 350, the packet is examined to
`
`determine if it meets a selection criteria for encryption. … [T]his selection criteria
`
`is the presence of a PES header as a portion of the packet payload.” Id. ¶[0091].
`
`Candelore-666 teaches encrypting the PES header and a portion of the video frame
`
`(PES payload)—“as much of the content as the bandwidth will allow.” Id. ¶[0094];
`
`EX1032 ¶¶22-23.
`
`Candelore-666 also teaches encrypting only I frames within a stream of TS
`
`packets. See EX1018 ¶¶[0079]-[0081], [0090]. Candelore-010 explains that
`
`portions of video frames can be encrypted either before or after packetization. See
`
`infra §I.D.1; EX1032 ¶¶24-25.
`
`A POSITA knew how to scramble and descramble video frames inside a TS
`
`packet stream based on indicators in TS packet headers and payloads. See EX1018
`
`
`
`
`
`
`-14-
`
`
`
`
`
`
`
`¶[0086]. A TS packet is a portion of a frame. EX1030, 34:2-7. Video frames exist
`
`inside the TS. Frames do not disappear when packaged into TS packets; the frames
`
`exist in packet payloads, and the prior art confirms that POSITAs knew how to
`
`encrypt/decrypt6 select portions of frames inside TS packets. EX1032 ¶¶24-25.
`
`D. Actual POSITAs combined CA systems with partial frame
`encryption.
`DivX contends “scrambling is applied to TS packets.” POR, 7. Dr. Nielson
`
`contends that “[e]ncryption at the frame level was not done” in CA systems and
`
`questions “if it had ever been performed.” EX2024 ¶11; EX1030, 56:8-15
`
`(encryption “is only performed across the entire elementary stream.”). But DivX’s
`
`own evidence confirms that “elementary streams or parts thereof may be
`
`scrambled for conditional access.” EX2019, 41 §2.4.4.
`
`The prior art included numerous examples where POSITAs partially
`
`encrypted video frames in CA systems providing express teaching, suggestion, and
`
`motivation to combine Chen’s CA teachings (for distributing keys) with Grab-
`
`333’s teachings (for partially encrypting frames). EX1032 ¶27. This includes
`
`
`Candelore-666 focuses on encryption. It was understood that decryption
`
`6
`
`followed the inverse process. Therefore, Candelore-666’s teachings—regarding
`
`partial encryption of video frames—encompassed the inverse process of decrypting
`
`those portions of video frames. EX1032 ¶26.
`
`
`
`
`
`
`-15-
`
`
`
`
`
`
`
`proprietary systems and Internet applications. See id.; EX1018 ¶¶[0030], [0163];
`
`EX1006 ¶[0021]. A motivation to combine may be found in the “interrelated
`
`teachings of multiple patents” and “the background knowledge, creativity, and
`
`common sense of the person of ordinary skill.” ZUP, LLC v. Nash Mfg., Inc., 896
`
`F.3d 1365, 1371 (Fed. Cir. 2018).
`
`Candelore-010 teaches encrypting selected portions of video frames in CA
`
`set-top boxes. See EX1020, 3:9-16. Fig. 6 illustrates a video frame with encrypted
`
`slice headers, and Fig. 7 illustrates encrypted headers and first macroblocks for
`
`each slice.7 EX1020, 8:65, 9:17, Figs. 6, 7:
`
`
`Slices are portions of frames. Each slice is a group of macroblocks. EX1032
`
`7
`
`¶29. Dr. Nielson does not consider himself to be an expert in video encoding
`
`(EX1030, 13:23-14:8, 17:22-18:10) and did not know basic concepts such as
`
`macroblocks or slices. EX1030, 49:2-15, 55:16-56:7, 52:17-53:13.
`
`
`
`
`
`
`-16-
`
`
`
`
`
`
`
`FIG. 6
`
`
`
`
`
`I W
`
`gm
`;
`I'
`3.. .
`,
`
`3%
`u! E.
`
`J
`riff-(I?
`gig-£4
`
`
`
`
`
`
`
`
`-17-
`
`
`
`
`
`
`
`Fig. 10 illustrates “full encryption of all data in the active slices” to encrypt the
`
`middle of a video frame. See, EX1020, 10:7-22, Fig. 10:
`
`
`
`EX1032 ¶¶28-30.
`
`Candelore-010 confirms that POSITAs knew how to encrypt portions of
`
`video frames before or after they are divided into TS packets. See EX1030, 34:2-7.
`
`Fig. 8 teaches encrypting an active region in the center of the frame without regard
`
`to packet boundaries, while Fig. 9 teaches adjusting the encrypted portion based on
`
`TS packets. See EX1032 ¶31; EX1020, 9:36, 9:52-67, Figs. 8, 9:
`
`
`
`
`
`
`-18-
`
`
`
`
`
`
`
`FIG. 8
`
`
`
`266
`
`
`
`FIG. 9
`
`
`
`
`
`
`
`
`WWII/WWfl/W
`
`
`
`WWI/IMIMI/mi
`
`
`§k
`WWWI’/ IAé/I/2W
`
`WIIIIIII/WWWIII/IIII
`
`
`
`”WWII/I/I"1%sz
`
`
`
`
`W/IW
`
`
`
`
`IIIIIIIWWIIAIIII/I WW
`
`
`
`
`
`II/IIWWA’{gI/II”mg
`IIIII
`
`
`ly’II’fII/IIIA/z’III/1% IMM’Z
`
`
`
`
`
`
`WWWW JIIWII
`
`
`
`
`MIA/WI AW
`1’1’/II/.///WW1WE
`
`
`
`
`WWW
`
`
`WIIWW/MW
`WIIIJAIII III/WWA
`
`
`W/W/1I1§I’1W/1WW/
`
`
`
`
`WWI’M/{WI
`
`
`WWI IIWWIII1m
`
`
`
`
`
`2TB
`
`
`
`
`
`-19-
`
`
`
`
`
`
`
`
`
`
`
`
`Dr. Nielson contends CA systems cannot encrypt just the central area of a
`
`video frame at the TS level. EX1030, 48:5-15. But that is precisely what
`
`Candelore-010 Fig. 9 teaches. Moreover, Candelore-010 teaches that, “[w]hile the
`
`above embodiments describe encryption of packets containing the selected data
`
`type, it is also possible to encrypt the raw data prior to packetizing … such
`
`encryption is considered equivalent thereto.” EX1020, 13:38-42. Candelore-010
`
`teaches encrypting at the network level (encrypting the “payload of a packet”) or
`
`frame level (directly encrypting “the data itself”). EX1020, 8:54-58; EX1032 ¶32.
`
`POSITAs had the capability and motivation to apply partial encryption to
`
`CA systems, and TS packetization did not discourage or prevent them from doing
`
`so. See EX1032 ¶34.8 Indeed, Grab-333 teaches mechanisms that allow precise
`
`control allowing encryption to be applied to selected portions of the video frame,
`
`including by location and size, which a POSITA knew how to use to configure
`
`encryption based on TS packet boundaries, as known in the art and illustrated by
`
`Candelore-010. See EX1032 ¶33; EX1005 ¶¶[0038], [0040], [0045].
`
`Kocher also teaches CA systems scrambling video frames, teaching that
`
`“MPEG ‘I’ frames could be encrypted, while other communications could be
`
`unencrypted.” EX1007, 116:3-18. Moreover, Kocher teaches forensics for CA
`
`8 Dr. Nielson generally agrees with Dr. McDaniel on the POSITA’s background
`
`knowledge. See EX1030, 120:14-121:2, 124:9-14, 125:2-126:21.
`
`
`
`
`
`
`-20-
`
`
`
`
`
`
`
`systems where blocks in video frames—not packets—are encrypted/decrypted. See
`
`EX1007, 102:25, 103:10-13 (explaining that CA set top boxes had “decryption
`
`keys used to decrypt the video”). Because Kocher’s encryption process evaluates
`
`the “[q]uality of the resulting video,” Kocher’s CA system encrypts video frames,
`
`not TS packets. See EX1032 ¶¶35-38; EX1007, 106:15-29.
`
`E. Ground 1 does not require frame-by-frame transmission.
`DivX argues frame-by-frame transmission is infeasible (POR, 4), but
`
`nothing in Ground 1 (or the challenged claims) requires it. Chen does not limit its
`
`teachings to particular transmission mechanisms, and a POSITA would have
`
`implemented Chen (and the combination of Chen and Grab-333), using known
`
`transmission techniques, which includes but does not require TS packets, TCP/IP,
`
`or other protocols. EX1032 ¶39.
`
`The exact network implementation is irrelevant to the challenged claims.
`
`Here, DivX argues more bodily incorporation, treating Chen Fig. 1 as a physical
`
`schematic with a physical wire connecting the scrambler and descrambler. But
`
`Chen’s Fig. 1 is a conceptual “block diagram,” not a schematic. See EX1004
`
`¶[0010]; EX1025; EX1026. As DivX admits, Chen is not concerned with and
`
`therefore does not illustrate transmission details. POR, 11; EX2009 ¶74. That does
`
`not mean transmission mechanisms are missing; it simply means known
`
`mechanisms are applied, for example with packetization between the scrambler
`
`
`
`
`
`
`-21-
`
`
`
`
`
`
`
`and descrambler. See EX1032 ¶40; EX1020, 13:38-42. It was common for such
`
`details to be abstracted as part of the channel. See EX1032 ¶41; EX1021, Fig. 1.7:
`
`
`
`The ’920 patent likewise includes diagrams without transmission or
`
`networking details (see EX1001, Figs. 2, 5, 7; EX1032 ¶42), suggesting they were
`
`known in the art. See Uber Techn., Inc. v. X One, Inc., 957 F.3d 1334, 1339 (Fed.
`
`Cir. 2020); In re Epstein, 32 F.3d 1559, 1568 (Fed. Cir. 1994).
`
`II. Ground 1
`As the Petition explained, Ground 1 combines Chen’s teachings for
`
`distributing keys with Grab-333’s teachings for partially encrypting video frames.
`
`See Pet., 15-19, 25-28; EX1003 ¶103. Chen and Grab both teach
`
`encrypting/decrypting video frames. See supra §I.A (explaining Chen); EX1003
`
`¶¶112, 114-115; EX2016, 65:3-5; EX1005, Figs. 5, 6, 8, 11-12, ¶¶[0016], [0019],
`
`[0048], [0055]-[0056]. Video frames exist at the playback device and it would
`
`have been obvious to decrypt them (in the ES, PES or TS) because Grab-333
`
`
`
`
`
`
`-22-
`
`
`
`
`
`
`
`teaches decrypting video frames. See id. For the combination of Ground 1,
`
`encrypted frames are obtained and decrypted at the RX (playback device). EX1032
`
`¶47. DivX’s arguments for limitations [1c], [1g], [1j] assume that no frames are
`
`obtained by the RX for decryption, which is not correct. EX1032 ¶48; see supra
`
`§I.A, §I.C.
`
`Partial encryption was commonly used to reduce computational overhead,
`
`including for streaming applications. Pet., 17; EX1003 ¶102; EX1005 ¶[0054]. The
`
`prior art provides express motivation to combine CA systems with partial
`
`encryption, and demonstrates a reasonable expectation of success because actual
`
`POSITAs made the combination while accounting for packetization or using
`
`proprietary encryption/decryption systems. See supra §I.C, §I.D; EX1032 ¶¶43-46.
`
`A. DivX attempts to distinguish the prior art with unclaimed audio
`features.
`The Federal Circuit has rejected arguments based on alleged implementation
`
`incompatibilities that are not required by the claims. See Packers Plus, 773 F.
`
`App’x at 1089; Smith & Nephew, 721 F.3d at 1381. DivX faults the Petition for not
`
`explaining how audio is encrypted, but the challenged claims do not recite any
`
`audio limitations. The ’920 specification likewise does not explain how audio is
`
`encrypted, instead relying on the POSITA’s background knowledge. EX1032
`
`¶¶51-52; see Uber, 957 F.3d at 1339; Epstein, 32 F.3d at 1568.
`
`
`
`
`
`
`-23-
`
`
`
`
`
`
`
`It was common for video patents to focus on video because the handling of
`
`audio was well known in the art. Kocher confirms this, explaining video
`
`encryption and stating, without additional detail, that “[a]lthough the examples
`
`above are described primarily with respect to the video portion of audiovisual
`
`content, the same techniques are equally applicable to audio and other kinds of
`
`content.” EX1007 108:5-7. Similarly, Candelore-666 teaches partial encryption of
`
`video while explaining that a POSITA would have known several variations for
`
`handling audio. See EX1018 ¶0090; EX1032 ¶¶53-54.
`
`POSITAs knew how to handle alleged signal quality issues.
`B.
`DivX argues that the combination would be inferior for unclaimed “signal
`
`quality” metrics if “unknown” packet loss occured on encrypted data. See POR,
`
`22-23. But DivX admits a POSITA knew how to transmit packets without loss.
`
`POR, 24-25; EX1032 ¶57. Dr. Nielson admits that various techniques were known
`
`for detecting and addressing packet loss (see EX1030, 76:16-20, 80:4-7) and that
`
`“TS descrambling is not affected by packet loss.” EX2024 ¶39. Nothing in the
`
`Grounds precludes those known techniques, such as using TS packets. See supra
`
`§I.A, §I.D. And a POSITA knew how to use protocols, e.g., TCP/IP, to detect and
`
`correct packet loss, including for video streaming. See EX1023; EX1029.
`
`Therefore, a POSITA would have combined the references while avoiding DivX’s
`
`“unknown” packet loss scenario. EX1032 ¶¶55-57.
`
`
`
`
`
`
`-24-
`
`
`
`
`
`
`
`The Federal Circuit has rejected similar arguments regarding “signal loss”
`
`when the challenged claims did not require any particular thresholds for it. See Slot
`
`Speaker Technologies, Inc. v. Apple Inc., 680 F. App’x 932, 940 (Fed. Cir. 2017)
`
`(“[The] claims do not require … any specific threshold with respect to signal loss.
`
`… It is irrelevant whether Tomonori and Sadaie together would be less effective
`
`than Sadaie alone …”); Packers Plus, 773 F. App’x 1089. The challenged claims
`
`do not recite signal quality requirements. DivX’s arguments are irrelevant.
`
`Even if there were alleged drawbacks in certain scenarios, they did not
`
`dissuade POSITAs who in fact used partial encryption with CA systems. See supra
`
`§I.D. “[A] given course of action often has simultaneous advantages and
`
`disadvantages, and this does not necessarily obviate motivation to combine.” Allied
`
`Erecting & Dismantling Co. v. Genesis Attachments, LLC, 825 F.3d 1373, 1381
`
`(Fed. Cir. 2016) (citations omitted); In re Mouttet, 686 F.3d 1322, 1334 (Fed. Cir.
`
`2012) (“[T]hat ‘better alternatives exist in the prior art does not mean that an
`
`inferior combination is inapt for obviousness purposes.’”); Novartis Pharm. Corp.
`
`v. W.-Ward Pharm. Int’l Ltd., 923 F.3d 1051, 1059 (Fed. Cir. 2019); Bayer
`
`Pharma AG v. Watson Labs., Inc., 874 F.3d 1316, 1327, (Fed. Cir. 2017) (“[T]hat
`
`there may be reasons a skilled artisan would prefer one over the other does not
`
`amount to a teaching away from the lesser preferred but still workable option.”);
`
`KSR, 550 U.S. at 424.
`
`
`
`
`
`
`-25-
`
`
`
`
`
`
`
`Despite agreeing that “common sense is important and should be
`
`considered” under KSR (EX1030, 118:4-7), Dr. Nielson argues a POSITA would
`
`only be able to “follow[] the instructions to apply the cryptography in the context
`
`in which it’s described by a standard … or some other kind of guide.” EX1030,
`
`121:3-24, 123:8-16 (“If there is a guide and the guide is applicable to the situation
`
`that they’re working with….”), 122:14-21, 123:17-124:8. However, by requiring
`
`an explicit teaching in the art, Dr. Nielson applies the rigid TSM requirement that
`
`the Supreme Court rejected in KSR. See KSR Int’l Co. v. Teleflex Inc., 550 U.S.
`
`398, 421 (2007). Contrary to Dr. Nielson’s opinions, “[o]ne of ordinary skill in the
`
`art is not an ‘automaton.’” Id.; In re Applied Materials, Inc., 692 F.3d 1289, 1298
`
`(Fed. Cir. 2012). Dr. Nielson’s opinions should be afforded minimal weight
`
`because he applies an improper legal standard.9
`
`Moreover, DivX’s signal-loss argument assumes a specific cryptographic
`
`mode (CBC). See EX1030, 80:12-24. But Chen and Grab-333 are not limited to
`
`any one mode. While Grab-333 may disclose CBC, it is just an “example”. See
`
`EX1005 ¶[0047]. According to then-available best practices, other modes such as
`
`This would not be the first time Dr. Nielson’s opinions were afforded little
`
`9
`
`weight. See, e.g., Sensus USA, Inc. v. Certified Measurement, LLC, IPR2015-
`
`01262, Paper 20, 13 (PTAB Dec. 9, 2015) (according “little weight because Dr.
`
`Nielson provides insufficient underlying facts or data to support his conclusion.”).
`
`
`
`
`
`
`-26-
`
`
`
`
`
`
`
`CTR were appropriate for potentially lossy channels. CTR was one of the most
`
`popular modes for AES (to which Chen refers) and was recommended by NIST.
`
`See EX1022, 15. Dr. Nielson admits that CTR is recommended “[i]n almost all
`
`circumstances.” EX1015, 90. With CTR mode, the loss of one block does not
`
`preclude decryption of subsequent blocks. See id.; EX1032 ¶¶58-61.
`
`III. Ground 2
`A. Candelore teaches encrypting multiple service (content) keys in an
`EMM.
`Ground 2 applies Candelore’s service key teachings to Chen’s work key.
`
`Pet., 56-57. DivX contends that Candelore “fails to disclose a configuration with
`
`more than one work key per EMM,” and there is no explanation for how multiple
`
`work/service keys would be used. POR, 34-35. But Candelore provides this
`
`express teaching and explanation.
`
`As DivX admits, Candelore teaches that each EMM may contain multiple
`
`keys. EX1006 ¶[0039]; POR, 34; EX1003 ¶171; Pet., 55; EX1030, 89:2-15.
`
`Candelore teaches sending “one or more Service Keys … in the EMMs.”10
`
`EX1006 ¶[0052]; see EX1003 ¶171-172; Pet., 86-88. Why would multiple service
`
`keys be included in a single EMM? Candelore explains that the current and next
`
`service key may be delivered together: “Depending on the duration of the current
`
`
`Candelore teaches the keys are sent encrypted. EX1006 ¶[0058].
`
`-27-
`
`
`
`
`
`10
`
`
`
`
`
`
`
`
`Service Key (e.g., SK), the next Service Key (SKnext) may be delivered along with
`
`the SK.” EX1006 ¶[0064]. Additionally, Candelore teaches sending service keys
`
`for individual shows or movies ahead of time, and it would have been obvious
`
`based on Candelore’s teachings for these service keys to be included in a key table
`
`in an EMM. See EX1006 ¶[0054] (“service keys for individual shows or movies
`
`may be requested by subscriber terminal device 110 and delivered ahead of time”);
`
`EX1003 ¶¶95, 171-172; Pet., 86-88. Therefore, it would have been obvious to send
`
`multip
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
Still Working On It
This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.
Give it another minute or two to complete, and then try the refresh button.
A few More Minutes ... Still Working
It can take up to 5 minutes for us to download a document if the court servers are running slowly.
Thank you for your continued patience.
This document could not be displayed.
We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.
Your account does not support viewing this document.
You need a Paid Account to view this document. Click here to change your account type.
Your account does not support viewing this document.
Set your membership
status to view this document.
With a Docket Alarm membership, you'll
get a whole lot more, including:
- Up-to-date information for this case.
- Email alerts whenever there is an update.
- Full text search for other cases.
- Get email alerts whenever a new case matches your search.
One Moment Please
The filing “” is large (MB) and is being downloaded.
Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.
Your document is on its way!
If you do not receive the document in five minutes, contact support at support@docketalarm.com.
Sealed Document
We are unable to display this document, it may be under a court ordered seal.
If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.
Access Government Site
