British Telecommunications plc

Patent Trials and Appeals Board

Aug 5, 2020

IPR2019-00833 (P.T.A.B. Aug. 5, 2020)

Trials@uspto.gov Paper No. 20
571-272-7822 Date: August 5, 2020

UNITED STATES PATENT AND TRADEMARK OFFICE

BEFORE THE PATENT TRIAL AND APPEAL BOARD

VIMEO, INC. and IAC/INTERACTIVECORP,
Petitioners,
v.
BRITISH TELECOMMUNICATIONS PLC,
Patent Owner.

IPR2019-00833
Patent 7,974,200 B2

Before JESSICA C. KAISER, MONICA S. ULLAGADDI, and
SCOTT RAEVSKY, Administrative Patent Judges.
RAEVSKY, Administrative Patent Judge.

JUDGMENT
Final Written Decision
Determining Some Challenged Claims Unpatentable
35 U.S.C. § 318(a)

IPR2019-00833
Patent 7,974,200 B2
2
I. INTRODUCTION
Vimeo, Inc. and IAC/InterActiveCorp (“Petitioners”) filed a Petition
(Paper 2, “Petition” or “Pet.”) requesting an inter partes review of claims 1–
7 and 10–12 of U.S. Patent No. 7,974,200 B2 (Ex. 1001, “the ’200 patent”).
British Telecommunications plc (“Patent Owner”) filed a Preliminary
Response. Paper 6 (“Prelim. Resp.”). In view of those submissions, we
instituted an inter partes review of claims 1–7 and 10–12. Paper 7
(“Decision”). Subsequent filings include a Patent Owner Response (Paper 9,
“PO Resp.”), a Petitioner Reply (Paper 12, “Pet. Reply” or “Reply”), and a
Patent Owner Sur-Reply (Paper 13, “Sur-Reply”). An oral hearing was held
on June 4, 2020, and a copy of the transcript was entered into the record.
Paper 19 (“Tr.”).
We have jurisdiction over this proceeding under 35 U.S.C. § 6(b).
After considering the evidence and arguments of the parties, we determine
that Petitioners have proven by a preponderance of the evidence that
claims 1–3, 5–7, and 10–12 of the ’200 patent are unpatentable. See 35
U.S.C. § 316(e) (2018). We also determine that Petitioners have not proven
by a preponderance of the evidence that claim 4 is unpatentable. We issue
this Final Written Decision pursuant to 35 U.S.C. § 318(a).
II. BACKGROUND
A. Related Matters
Petitioners and Patent Owner identify a related litigation involving the
’200 patent: British Telecommunications plc v. IAC/InterActiveCorp, No.
18-cv-00366 (D. Del.), filed March 8, 2018. Pet. 2; Paper 4, 2; Ex. 1002.
IPR2019-00833
Patent 7,974,200 B2
3
B. The ’200 Patent
The ’200 patent concerns a method of streaming video to a client
while reducing the start-up delay associated with buffering data. Ex. 1001,
1:14–17. Figure 1 of the ’200 patent is reproduced below.

Figure 1 illustrates a video streaming system. Id. at 3:30–31, 40–41, 4:57.
Dual rate encoder 1 encodes video data into two sets of data, including low
bit rate data (RL) and high bit rate data (RH) having higher quality than the
low bit rate data. Id. at 3:41–44, 4:1–3. Encoder 1 supplies the low bit rate
data and the high bit rate data to video streamer 2, which supplies the data to
client devices a, b, c over the Internet. Id. at 3:48–58.
Video streamer 2 reduces start-up delay by transmitting the low bit
rate data at a transmission rate faster than the low bit rate. Id. at 5:11–
26, 3:62–67, 4:10–29. Client device a, b, or c receives the low bit rate data
in a client buffer and promptly reads out the data for display. Id. at 5:19–24.
Because video streamer 2 transmits the low bit rate data faster than its bit
rate, the client can display the video without first filling the buffer. Id.
at 5:11–26. When the buffer reaches a specified level of low bit rate data,
video streamer 2 switches to transmitting the high bit rate data. Id. at 5:27–
35. The client buffer then begins to fill with the high bit rate data, placed
IPR2019-00833
Patent 7,974,200 B2
4
behind the low quality data. Id. at 5:36–37. The client eventually decodes
the high bit rate data for display, whereupon the user will perceive an
increase in picture quality. Id. at 5:37–40.
C. Illustrative Claims
Claims 1, 4, 5, 7, and 10 are independent. Each of claims 2, 3, 6, 11,
and 12 depends from one of independent claims 1, 4, 5, 7, and 10. Claims 1
and 4 are illustrative of the challenged claims and recite:1
1. [1.0] A method of operating a real-time communication
apparatus comprising [a] a real-time data sender, [b] a real-time
data display device having [c] a store, and [d] a network
connecting said sender and said display device, said method
comprising:
[1.1] operating said sender to transmit a plurality of first-
encoding-rate data packets [a] encoded at a first encoding rate
and [b] representing a first part of a real-time presentation to
said display device [c] at a transmission rate higher than said
first encoding rate of said first-encoding-rate data packets;
[1.2] operating said display device to:
[1.3] receive said first-encoding-rate data packets into
said store;
[1.4] read-out said received first-encoding-rate data
packets from said store [a] at a data rate equal to said first
encoding rate of said first-encoding-rate data packets [b] for
decoding [c] so as to present said first part of said real-time
presentation to a user at a first level of quality;
[1.5] on said store being filled with said first-encoding-
rate data to a predetermined level, sending an indication that
said level has been reached to said data sender;
[1.6] operating said data sender, [a] on receipt of said
indication from said display device, to send second-encoding-
rate data packets [b] encoded at a second encoding rate and
[c] representing subsequent parts of said real-time presentation
to said display device, [d] said second encoding rate of said

1 For convenience, we add in brackets the numbering and lettering of the
claim limitations provided in the Petition. See Pet. 29–42, 44–57, 73–80.
IPR2019-00833
Patent 7,974,200 B2
5
second-encoding-rate data packets being higher than said first
encoding rate of said first-encoding-rate data packets;
[1.7] operating said display device to:
[1.8] receive [a] the second-encoding-rate data packets
[b] representing the subsequent part of real-time presentation
into said store;
[1.9] read out said second-encoding-rate data packets
from said store [a] at a data rate equal to said second encoding
rate of said second-encoding-rate data packets [b] for decoding
[c] so as to present said subsequent part of said real-time
presentation to said user [d] at a second level of quality which
is higher than said first level of quality.
Ex. 1001, 7:18–55.
4. [4.0] A method of operating a communication apparatus
comprising [a] a data sender, [b] a data presentation device
having [c] a store, and [d] a network connecting said data
sender and said data presentation device, said method
comprising:
[4.1] operating said data sender to transmit a first
plurality of first-encoding-rate data packets [a] encoded at a
first encoding rate to said data presentation device, wherein
[b] said first plurality of first-encoding-rate data packets are
transmitted at a first transmission rate [c] which is higher than
said first encoding rate of said first-encoding-rate data packets;
[4.2] operating said data presentation device to:
[4.3] receive said first plurality of first encoding rate data
packets into said store;
[4.4] read out said received first-encoding-rate data
packets from said store [a] at a data rate equal to said first
encoding rate of said first-encoding-rate data packets [b] for
decoding [c] so as to present to a user at a first level of quality,
[d] wherein the removing of said first-encoding-rate data
packets from said store is initiated when said first-encoding-rate
data packets first arrive at said store;
[4.5] on said store being filled with data packets to a
predetermined level, sending an indication to said data sender
that said predetermined level has been reached;
[4.6] operating said data sender, [a] on receipt of said
indication from said data presentation device, to transmit a
IPR2019-00833
Patent 7,974,200 B2
6
second plurality of second-encoding-rate data packets
[b] encoded at a second encoding rate to said data presentation
device, [c] wherein said second plurality of second-encoding-
rate data packets are transmitted at a second transmission rate
which is higher than said first transmission rate; and
[d] wherein said second encoding rate is higher than said first
encoding rate;
[4.7] operating said data presentation device to:
[4.8] receive said second plurality of second-encoding-
rate data packets into said store;
[4.9] read out said second-encoding-rate data packets
from said store [a] at a data rate equal to said second encoding
rate of said second-encoding-rate data packets [b] for decoding,
[c] so as to present to said user at a second level of quality,
wherein said second level of quality is higher than said first
level of quality.
Id. at 7:64–8:39.
D. The Asserted Grounds of Unpatentability
Petitioners advance the following challenges, relying on the
declaration of Dr. Clifford Reader (Ex. 1008, “Reader Decl.”):
Claim(s) Challenged 35 U.S.C. § Reference(s)/Basis
1–4, 7, 10 103(a)2 Lippman3, Chou4, Muroi5

2 The Leahy-Smith America Invents Act (“AIA”), Pub. L. No. 112-29,
125 Stat. 284, 287–88 (2011), amended 35 U.S.C. § 103. Because the ’200
patent was filed before the effective date of the relevant amendment, the pre-
AIA version of § 103 applies.
3 “Video Coding for Multiple Audiences,” SPIE vol. 3653, pp. 780–782,
pub. Jan. 1999 (Ex. 1010).
4 U.S. Patent No. 6,637,031 B1, iss. Oct. 21, 2003 (Ex. 1011).
5 JP H11-187367, pub. July 9, 1999 (Ex. 1012). Petitioners provided a
certified translation of Muroi as Exhibit 1013. When referring to “Muroi,”
the remainder of this Decision refers to Exhibit 1013. Patent Owner does
not challenge the accuracy of the translation in Exhibit 1013. Accordingly,
we treat Exhibit 1013 as an accurate translation of Muroi.
IPR2019-00833
Patent 7,974,200 B2
7
Claim(s) Challenged 35 U.S.C. § Reference(s)/Basis
5, 6, 11, 12 103(a) Lippman, Chou, Muroi, Lee6
III. ANALYSIS
A. Legal Standards
A claim is unpatentable under § 103(a) if the differences between the
claimed subject matter and the prior art are such that the subject matter, as a
whole, would have been obvious at the time the invention was made to a
person having ordinary skill in the art to which said subject matter pertains.
KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 406 (2007). The question of
obviousness is resolved on the basis of underlying factual determinations,
including (1) the scope and content of the prior art; (2) any differences
between the claimed subject matter and the prior art; (3) the level of skill in
the art; and (4) when in evidence7, objective indicia of non-obviousness (i.e.,
secondary considerations). Graham v. John Deere Co., 383 U.S. 1, 17–18
(1966). We analyze the asserted obviousness grounds with the principles
identified above in mind.
B. Level of Ordinary Skill in the Art
Petitioners assert that a person of ordinary skill “would have had
either a bachelor’s degree in electrical engineering or computer science, or
an equivalent degree, and two to three years of experience in the field of
audiovisual systems design, or a graduate degree in engineering or computer
science focusing on audiovisual systems design.” Pet. 9–10. Patent Owner

6 “An Integrated Source Coding and Congestion Control Framework for
Video Streaming in the Internet,” Proc. IEEE INFOCOM, Vol. 2, pp. 747–
756, pub. April 12, 2000 (Ex. 1014).
7 Patent Owner has not submitted any evidence of objective indicia of non-
obviousness in this case.
IPR2019-00833
Patent 7,974,200 B2
8
does not address the level of ordinary skill in its Patent Owner Response.
We adopt Petitioners’ assessment of the level of ordinary skill in the art
because it is consistent with the prior art cited and the evidence of record.
C. Claim Construction
In an inter partes review filed on or after November 13, 2018, we
apply the same claim construction standard used by Article III federal courts
in civil actions under 35 U.S.C. § 282(b), which follow Phillips v. AWH
Corp., 415 F.3d 1303 (Fed. Cir. 2005) (en banc), and its progeny. Changes
to the Claim Construction Standard for Interpreting Claims in Trial
Proceedings Before the Patent Trial and Appeal Board, 83 Fed. Reg. 51,340
(Oct. 11, 2018) (amending 37 C.F.R. § 42.100(b) effective November 13,
2018). Because the instant Petition was filed on March 22, 2019, we apply
that standard here. Accordingly, we construe each challenged claim of the
’200 patent to generally have “the ordinary and customary meaning of such
claim as understood by one of ordinary skill in the art and the prosecution
history pertaining to the patent.” 37 C.F.R. § 42.100(b).
Petitioners propose a construction for the terms “first-encoding-rate
data packets . . . representing a first part of a real-time presentation” and
“second-encoding-rate data packets . . . representing subsequent parts of said
real-time presentation.” Pet. 10–11. Petitioners assert these clauses “refer,
respectively, to the onset (the ‘first part’) of a real-time presentation, and to a
portion of the presentation that follows this first part (the ‘subsequent
parts’).” Id. at 10.
Petitioners do not argue that their proffered construction is dispositive
of the issue of patentability of the challenged claims, and Patent Owner does
not dispute Petitioners’ construction. See generally Pet.; PO Resp.; Pet.
Reply; Sur-reply. We determine that we need not explicitly construe these
IPR2019-00833
Patent 7,974,200 B2
9
or any other terms. See Nidec Motor Corp. v. Zhongshan Broad Ocean
Motor Co., 868 F.3d 1013, 1017 (Fed. Cir. 2017) (“[W]e need only construe
terms that are in controversy, and only to the extent necessary to resolve the
controversy.” (internal quotation and citation omitted)).
D. Overview of the Prior Art
Before turning to Petitioners’ asserted grounds, we provide an
overview of the asserted references.
1. Lippman
Lippman describes video streaming software that measures available
network bandwidth and switches between separate video encodings to match
channel capacity. Ex. 1010, 780. A player receiving a video stream from
the software collects a minimum number of packets, termed a “pre-roll
amount,” and then decodes the packets and combines the decoded results
into a multimedia presentation. Id. Content creators using the video
streaming system select which target audiences to provide content for,
including those having a 28.8k modem, 56k modem, ISDN, Dual ISDN,
Corporate LAN, XDSL, or Cable Modem. Id. at 781. There is an associated
target bit rate for each target audience. Id. If the content creator chooses
only a single target audience, the system creates three video streams—one at
the target bit rate, one at approximately 75% of the target rate, and one at no
more than 50% of the target rate that only contains key frames. Id.
The 75% and 50% bit rate streams provide alternate streaming experiences
for those users experiencing bad or very bad connections, respectively. Id.
If the content creator chooses multiple target audiences, the system attempts
to nest the bit rates created so that the fallback stream for one target
audience is the target bit rate for another. Id.
IPR2019-00833
Patent 7,974,200 B2
10
2. Chou
Chou discloses a server that provides at least two different data
streams, including a low resolution stream encoded below a transmission bit
rate and a normal resolution stream encoded at the transmission bit rate.
Ex. 1011, 3:29–33. The server initially transmits the low resolution stream
to a client, which decodes and presents the stream in real time. Id. at 3:33–
37. The client does not need to wait for its buffer to fill before beginning to
decode, thereby reducing delay due to buffering to nearly zero. Id. at 3:38–
46. When the client buffer has grown sufficiently large, the server halts
transmission of the low resolution stream and transmits the normal
resolution stream. Id. at 3:47–51.
This process is depicted graphically in Chou’s Figure 9, reproduced
below.

Figure 9 illustrates a graph representing a method of constructing an
encoded bit stream by splicing together a low quality initial portion of
content with a normal quality subsequent portion of the content. Id.
at 10:35–39. In Figure 9, the server transmits the low bit rate encoding until
its schedule intersects a full bit rate encoding schedule, and then the server
transmits the full bit rate encoding. Id. at 11:22–24; see also id. at Fig. 8
(superimposing a low bit rate encoding and a full bit rate encoding); id.
at 11:10–13 (describing Figure 8).
IPR2019-00833
Patent 7,974,200 B2
11

3. Muroi
Muroi discloses a video transmission device and receiver that operate
to address decreases in video delivery speed caused by changes in traffic on
the delivery path. Ex. 1013 ¶ 1. Muroi’s transmission device stores video at
three separate data rates. Id. ¶ 16. The transmission device transmits video
to the receiver at an arbitrarily selected rate. Id. ¶ 20. The receiver stores
the received video in a buffer and takes video data out from the buffer to
reproduce video. Id. ¶¶ 20–21. Figure 2 of Muroi, reproduced below,
depicts an example buffer.

Figure 2 shows data volume in a buffer memory and reference values. Id.
at 4–5. When the volume of data stored in the buffer falls below a first
reference value (a) (see Figure 2), the receiver requests the transmitter to
switch to a lower video rate. Id. ¶ 24. Conversely, when the volume of data
stored in the buffer exceeds a second reference value (b) (see Figure 2), the
receiver requests the transmitter to switch to a higher video rate. Id.
4. Lee
Lee discloses a transport-layer congestion control algorithm that
IPR2019-00833
Patent 7,974,200 B2
12
adjusts a video transmission rate in a TCP-friendly manner. Ex. 1014, 747,
750–51. Lee describes a typical Internet congestion control algorithm, linear
increase multiplicative decrease (LIMD). Id. at 750. LIMD changes the
sending rate of a connection by slowly increasing the transmission rate by an
additive constant α upon observing no packet loss and aggressively
decreasing the transmission rate by a multiplicative constant β upon
observing packet losses. Id. Lee proposes an augmented LIMD algorithm,
termed LIMD/H, which uses the history of packet losses in order to
distinguish between persistent congestion and non-persistent congestion. Id.
The LIMD/H algorithm reacts gently to non-persistent congestion and
aggressively to persistent congestion. Id.
E. Patentability Analysis
1. Challenges Applying Lippman, Chou, and Muroi
Petitioners contend that claims 1–4, 7, and 10 are obvious over
Lippman, Chou, and Muroi under 35 U.S.C. § 103(a). Pet. 25–70. For the
reasons that follow, we are persuaded that Petitioners establish, by a
preponderance of the evidence, that claims 1–3, 7, and 10 are unpatentable
as obvious over Lippman, Chou, and Muroi. We are not, however,
persuaded that Petitioners establish, by a preponderance of the evidence, that
claim 4 is unpatentable as obvious over Lippman, Chou, and Muroi.
a) Independent Claim 1
For claim 1, Patent Owner does not dispute that the references
disclose all the claim limitations. Instead, Patent Owner challenges
Petitioners’ rationale to combine the references. PO Resp. 27–54. We first
address Petitioners’ contentions as to the limitations of claim 1 and then
address Patent Owner’s arguments. We find Petitioner’s contentions as to
IPR2019-00833
Patent 7,974,200 B2
13
the limitations of claim 1, as detailed below and not disputed by Patent
Owner, persuasive.
(1) “A method of operating [a] a real-time communication
apparatus comprising a real-time data sender, [b] a real-
time data display device having [c] a store, and [d] a
network connecting said sender and said display device.”
Petitioners contend that each of Lippman and Chou discloses the
preamble of claim 1.8 Pet. 29–30 (citing Ex, 1008 § IX.B.1). Specifically,
Petitioners contend Lippman discloses “delivery of streaming video from [a]
a server to [b] a player over [d] the Internet” and “a ‘pre-roll’ amount of
data, . . . which teaches use of [c] a data store at the data presentation
device.” Id. at 29 (citing Ex. 1010, 780). Petitioners also assert that
Lippman describes applications to “real-time encoding” and to “live
content.” Id. (citing Ex. 1010, 780–81). Similarly, Petitioners assert Chou
“teaches a communication system 200 comprising: [a] a data sender 204
. . . , [b] a data display device 208 . . . having [c] a store 308 . . . , and [d] a
network 206 . . . connecting said sender and said display device.” Id. at 29–
30 (citing Ex. 1011, Figs. 2, 3, 6:34–7:8, 7:60–63, 8:25). Petitioners also
assert that Chou’s “data sender and data display devices are real-time
devices.” Id. at 30 (citing Ex. 1011, 7:43–55 (referring to real-time video
sources and live transmission)).

8 We take no position as to whether the preamble is limiting. Regardless, we
find Petitioner has shown that the recitation in the preamble is satisfied by
the prior art.
IPR2019-00833
Patent 7,974,200 B2
14
(2) “operating said sender to transmit a plurality of first-
encoding-rate data packets [a] encoded at a first encoding
rate and [b] representing a first part of a real-time
presentation to said display device [c] at a transmission
rate higher than said first encoding rate of said first-
encoding-rate data packets”
Petitioners contend that Chou discloses this limitation. Id. at 31–32
(citing Reader Decl. § IX.B.2). Petitioners assert Chou discloses:
A server provides at least two different data streams. A first
data stream is a low resolution stream encoded at a bit rate
below the transmission bit rate. . . . The server initially
transmits the low resolution stream faster than real time, at a bit
rate equal to the transmission bit rate.

Id. (citing Ex. 1011, 4:42–48, 11:6–7, Fig. 7); see also id. at 30 (asserting
Chou discloses “real time video sources and live transmission”).
(3) “operating said display device to: receive said first-
encoding-rate data packets into said store”
Petitioners contend that each of Lippman, Chou, and Muroi discloses
this limitation. Id. at 33 (citing Reader Decl. § IX.B.4). Petitioners contend
Lippman discloses this limitation because Lippman’s “player collects a
certain minimum amount of packets (specified by a pre-roll amount) and
then starts to decode the packets and combine the decoded results into a
multimedia presentation.” Id. (citing Ex. 1010, 780). Petitioners contend
Chou discloses this limitation because Chou discloses “decoder buffer 308”
that “receives encoded data from the network 306.” Id. (citing Ex. 1011,
7:60–63, 12:27–37). Petitioners also contend Muroi discloses this limitation
because Muroi teaches that “buffer memory 22 temporarily stores received
video data.” Id. (citing Ex. 1013 ¶ 20).
(4) “operating said display device to: . . . read-out said
received first-encoding-rate data packets from said store
IPR2019-00833
Patent 7,974,200 B2
15
[a] at a data rate equal to said first encoding rate of said
first-encoding-rate data packets [b] for decoding [c] so as
to present said first part of said real-time presentation to a
user at a first level of quality”
Petitioners contend that Chou discloses this limitation based on
Figures 3 and 4D of Chou, reproduced below with annotations added by
Petitioners:

Chou, Fig. 3 (annotations added by Petitioners)

Chou, Fig. 4D (annotations added by Petitioners)
Id. at 34–35 (citing Reader Decl. § IX.B.5). Figure 3 depicts points A, B, C,
and D in a streaming media system, and Figure 4D depicts data flow
schedules corresponding to points A and D in Figure 3. Ex. 1011, 7:41,
7:66–8:8, 8:41–47. Petitioners assert Chou discloses that “schedule A is
encoding rate and schedule D is read out rate.” Pet. 35 (citing Ex. 1011,
7:66–8:5). Petitioners contend that schedule A and schedule D represent
equal rates. See id. at 34–35; Ex. 1008 ¶ 215.
IPR2019-00833
Patent 7,974,200 B2
16
(5) “on said store being filled with said first-encoding-rate
data to a predetermined level, sending an indication that
said level has been reached to said data sender”
Petitioners contend that Muroi discloses this limitation. Pet. 35–36
(citing Ex. 1008 § IX.B.6). Specifically, Petitioners reproduce Figure 2 of
Muroi as annotated below:

Id. at 36 (annotations in Petition). Figure 2 depicts Muroi’s buffer at the
video receiver device. Ex. 1013 ¶¶ 18, 20. Petitioners contend Muroi
discloses that “when a buffer is filled with first-encoding-rate data to
threshold (b), a display device 2 sends a rate switch request . . . to the data
sender 1 . . ., which indicates that the threshold has been reached.” Pet. 36
(citing Ex. 1013 ¶¶ 6, 7, 24).
(6) “operating said data sender, [a] on receipt of said
indication from said display device, to send second-
encoding-rate data packets [b] encoded at a second
encoding rate and [c] representing subsequent parts of said
real-time presentation to said display device, [d] said
second encoding rate of said second-encoding-rate data
packets being higher than said first encoding rate of said
first-encoding-rate data packets”
Petitioners contend that Chou discloses this limitation. Id. at 36–38
(citing Ex. 1008 § IX.B.7). As to parts [a]–[c] of this limitation, Petitioners
contend Chou discloses those parts because Chou teaches: “When the client
buffer has grown sufficiently large to guard against future underflow by the
normal resolution stream, the server stops transmission of the low resolution
IPR2019-00833
Patent 7,974,200 B2
17
stream and begins transmission of the normal resolution stream.” Id. at 37
(citing Ex. 1011, 4:51–55). As to parts [b] and [d] of this limitation,
Petitioners point to an annotated version of Figure 5 of Chou as depicted
below:

Id. Figure 5 illustrates a decoding schedule for multimedia content pre-
encoded at a full bit rate. Ex. 1011, 4:7–8. According to Petitioners, Figure
5 depicts a “decoding schedule for full bit rate content,” which “teaches that
the second-encoding-rate data packets are [b] encoded at a second encoding
rate, [d] wherein said second encoding rate is higher than said first encoding
rate.” Pet. 37–38 (citing Ex. 1011, 10:64–65).
(7) “operating said display device to: receive [a] the second-
encoding-rate data packets [b] representing the subsequent
part of real-time presentation into said store”
Petitioners contend that each of Lippman, Chou, and Muroi discloses
this limitation. Id. at 38–39 (citing Ex. 1008 § IX.B.9). Petitioners contend
Lippman discloses these features because Lippman’s “player stores packets
into a pre-roll buffer even after stream switching occurs.” Id. at 38 (citing
Ex. 1010, 781). Petitioners contend Chou’s “decoder buffer 308 receives
encoded data from the network,” including a “second plurality of second-
encoding-rate data packets.” Id. at 39 (citing Ex. 1011, 7:60–63, 12:27–37).
In addition, Petitioners contend Muroi’s “buffer memory 22 temporarily
stores received video data.” Id. (citing Ex. 1013 ¶ 20).
IPR2019-00833
Patent 7,974,200 B2
18
(8) “operating said display device to: . . . read out said
second-encoding-rate data packets from said store [a] at a
data rate equal to said second encoding rate of said
second-encoding-rate data packets [b] for decoding [c] so
as to present said subsequent part of said real-time
presentation to said user [d] at a second level of quality
which is higher than said first level of quality”
Petitioners contend that Chou discloses this limitation, relying on
similar arguments as for similar limitations in element [1.4], discussed
above. Id. at 39–42 (citing Ex. 1008 § IX.B.10). In addition, Petitioners
point to Chou’s disclosure that “the second data stream is sent to improve
the quality as soon as the playback buffer can handle it.” Id. at 41 (citing
Ex. 1011, 12:32–37).
(9) Summary
As we noted above, Patent Owner does not dispute that the references
disclose all the claim limitations. We are persuaded by Petitioners’
arguments and cited evidence, as described above, and accordingly, we find
that the combination of Lippman, Chou, and Muroi discloses each and every
limitation of claim 1.
(10) Rationale to Combine
Petitioners assert that a person of ordinary skill in the art would have
modified the operation of Lippman “to perform fast start-up operations as
taught by Chou and to switch streams according to buffer status indications
as taught by Muroi.” Id. at 25 (citing Ex. 1008 § IX.A).
With respect to fast start-up, Petitioners point to Lippman’s player
collecting “pre-roll” packets, which result in pre-roll times of 1.7 to 5.8
seconds. Id. at 26 (citing Ex. 1010, 780, 782). Petitioners then point to
Chou’s disclosure that describes “start-up delays of 2–10 seconds” as
“annoying” and “intolerable.” Id. (citing Ex. 1011, 3:8–19). According to
IPR2019-00833
Patent 7,974,200 B2
19
Petitioners, Chou “proposes solutions to avoid such delays” and also
“identifies RealNetworks’ RealAudio and RealVideo players,” which
Petitioners assert are the subject of Lippman. Id. Petitioners assert,
therefore, that a person of ordinary skill in the art would have applied
Chou’s techniques in Lippman’s architecture to avoid these “annoying”
delays. Id. Specifically, Petitioners assert that “a skilled worker would
cause Lippman’s system to fall back to a lower-rate coding, for example a
‘28.8 Modem’ coding when starting streaming to a ‘56K Modem’ target
audience member.” Id. at 26–27.
With respect to stream switching, Petitioners assert, “both Lippman
and Muroi prepare multiple codings of video at different coding rates and
switch among them to match changing bandwidth.” Id. at 27. Petitioners
assert that “Lippman does not explain how bandwidth is measured in his
system” but that “Muroi explains that switches between different data-rate
encodings can be accomplished by measuring a volume of data stored in the
receiver’s storage buffer.” Id. Petitioners, therefore, assert that a “skilled
worker would recognize readily that Muroi’s switching technique finds
application with a system such as Lippman’s.” Id. at 27–28 (citing KSR, 550
U.S. at 418 (“[I]f a technique has been used to improve one device, and a
person of ordinary skill in the art would recognize that it would improve
similar devices in the same way, using the technique is obvious unless its
actual application is beyond that person’s skill.”)); see also Ex. 1008
¶¶ 148–150 (“Lippman . . . does not explain how such measurements are to
be performed” but “Muroi also discloses techniques for switching between
redundant codings of video in response to changing bandwidth conditions of
a network.”).
IPR2019-00833
Patent 7,974,200 B2
20
Patent Owner makes several arguments attacking the Lippman, Chou,
Muroi combination, supported by the testimony of Dr. Nathaniel Polish,
which we address in detail below. PO Resp. 28–53; Ex. 2001 (“Polish
Decl.”), Ex. 2002 (“2nd Polish Decl.”).
(a) Modifying Lippman with Muroi
Patent Owner first contends it would not have been obvious to modify
Lippman to implement Muroi’s technique for determining when to switch
between different versions of video. PO Resp. 27–28. Specifically, Patent
Owner contends that one of ordinary skill in the art would not have modified
Lippman with Muroi “for the purposes of solving a problem that is already
solved by” Lippman. Id. at 28 (citing Ex parte Rinkevich, 2007 WL
1552288 (BPAI May 29, 2007) (nonprecedential)).
Patent Owner contends that in Rinkevich, the primary reference
identified a commercially-available product that solved a user authentication
problem and identified where to find that product online, without providing
specific details of how to solve the problem. Id. at 30. According to Patent
Owner, the secondary reference in Rinkevich disclosed specific details of
how to solve the problem. Id. at 29. In Rinkevich, Patent Owner asserts, the
Board determined there was no motivation to combine the references
because “the problem [to be solved by the secondary reference] is already
solved by [the primary reference].” Id. at 30 (quoting Rinkevich, 2007 WL
1552288, at *4). Patent Owner asserts that like the primary reference in
Rinkevich, Lippman identifies a commercially-available product
(“RealSystem’s SureStream”) for switching video versions and where to find
the product online without specifying how the product switches video
versions. Id. (citing Ex. 1010, 780). Patent Owner contends Muroi is like
the secondary reference in Rinkevich that disclosed specific details for
IPR2019-00833
Patent 7,974,200 B2
21
solving a user authentication problem because Muroi discloses specific
details for determining how to switch video versions. Id. at 30–31.
Accordingly, Patent Owner contends that “a [person of ordinary skill in the
art] having common sense at the time of the invention would not have
reasonably looked to [Muroi] to solve a problem already solved by
[Lippman].” Id. at 31 (quoting Rinkevich, 2007 WL 1552288, at *4).
In the Reply, Petitioners contend that the present facts differ from
those in Rinkevich and instead resemble those in Unwired Planet, LLC v.
Google Inc., “where a primary reference (Brohoff) provided search results
that grouped service providers by location within a geographic zone, but did
not explain how service providers are prioritized within those zones.” Pet.
Reply 10 (citing Unwired Planet, LLC v. Google Inc., 841 F.3d 995, 1003–
1004 (Fed. Cir. 2016)). According to Petitioners, the “Federal Circuit
affirmed the Board’s holding [in Unwired Planet] that it would have been
obvious to apply teachings of a secondary reference (Galitz), which taught
different ordering techniques and the benefits of using them to the Brohoff
system.” Id. Petitioners contend the facts here are analogous to Unwired
Planet because Lippman’s system measures available bandwidth without
explaining how to do so and because Muroi teaches how to measure
bandwidth. Id. at 11. Accordingly, Petitioners contend that it would have
been obvious to combine the two references. Id.
Patent Owner’s Sur-Reply argues Unwired Planet is inapposite
because it refers to prior art references that disclosed “different,
complementary features.” Sur-Reply 8 (emphasis omitted). In contrast,
Patent Owner contends, the facts here correspond more closely with
Rinkevich because “Lippman and Muroi disclose the same feature”—video
stream switching. Id. at 8–9 (citing Ex. 1010, 780; Ex. 1013 ¶¶ 23–24).
IPR2019-00833
Patent 7,974,200 B2
22
Turning to our analysis, we first note that Patent Owner’s exclusive
reliance on Rinkevich is misplaced. The Board’s decision in Rinkevich is
nonprecedential and, therefore, not binding. See PTAB Standard Operating
Procedure (SOP) 2 (Rev.10), available at http://www.uspto.gov/patents-
application-process/patent-trial-and-appeal-board/resources. Rinkevich also
differs from this case in at least two significant respects—in burden of proof
and level of detail. First, Rinkevich is an ex parte appeal decision, which
involves a different prima facie burden than Petitioners’ preponderance of
the evidence burden here. See, e.g., Manual of Patent Examining Procedure
(“MPEP”) § 2142 (describing the burden-shifting framework of the prima
facie case in examination); 35 U.S.C. § 316(e) (“In an inter partes review
instituted under this chapter, the petitioner shall have the burden of proving a
proposition of unpatentability by a preponderance of the evidence.”).
Second, in Rinkevich, the Examiner’s rationale to combine was terse, but
here, Petitioners supply detailed arguments with expert testimony to support
Petitioners’ reasoning. Compare Rinkevich, 2007 WL 1552288, at *4 (“The
Examiner asserts that the nature of the problem to be solved would have led
an artisan, having knowledge of Savill, to look to Wu to solve the purported
deficiencies of Savill.”) (emphasis omitted), with Pet. 25–29; Ex. 1008
¶¶ 137–160. Consequently, we view Rinkevich as having limited
applicability to the facts of this case.
Far more pertinent to our decision are the principles of obviousness
articulated by our reviewing courts, including the Supreme Court’s KSR
decision. There, the Supreme Court explained that a rationale to combine
may arise from various sources, including the “interrelated teachings of
multiple patents.” KSR, 550 U.S. at 418. The Court also stated, “if a
technique has been used to improve one device, and a person of ordinary
IPR2019-00833
Patent 7,974,200 B2
23
skill in the art would recognize that it would improve similar devices in the
same way, using the technique is obvious unless its actual application is
beyond his or her skill.” Id. at 417.
We also consider the Federal Circuit’s guidance in Unwired
Planet, stating that “[f]or [a] technique’s use to be obvious, the skilled
artisan need only be able to recognize, based on her background knowledge,
its potential to improve the device and be able to apply the technique.” 841
F.3d at 1003. The primary reference in Unwired Planet (Brohoff) described
a system that allowed users to search for information and provide results
based on their location, but it did not explain how those search results were
ordered. Id. The secondary reference (Galitz) described various ordering
techniques and their benefits. Id. The court concluded these were
techniques “that a person of ordinary skill in the art would recognize could
be used to improve [Brohoff’s] information display.” Id. Expert testimony
supported this conclusion. Id. The patentee also “d[id] not dispute that a
person of ordinary skill in the art would have been able to apply the
teachings of Galitz to make the improvement it suggests.” Id.
We agree with Petitioners that this case is similar to Unwired Planet.
Like the primary reference in Unwired Planet, Lippman does not describe
certain details, such as how to switch streams based on measuring
bandwidth. See Ex. 1010, 780–781. Like the secondary reference in
Unwired Planet, Muroi offers a detailed solution, namely how to determine
when to switch video streams. See Ex. 1013 ¶¶ 23, 24. Also like the
secondary reference in Unwired Planet, Muroi describes the benefit of its
approach, including “realiz[ing] video reproduction with the highest picture
quality possible by changing the video rate dynamically according to the
state of the network at a given point in time.” Id. ¶ 30 (emphasis added).
IPR2019-00833
Patent 7,974,200 B2
24
We agree with Petitioners that one of ordinary skill in the art would have
looked to Muroi’s detailed disclosure to fill in details absent from Lippman’s
disclosure. See Pet. 27–28; Reply 10–11. We also find Muroi’s “highest
picture quality” benefit would have led one of ordinary skill to combine
Muroi’s teachings with Lippman because Lippman likewise expresses a
desire to “maximize[] the quality of a streaming multimedia experience
regardless of the vagaries of a user[’]s Internet connection.” See Ex. 1013
¶ 30; Ex. 1010, 782.
Petitioner’s contention is supported by the testimony of Dr. Reader,
which we credit. He testifies that “Lippman teaches that a switch between
redundant codings of video may be performed based on measurements of
available bandwidth, even though he does not explain how such
measurements are to be performed.” Ex. 1008 ¶ 148. Dr. Reader further
testifies that “Muroi also discloses techniques for switching between
redundant codings of video in response to changing bandwidth conditions of
a network” and “discloses how measurements are to be performed.” Id.
¶¶ 149–150. Dr. Reader also testifies about the benefits of Muroi’s
approach, which “enables a data sender to switch between streams in
response to fluctuating network delivery speeds.” Id. ¶ 151. Patent Owner
does not cite Dr. Polish’s testimony to support its arguments, nor does Patent
Owner dispute that one of ordinary skill in the art would have been able to
apply the teachings of Muroi to make the improvement it suggests. See
generally PO Resp.; Unwired Planet, 841 F.3d at 1003 (“For [a] technique’s
use to be obvious, the skilled artisan need only be able to recognize, based
on her background knowledge, its potential to improve the device and be
able to apply the technique.”); KSR, 550 U.S. at 417 (“using the technique is
obvious unless its actual application is beyond that person’s skill.”).
IPR2019-00833
Patent 7,974,200 B2
25
Accordingly, Petitioners persuade us that one of ordinary skill in the art
would have looked to Muroi to solve the problem of missing details in
Lippman’s stream switching solution.
Contrary to Patent Owner’s assertion, the fact that Lippman provides
a link to download its player would not have led one of ordinary skill in the
art away from combining with Muroi. See PO Resp. 30 (arguing against the
combination because “Lippman identifies a commercially-available product
that determines when to switch between different versions of video”).
Lippman merely states, “[f]or those interested in experimenting further we
encourage the download of our freely available RealProducer and
RealPlayer™ available at www.real.com.” Ex. 1010, 781. Nothing in
Lippman or in any other evidence of record suggests that this download link
is for anything more than downloading a functioning player. Significantly,
Lippman does not state that the download link includes any additional
details on how to implement stream switching. We find the mere mention of
Lippman’s download link would not have discouraged one of ordinary skill
in the art from combining Lippman’s teachings with Muroi’s disclosure.
Based on the preponderance of the evidence of record, we agree with
Petitioner that the skilled artisan would have combined Lippman and Muroi
as proposed. For the reasons explained above, Patent Owner’s arguments to
the contrary above are unavailing. We address Patent Owner’s additional
arguments against combining Muroi with Lippman in the following section.
(b) Modifying Lippman with Chou and Muroi
Patent Owner next contends that modifying Lippman with Chou and
Muroi would have changed Lippman’s principle of operation. PO Resp. 32–
44. Patent Owner raises three main arguments: (1) “modifying Lippman to
transmit low quality video at startup as taught in Chou would change
IPR2019-00833
Patent 7,974,200 B2
26
Lippman’s principle of operation,” (2) “modifying Lippman to transmit low
quality video at startup as taught in Chou would frustrate a main goal of
Lippman,” and (3) “modifying Lippman to use Muroi’s client-side control
technique would further change the principle of operation of Lippman.” Id.
at 36 (citing Ex. 2001 ¶¶ 173–187; Ex. 2002 ¶¶ 6–78), 32 (citing In re Ratti,
270 F.2d 810 (C.C.P.A. 1959); Ex parte Böger, Appeal 2017-001586 (PTAB
Aug. 31, 2017); Ex parte Salinas, Appeal 2013-00433 (PTAB Jan. 22,
2016); Ex parte Mandrusov, Appeal 2008-0114 (BPAI, July 23, 2008); Plas-
Pak Indus., Inc. v. Sulzer Mixpac AG, Appeal 2013–007786 (PTAB Jan. 17,
2014) (appeal from inter partes reexamination), aff’d, 600 Fed. Appx. 755,
759–60 (Fed. Cir. 2015) (nonprecedential)). The next section addresses
Patent Owner’s first two arguments, and the following section addresses
Patent Owner’s third argument.
(i) Whether Chou changes the principle of operation
and frustrates a main goal of Lippman
We begin by summarizing Patent Owner’s first two arguments. First,
Patent Owner contends that modifying Lippman with Chou would change
Lippman’s principle of operation. PO Resp. 36–37. Specifically, Patent
Owner contends that Chou would change Lippman’s operation from “an
optimistic assumption of network conditions by initially transmitting the
highest quality video . . . that [a] client’s Internet connection will support” to
“a pessimistic assumption of network conditions by initially transmitting
lower-quality video to the client.” Id. at 36–37 (citing Ex. 2001 ¶¶ 177–
180). Patent Owner also contends that Chou’s “principle of operation at
startup (sacrificing video quality for faster startup) is the opposite of
Lippman’s principle of operation at startup (sacrificing faster startup for
higher video quality).” Id. at 37 (citing Ex. 2002 ¶ 49).
IPR2019-00833
Patent 7,974,200 B2
27
Second, Patent Owner contends that Chou would “frustrate Lippman’s
main goal of maximiz[ing] the quality of a streaming multimedia experience
regardless of the vagaries of a [user’s] Internet connection.” Id. at 38 (citing
Ex. 1010, 782). By transmitting low quality video at startup, Patent Owner
contends, “Chou would reduce rather than maximize the quality of the
streaming multimedia experience.” Id. (citing Ex. 2002 ¶¶ 56–62). Patent
Owner contends that although Chou seeks to avoid start-up delays,
Lippman’s pre-roll buffer is a “primary feature” rather than a shortcoming
“because buffering pre-roll is critical to Lippman’s stated goal” of
maximizing streaming quality. Id. (citing Ex. 1010, 782; Ex. 2001 ¶¶ 92–
93; Ex. 2002 ¶ 40). Thus, Patent Owner contends, “transmitting lower-
quality video at startup to avoid buffering pre-roll reduces the quality of the
user’s video experience rather than maximizes the quality of the user’s video
experience, especially when the user’s Internet connection would support a
higher-quality video stream.” Id. at 38–39 (citing Ex. 1010, 780–82;
Ex. 2002 ¶ 49).
Petitioners in their Reply contend motivation is not lacking to modify
Lippman because neither Lippman nor Chou “describes any ‘assumptions’
of network conditions.’” Reply 6. Rather, Petitioners contend, Lippman is
not “optimistic” but merely “measures available bandwidth and switches
between separate (non-layered) video encodings to match the channel
capacity.” Id. at 12 (quoting Ex. 1010, 780). Chou, Petitioners contend,
“does not make ‘pessimistic assumptions’ regarding network conditions and,
instead, it estimates the state of playback buffers at client devices.” Id.
(citing Ex. 1011, Abstract, 3:47–51, 10:52–11:29). Petitioners further
contend, “[a]lthough applying Chou’s technique would change Lippman’s
system—as all obvious combinations do—a skilled worker would recognize
IPR2019-00833
Patent 7,974,200 B2
28
that it would change Lippman’s system for the better.” Id. Specifically,
Petitioners contend that “[a] skilled worker would be motivated to apply
Chou’s technique to Lippman’s system . . . to avoid the annoying latencies
represented by Lippman’s prolonged buffer times.” Id. at 13.
Petitioners also contend that Chou would not frustrate Lippman’s
goals, asserting “Lippman never states that the quality of video at initial
playback must be maintained at the highest quality, or that he sacrifices
faster startup for higher video quality, as [Patent Owner] suggests.” Id.
at 14. Rather, Petitioners contend, “Lippman focuses on maintaining the
quality of streaming experience overall.” Id. (citing Ex. 1010, 780).
Petitioners also contend that Patent Owner “impl[ies] incorrectly that
Lippman requires buffered video be coded at the highest possible coding rate
at startup.” Id. Petitioners assert, “Lippman teaches to apply pre-roll at a
variety of coding rates, not just the highest coding rate possible” because
Lippman’s pre-roll buffer merely collects “a certain minimum amount of
packets” and permits a presentation to “continue seamlessly until the player
is stopped or the content finishes.” Id. at 14–15 (citing Ex. 1010, 780–81;
Ex. 1011, 2:32–34). Moreover, Petitioners contend, “Lippman invites others
to find new ways to improve streaming experiences” by stating that “[o]ur
goal is to present one approach to this problem . . . in the hopes of
encouraging progress in the development of the best possible video
streaming experience over the Internet.” Id. at 15 (quoting Ex. 1010, 780).
Petitioners conclude, “[a] skilled worker . . . would recognize that Chou’s
fast startup technique represents further progress in streaming experiences
by avoiding . . . ‘intolerable’ startup delays” and therefore “promote[s]
Lippman’s goals, rather than frustrate[s] them.” Id. (citing Ex. 1011,
Abstract, 3:8–18, 3:29–46, 12:27–36).
IPR2019-00833
Patent 7,974,200 B2
29
In its Sur-Reply, Patent Owner responds that “Dr. Polish never opined
that the systems in Lippman or Chou affirmatively ‘make’ any assumption
as Petitioners contend”; rather, according to Patent Owner, Dr. Polish
explained that Lippman’s and Chou’s protocols are “based on” optimistic
and pessimistic assumptions, respectively. Sur-Reply 4–5 (citing Reply 6–7;
Ex. 2001 ¶¶ 102, 175, 180). Patent Owner also disputes that Chou’s
approach would change Lippman for the better, as “[t]he issue is not whether
Chou would improve Lippman as Petitioners’ allege, but whether modifying
Lippman with the teachings of Chou would change Lippman’s principle of
operation.” Id. at 10.
Further, Patent Owner disagrees that “Lippman never states that the
quality of video at initial playback must be maintained at the highest
quality” because Lippman “maximiz[es] bandwidth usage,” and
“[t]ransmitting the highest quality of video that the client’s Internet
connection will support (including at startup) maximizes the bandwidth
usage.” Id. at 11–12 (citing Reply 13–14; Ex. 1010, 780) (emphasis
omitted). Patent Owner also contends that “Lippman’s principle of
operation at startup (sacrificing faster startup for higher video quality) is the
exact opposite of Chou’s principle of operation at startup (sacrificing video
quality for faster startup).” Id. at 12 (citing Ex. 2002 ¶ 49). Finally, Patent
Owner contends that “buffering pre-roll is a primary feature of Lippman—
not a shortcoming.” Id. at 13 (citing PO Resp. 38–39; Ex. 2001 ¶¶ 92–93;
Ex. 2002 ¶ 40).
Turning to our analysis, “[w]e start from the self-evident proposition
that mankind, in particular, inventors, strive to improve that which already
exists.” Pro-Mold and Tool Co., Inc. v. Great Lakes Plastics, Inc., 75 F.3d
1568, 1573 (Fed. Cir. 1996). “Under the correct analysis, any need or
IPR2019-00833
Patent 7,974,200 B2
30
problem known in the field of endeavor at the time of invention and
addressed by the patent can provide a reason for combining the elements in
the manner claimed.” KSR, 550 U.S. at 420. Moreover, “[a] person of
ordinary skill is also a person of ordinary creativity, not an automaton.” Id.
at 421. With respect to combining Chou with Lippman, Petitioners provide
evidence to show that a person of ordinary skill in the art would have been
motivated to apply Chou to solve a problem with Lippman’s approach.
Specifically, we credit Dr. Reader’s testimony that Chou recognizes start-up
delays, such as Lippman’s delay due to pre-roll buffering, are “annoying”
and “intolerable,” and that a person of ordinary skill would have “appl[ied]
Chou’s techniques to . . . Lippman’s to avoid such delays.” Ex. 1008 ¶ 147.
We find Patent Owner’s arguments and evidence to the contrary unavailing.
Patent Owner relies on In re Ratti and several of the Board’s
nonprecedential decisions in ex parte appeals to support Patent Owner’s
argument that if a secondary reference changes the principle of operation or
frustrates a main goal of the primary reference, the secondary reference
cannot provide a motivation to combine. PO Resp. 32 (citing Ratti, 270
F.2d at 810; Ex parte Böger, Appeal 2017-001586; Ex parte Salinas, Appeal
2013-00433; Ex parte Mandrusov, Appeal 2008-0114; Plas-Pak, Appeal
2013–007786). These cases more accurately support that the issue is
whether the combination would impermissibly change the principle of
operation of the references or impermissibly frustrate a primary purpose of
them. For example, a skilled artisan has no reason to combine references in
a way that renders a device inoperable. Cf. In re Gordon, 733 F.2d 900, 902
(Fed. Cir. 1984) (one of ordinary skill in the art would not be motivated to
modify a reference by making it “inoperable for its intended purpose”);
McGinley v. Franklin Sports, Inc., 262 F.3d 1339, 1354 (Fed. Cir.
IPR2019-00833
Patent 7,974,200 B2
31
2001) (noting that producing a “seemingly inoperative device” is a type of
teaching away from a combination). Even if the proposed combination
would diminish the primary reference’s performance, but not make it
inoperable, the diminished performance would merely be one of several
competing considerations in the obviousness analysis. See Winner Int’l
Royalty Corp. v. Wang, 202 F.3d 1340, 1349 n.8 (Fed. Cir. 2000) (“The fact
that the motivating benefit comes at the expense of another benefit . . .
should not nullify its use as a basis to modify the disclosure of one reference
with the teachings of another. Instead, the benefits, both lost and gained,
should be weighed against one another.”).9
Patent Owner contends that Chou would “change Lippman’s principle
of operation” or “frustrate a main goal of Lippman.” See, e.g., PO Resp. 36.
But Patent Owner’s arguments amount to asserting that Chou would merely
diminish Lippman’s performance. Even if it were true that modifying
Lippman with Chou would change Lippman’s operation from “an optimistic
assumption of network conditions” to “a pessimistic assumption of network
conditions,” Patent Owner does not argue that such a change would render
Lippman inoperable or even that such a change would not be
counterbalanced by motivating benefits. See PO Resp. 36–37 (citing Ex.
2001 ¶¶ 177–180). Likewise, Patent Owner contends that Lippman
maximizes streaming quality and that Chou’s fast startup would merely
reduce that quality. PO Resp. 38 (citing Ex. 1010, 782). And although
Patent Owner contends that “buffering pre-roll is critical to Lippman’s main
goal regarding maximizing streaming quality,” Patent Owner does not

9 We also note that the cases upon which Patent Owner relies involve
mechanical and electrical inventions that are not directly analogous to the
facts of this case.
IPR2019-00833
Patent 7,974,200 B2
32
contend that eliminating this pre-roll would sufficiently outweigh the
advantages that Petitioners identify based on the combination with Chou, to
undermine Petitioner’s obviousness analysis. See PO Resp. 38.
We find that the alleged different “assumptions,” “opposite principles
of operation,” reduction in quality, and eliminating buffering pre-roll are
merely competing considerations against Chou’s teaching that its fast startup
technique would avoid the “‘particularly annoying,’ ‘intolerable’ startup
delays” that may be present in systems such as Lippman’s. See Ex. 1008
¶¶ 144–147 (citing, e.g., Ex. 1011, 2:59–3:16). Weighing all these
competing considerations together, we are persuaded one of ordinary skill in
the art would have been motivated to use Chou’s fast startup technique in
Lippman’s system to avoid those intolerable delays, even at the possible
expense of some quality at start-up. We emphasize that any reduction in
streaming quality from the combination is merely a possibility, as it appears
to us that Chou’s fast playback would represent higher quality than no
playback at all during Lippman’s buffering period. Thus, we agree with
Petitioners that “[a]lthough applying Chou’s technique would change
Lippman’s system—as all obvious combinations do—a skilled worker
would recognize that it would change Lippman’s system for the better.”
Reply 12.
In summary, we find that replacing Lippman’s pre-roll buffering with
Chou’s fast startup would have been the very kind of change one of ordinary
skill in the art would have found obvious to make based on the evidence of
record before us, including the testimony of Dr. Reader which we find more
persuasive than the testimony of Dr. Polish. We, therefore, determine that
Petitioners have persuasively articulated a rationale to combine Chou with
Lippman, supported by the testimony of Dr. Reader.
IPR2019-00833
Patent 7,974,200 B2
33
(ii) Whether Muroi changes Lippman’s principle of
operation
Turning to Patent Owner’s third argument, Patent Owner contends
that modifying Lippman with Muroi would change Lippman from a
“centralized, server-controlled scheme” to a “distributed, client-controlled
scheme” for stream switching. PO Resp. 40 (citing Ex. 2001 ¶¶ 181–186).
Patent Owner first addresses a remark we made in our Institution
Decision. Id. There, we discussed Patent Owner’s Preliminary Response
argument that “Muroi’s stream switching feature would require an even
further fundamental change in Lippman’s principle of operation.” Decision
25 (quoting Prelim. Resp 43). At that stage, we disagreed with Patent
Owner that Lippman and Muroi were incompatible, in part, because “Dr.
Reader testifies that one of ordinary skill in the art would have recognized
that Lippman’s server would have become overloaded if it diagnosed
bandwidth individually for every client.” Id. at 26 (citing Ex. 1008 ¶¶ 105–
107). Patent Owner disputes Dr. Reader’s assertion. PO Resp. 40 (citing
Decision 26; Ex. 2001 ¶¶ 105–107). Patent Owner argues that Dr. Reader
addresses “a server becoming overloaded when encoding video,” not when
implementing Lippman’s switching scheme. Id. at 41 (citing Ex. 2002 ¶ 9).
Patent Owner cites another reference, Lu (Ex. 1029), as describing
Lippman’s server-controlled scheme as “effective and requir[ing] only a
little server involvement.” Id. (citing Ex. 1029 § 3.2.1; id. n. 22 (Lippman)).
In light of Lippman’s “little server involvement” and Lu’s teachings, Patent
Owner contends that one of ordinary skill in the art would not have been
motivated to modify Lippman with Muroi’s teachings “for the purpose of
reducing the risk of video server overload caused by ‘increased processing
and I/O demands.’” Id. at 42 (citing Ex. 2002 ¶ 14; Ex. 1029 § 3.2.1).
IPR2019-00833
Patent 7,974,200 B2
34
Patent Owner also asserts that one of ordinary skill in the art would
not have “replac[ed] Lippman’s centralized, server-controlled scheme with
Muroi’s distributed, client-controlled scheme” because “a client-controlled
approach is not a guaranteed improvement over a server-controlled
approach.” Id. at 43. Instead, Patent Owner contends, in the client-
controlled approach, “a minority of ‘greedy’ clients can monopolize
bandwidth . . . and starve other clients of network bandwidth.” Id. at 43
(citing Ex. 2002 ¶¶ 19–24). Patent Owner asserts that “a video provider
would prefer an outcome where most clients receive ‘good’ quality video
rather than . . . where a few ‘greedy’ clients receive the ‘best’ quality video
while the rest of the clients receive ‘poor’ quality video (or perhaps no video
at all).” Id. at 43–44 (citing Ex. 2002 ¶¶ 25–29).
In the Reply, Petitioners contend that “Dr. Polish . . . contradicts
express disclosure from the prior art” when stating that Lippman’s video
server determines whether to switch streams. Reply 4 (citing Ex. 2001 ¶ 79,
185; Ex. 2002 ¶¶ 32, 38). Petitioners contend that Lippman actually
“explains that his player . . . controls the stream-switching process: ‘[T]he
player will switch to a lower or higher bandwidth stream when conditions
are appropriate.’” Id. (citing Ex. 1010, 781). Thus, Petitioners contend,
“Lippman expressly discloses that his player switches between streams, just
as Muroi’s player does, demonstrating that Dr. Polish’s opinions about the
basic facts are wrong.” Id. “In this regard,” Petitioners contend, “Lippman
and Muroi are alike because they both employ client-controlled stream-
switching techniques.” Id. at 16 (citing Ex. 1013 ¶¶ 23–24). Petitioners
further argue that Patent Owner’s “argument regarding the purported
location of switching ‘control’ is a red herring” because the “proper question
is whether it would have been obvious for a client to send ‘an indication that
IPR2019-00833
Patent 7,974,200 B2
35
[a predetermined level of fullness] has been reached to said data sender.” Id.
(quoting Ex. 1001, claim 1).
Petitioners also contend that even if Lippman performed server-
controlled stream switching, “it would have been obvious for a server to
switch streams in response to a client’s buffer-fullness messages that
represent network bandwidth.” Id. (citing Ex. 1010, 780; Decision 26).
Petitioners contend that it is undisputed that Lippman does not explain how
to measure bandwidth, and Muroi and other references (including Lu)
“suggest having clients measure bandwidth.” Id. at 16–17 (citing Ex. 1029,
8; Ex. 1034, 33:22–24:9, 69:24–70:10; Ex. 1013 ¶¶ 21, 23–24). Petitioners
cite Lu as teaching that “[d]ynamic stream switching relies on ‘estimated
available bandwidth reported by a client.’” Id. at 17 (citing Ex. 1029, 8).
Petitioners also assert that “people of skill recognized that it is preferable to
have systems operate using information of actual device state rather than
predictive models.” Id. (citing Ex. 1008 ¶ 113). Thus, according to
Petitioners, even if Lippman uses a server to control stream switching,
“applying Muroi’s measurement techniques to Lippman’s system merely
reflects application of well-known features operating according to their
established functions.” Id. (citing KSR, 550 U.S. at 417–418).
Petitioners further dispute that “concerns over server loading would
dissuade a skilled worker from applying Muroi’s technique to Lippman’s
system.” Id. (citing PO Resp. 42). Petitioners contend that Patent Owner’s
“greedy client” argument is based on speculative expert testimony and
should be ignored, as “Dr. Polish cites nothing that corroborates his
explanation of a ‘greedy client’ phenomenon.” Id. at 17, 18, 7. Petitioners
also contend that Dr. Polish also fails to address Dr. Reader’s testimony on
IPR2019-00833
Patent 7,974,200 B2
36
“prevailing Internet techniques that limit client transmission rates over the
Internet.” Id. at 7 (citing Ex. 1008 ¶¶ 86–87; Ex. 2002).
In the Sur-Reply, Patent Owner contends that Petitioners “cherry-
pick[] ambiguous passages from Lippman and ignore[] other disclosures,”
arguing that Lippman’s statement about the player switching “does not
address which end decides/determines whether to switch, only that the
switch is made.” Sur-Reply 3. Patent Owner relies on Lu to support its
argument that Lippman discloses server-controlled stream switching, as Lu
states “[t]he idea [in Lippman] is that the server can switch among multiple
streams” and it is “the server’s decision on whether it will respond.” Id.
at 3–4, 14 (citing Ex. 1029 § 3.2.1). Patent Owner also disputes that
Dr. Polish did not cite anything to support his explanation of a “greedy
client” phenomenon, as Dr. Polish cited three references in support. Id. at 5–
6 (citing Exs. 2006, 2007, 2008; Ex. 2002 ¶¶ 24–29; Ex. 1034, 99:9–
100:21).
Although the parties expend significant effort disputing whether
Lippman discloses server-centric or player-centric switching, we find it
unnecessary to resolve whether Lippman’s server or player controls stream
switching. Indeed, contradictory statements in the record make it difficult to
determine the extent to which Lippman’s server and player are involved with
stream switching.10 What is clear is that Lippman’s player provides at least
some input to the server for stream switching. Lippman states that “the
player will switch” streams “when conditions are appropriate,” and Lu

10 Lippman states that its “player will switch to a lower or higher bandwidth
stream,” which appears to describe a client-centric approach, but Lu states
that Lippman’s “server can switch among multiple streams.” Ex. 1010, 781;
Ex. 1029 § 3.2.1.
IPR2019-00833
Patent 7,974,200 B2
37
confirms11 that “the switching is dynamic, depending on the estimated
available bandwidth reported by a client,” that is, Lippman’s player. See
Ex. 1010, 781; Ex. 1029 § 3.2.1. Muroi’s client likewise provides input to
the server for stream switching. See Ex. 1013 ¶¶ 23–24. Muroi’s client
makes a rate switching request to the server “according to the state of the
network,” i.e., based on fluctuating network bandwidth. See Ex. 1013 ¶¶ 30,
21; Ex. 1008 ¶ 151. As both Lippman’s and Muroi’s players provide at least
some input to a server for stream switching, Patent Owner’s argument that
modifying Lippman with Muroi would change Lippman’s principle of
operation on this basis is unpersuasive. See PO Resp. 36, 40–44.
In any event, we agree with Petitioners that “even if” Lippman
performs server-controlled stream switching, “it would have been obvious
for a server to switch streams in response to a client’s buffer-fullness
messages that represent network bandwidth.” See Reply 16 (citing Ex.
1010, 780). Patent Owner contends that changing from a service-centric
approach to a client-centric approach would starve some clients’ bandwidth
based on a few “greedy” clients. See PO Resp. 43–44; Sur-Reply 5–7; Ex.
2002 ¶ 20 (testifying that Muroi’s “client-controlled architecture . . . may
result in other clients either (i) receiving very low quality video streams or
(ii) being unable to receive a video stream at all”). However, we find this
problem to be illusory because, according to Patent Owner, “[t]he idea [in
Lippman] is that the server can switch among multiple streams” and it is
“the server’s decision on whether it will respond.” Sur-Reply 3–4, 14
(citing Ex. 1029 § 3.2.1). As we understand it, Petitioners’ proposed

11 Neither party challenges Lu as mischaracterizing Lippman; to the
contrary, both parties rely on Lu. See, e.g., Pet. 14; PO Resp. 41; Reply 17;
Sur-Reply 4.
IPR2019-00833
Patent 7,974,200 B2
38
combination is not to bodily incorporate Muroi’s client-based control by
completely replacing whatever server control Lippman has; rather, it is to
incorporate Muroi’s “indication that [a predetermined level of fullness] has
been reached to [the] data sender” in Lippman. See Reply 16; Pet. 25–29,
35–36; In re Keller, 642 F.2d 413, 425 (CCPA 1981) (“The test for
obviousness is not whether the features of a secondary reference may be
bodily incorporated into the structure of the primary reference. . . . Rather,
the test is what the combined teachings of those references would have
suggested to those of ordinary skill in the art.”). Thus, even if Lippman’s
player using Muroi’s technique greedily requested too much bandwidth, in
the proposed combination Lippman’s server would still retain the decision
whether to switch streams. Because Lippman’s server would retain this
control, we find that Lippman’s server could avoid problems due to
potentially greedy clients.
Thus, we also find unpersuasive Patent Owner’s argument that Muroi
would change Lippman’s principle of operation on this basis.
(c) Chou and Muroi Rendering Lippman Unsatisfactory for
its Intended Purpose
Patent Owner additionally argues that “modifying Lippman with the
teachings of Chou and Muroi as Petitioners propose would make Lippman’s
system unsatisfactory for its intended purpose.” PO Resp. 45 (citing Ex.
2002 ¶¶ 30–78). This argument is largely redundant with Patent Owner’s
previous arguments.
Patent Owner contends modifying Lippman with Chou and Muroi
would reduce Lippman’s streaming quality as follows:
Modifying Lippman’s system with both Chou and Muroi
as proposed would render Lippman’s system unsatisfactory for
its intended purpose because the modified system would not
IPR2019-00833
Patent 7,974,200 B2
39
transmit the best-quality version of the video for a target
audience (e.g., a version encoded at the full bitrate for the target
audience) to a client device in that target audience even during
optimal network conditions.
. . . .
Even at startup, transmitting lower quality video to a user
in a target audience when the user’s Internet connection would
support a better quality version of the video is contrary to
Lippman’s intended purpose of “maximiz[ing] the quality of a
streaming multimedia experience regardless of the vagaries of a
[user’s] Internet connection.”
Id. at 48 (citing Ex. 2002 ¶¶ 56–58), 49 (citing Ex. 2002 ¶ 62).
Patent Owner further contends that Chou and Muroi would reduce
Lippman’s streaming quality for additional reasons:
Because Petitioners’ proposed system moves the function
of determining when to switch between video versions from the
server (as taught in Lippman) to the client (as taught in Muroi),
the video server would respond to the client’s rate switching
request by switching to a lower quality version of the video. . . .
And because the modified system implements Chou’s approach
of transmitting the lower quality video at the full transmission
rate at startup to fill the client buffer, the server would transmit
the third “keyframe only” version to the client at the
transmission rate for the normal-quality encoding of the video,
too.
. . . .
Immediately switching from the second “75% rate”
version at startup to the third “keyframe only” version of the
video (or from the “28.8 Modem” version to the “28.8 Modem
Duress” version, depending on the implementation) even
further diminishes the quality of the user’s experience, thereby
further frustrating Lippman’s intended purpose of
“maximiz[ing] the quality of a streaming multimedia
experience regardless of the vagaries of a [user’s] Internet
connection.”
IPR2019-00833
Patent 7,974,200 B2
40
Id. at 50 (citing Ex. 1013 ¶ 27; Ex. 2002 ¶¶ 65, 66), 51 (citing Ex. 2002
¶ 68). Finally, Patent Owner proposes additional scenarios (PO Resp. 51–
53) where a “user does not receive the best quality version of the video that
his or her Internet connection would otherwise support.” Id. at 53 (citing
Ex. 2002 ¶ 77). Patent Owner also contends that “[t]his sub-optimal
outcome is contrary to Lippman’s intended purpose of ‘maximiz[ing] the
quality of a streaming multimedia experience.’” Id. (citing Ex. 1010, 780–
782; Ex. 2002 ¶ 78).
Each of these arguments echoes Patent Owner’s argument above that
Chou would change Lippman’s principle of operation or frustrate Lippman’s
purpose by reducing Lippman’s quality in some way. None of Patent
Owner’s arguments in this section show any disadvantages purportedly
resulting from the combination that sufficiently undermines Petitioner’s
obviousness analysis, including the reasons why a person of ordinary skill
would have found the combination obvious (e.g., from advantages of the
combination). We, therefore, reject these arguments for reasons similar to
those stated above, as they also do not undermine Petitioner’s persuasive
showing.
(d) Summary
For the foregoing reasons, and based on our review of the complete
record, we are persuaded that Petitioners have established by a
preponderance of the evidence that claim 1 is rendered obvious by Lippman,
Chou, and Muroi.
b) Independent Claim 4
Independent claim 4 is similar to claim 1 in many respects, and
Petitioners provide similar arguments for similar portions of claims 4 and 1.
Pet. 44–57. However, claim 4 also recites “wherein said second plurality of
IPR2019-00833
Patent 7,974,200 B2
41
second-encoding-rate data packets are transmitted at a second transmission
rate which is higher than said first transmission rate,” which is not recited in
claim 1. Patent Owner contends that Petitioners have not shown this
limitation is taught or suggested by the asserted prior art. PO Resp. 54–57.
We agree with Patent Owner.
(1) The parties’ arguments
The Petition sets forth two theories for how the asserted prior art
discloses this limitation. Pet. 52–53. First, Petitioners contend Lippman
discloses this limitation based on Lippman’s video statistics table,
reproduced below with annotations added by Petitioners:

Id. (citing Reader Decl. § IX.E.7). Lippman describes the above table as
presenting aggregated statistics for two MPEG video clips. Ex. 1010, 781.
The Petition contends Lippman teaches that “playback should be seamless,
robust to missing packets and changing bandwidth conditions” and that
“stream switching [is] accomplished by ‘key frames.’” Id. at 53 (citing
Ex. 1010, 780–781).
The Petition’s second theory is that “[w]hen Chou’s teachings are
applied, it would cause a skilled worker to use Lippman’s server to switch to
a higher rate coding and transmit it at increasingly higher transmission rates
according to the probe-lose-throttle principle.” Pet. 53 (citing Ex. 1008
IPR2019-00833
Patent 7,974,200 B2
42
§ VII.A.1). The Petition describes the probe-lose-throttle principle in its
section entitled “Background Knowledge of a Skilled Worker.” Id. at 12
(“When packet loss is discovered, . . . source devices on the Internet reduce
their transmission rates for a time, then gradually increase their transmission
rates in an effort to determine if higher rate transmission can be
supported. . . . [This is] called the ‘probe-lose-throttle’ principle.”) (citing
Ex. 1001, 2:55–67; Ex. 1014 (Lee), 747). The Petition also contends that “a
skilled worker would recognize that, over the course of the streaming
session, transmission rates would rise from an initial, relatively-slow
transmission rate to a higher steady-state transmission rate.” Id. at 27.
In our Institution Decision, we determined preliminarily that
Petitioners’ arguments for this limitation were not persuasive. Decision 30.
We asked the parties to address the following two issues at trial:
First, the parties should address whether the “Target Total
Bitrates” in Lippman’s table are transmission rates, as Lippman
does not appear to describe the “Target Total Bitrates” as
transmission rates.
Second, the parties should address whether the asserted second-
encoding rate data packets in Lippman are transmitted at a
second transmission rate that is higher than the first
transmission rate.
Id. at 30–31. We also determined preliminarily that “the Petition cites over
four pages of Dr. Reader’s declaration for the proposition that Lippman and
Chou would transmit at increasingly higher rates according to the probe-
lose-throttle principle, but Petitioners do not explain in the Petition how this
would occur.” Id. at 31 (citing Pet. 53).
Patent Owner contends that the combination of Lippman, Chou, and
Muroi fails to teach or suggest “transmit[ting] a second plurality of second-
encoding-rate data packets encoded at a second encoding rate to said data
IPR2019-00833
Patent 7,974,200 B2
43
presentation device . . . at a second transmission rate which is higher than
said first transmission rate.” PO Resp. 54. Patent Owner contends that “the
‘Target Total Bitrates’ in Lippman’s table correspond to encoding rates—not
transmission rates.” Id. at 55 (citing Ex. 2002 ¶ 82). Patent Owner
addresses Petitioners’ “probe-lose-throttle” argument by contending that
“the addition of Chou fails to overcome the deficiencies of Lippman.” PO
Resp. 55–56 (citing Ex. 2002 ¶ 87). Patent Owner further contends, “both
Lippman and Chou are silent as to any purported ‘probe-loss-throttle
principle.’” Id. at 57.
In the Reply, Petitioners concede that Lippman does not disclose the
disputed limitation and instead rely exclusively on their probe-lose-throttle
theory. Reply 24. Petitioners state, “although Lippman does not disclose
that packets are transmitted at a second transmission rate that is higher than
the first transmission rate, the Petition explains that this feature is rendered
obvious by ‘probe-lose-throttle’ rate control techniques that are . . . part of
the background knowledge of the ordinarily skilled worker.” Id. Petitioners
contend that Patent Owner does not dispute “that the probe-lose-throttle
technique is familiar to the ordinarily skilled worker or that it would have
been obvious to consider the teachings of Lippman, Chou, and Muroi with
this background knowledge.” Id. at 25. Petitioners also contend that
Dr. Polish failed to challenge Dr. Reader’s findings on the probe-lose-
throttle principle. Id. at 26.
Petitioners’ Reply further contends that a reference introduced in
Dr. Polish’s second declaration, Rejaie (Ex. 2006), “further corroborates
Dr. Reader’s explanation of probe-lose-throttle techniques.” Id. According
to Petitioner, “Rejaie describes application of the probe-lose-throttle
technique to video-streaming systems that switch among layered codings.”
IPR2019-00833
Patent 7,974,200 B2
44
Id. at 27 (citing Ex. 2006 § 2). Petitioners contend that at low bandwidth
levels, Rejaie’s server sends fewer coded video layers to a player device than
at higher bandwidth levels. Id. (citing Ex. 2006 § 1.3). Petitioners further
assert, “[a]ll the while, Rejaie’s server probes for additional bandwidth by
increasing its transmission rate . . . until it determines that a packet is lost,
and it throttles its transmission rate to a lower level.” Id. (citing Ex. 2006
§ 2). The Reply also cites Lee (Ex. 1014) as supporting Petitioners’ probe-
lose-throttle argument. Id. at 24–25.
In its Sur-Reply, Patent Owner disagrees with the Reply’s assertion
that Patent Owner did not dispute that claim 4 is obvious in view of
Lippman, Chou, Muroi, and the probe-lose-throttle technique because its
Patent Owner Response “addressed the probe-lose-throttle technique . . . in
as much detail as Petitioners addressed the technique in the Petition.” Sur-
Reply 23. Patent Owner also contends that Petitioners’ probe-lose-throttle
argument in the Reply is untimely because it “advances new arguments
based on citations to additional prior art references.” Id. at 20–21.
Specifically, Patent Owner contends that the Petition merely makes “passing
mention of probe-lose-throttle” and that the Reply’s “over four new pages of
argument” and “extensive citations to new prior art references” is improper.
Id. at 23 (citing Intelligent Bio-Sys., Inc. v. Illumina Cambridge, Ltd., 821
F.3d 1359, 1369 (Fed. Cir. 2016)).
On the merits, Patent Owner’s Sur-Reply contends that although
probe-lose-throttle is a congestion control mechanism used in Transport
Control Protocol (TCP), “as Petitioners acknowledge, Lippman employs
User Datagram Protocol (UDP) rather than TCP to transmit video.” Id. at 24
(citing Reply 23; Ex. 1010, 780). According to Patent Owner, “UDP does
not implement TCP’s probe-lose-throttle mechanism or any other type of
IPR2019-00833
Patent 7,974,200 B2
45
congestion control.” Id. (citing Pet. 12; Ex. 1018 (RFC 1122), 82). Patent
Owner also contends that “Petitioners provide no explanation of how a
[person of ordinary skill in the art] might further modify the proposed
Lippman-Chou-Muroi combination to implement TCP’s probe-lose-throttle
mechanism.” Id. at 25. Finally, Patent Owner contends that other references
teach away from using TCP for multimedia applications. Id. at 25–26
(citing Ex. 1021 (Padhye), Abstract (“Using TCP for multimedia
applications is not practical . . . but inappropriate”); Ex. 1014 (Lee), 747
(TCP’s probe-lose-throttle “is inappropriate for video packet flows.”)).
(2) Analysis
The Petition’s probe-lose-throttle argument is short on details and
specific citations. The Petition does not cite to any supporting evidence for
its proposition that “a skilled worker would recognize that, over the course
of the streaming session, transmission rates would rise from an initial,
relatively-slow transmission rate to a higher steady-state transmission rate.”
Pet. 27. As we noted at institution, the Petition tersely contends that
“[w]hen Chou’s teachings are applied, it would cause a skilled worker to use
Lippman’s server to switch to a higher rate coding and transmit it at
increasingly higher transmission rates according to the probe-lose-throttle
principle.” Decision 30 (citing Pet. 53 (citing Ex. 1008 § VII.A.1)). But
neither the Petition nor the four pages of Dr. Reader’s declaration that it
cites explains how transmission rates rising “over the course of the
streaming session” would suggest to one of ordinary skill in the art that
second encoding rate packets would be sent at a higher transmission rate
than first encoding rate packets in the cited references. “A skeletal
argument, really nothing more than an assertion, does not preserve a claim.”
SmithKline Beecham Corp. v. Apotex Corp., 439 F.3d 1312, 1320 (Fed.
IPR2019-00833
Patent 7,974,200 B2
46
Cir. 2006) (internal quotation omitted). In short, the Petition fails to
persuade us that the probe-lose-throttle theory plugs the gap in its now-
conceded failure of Lippman to teach the disputed limitation. See Reply 24.
Our decision does not rest on the brevity of the Petition alone. Rather,
we determine that Petitioners’ cited evidence for the probe-lose-throttle
principle also fails to support the Petition’s argument. The Petition cites to
Lee (Ex. 1014 of Ground 2), other references (Exs. 1020–1024), the ’200
patent itself (Ex. 1001, 2:55–67), and Dr. Reader’s declaration to establish
the background knowledge regarding probe-lose-throttle. Pet. 12, 27. Lee is
the reference the Petition and Dr. Reader discuss in the most detail with
respect to this principle. See Pet. 21–23, 79–80 (discussing claim 5). In
contrast, the Petition does not discuss the other references in any detail (Exs.
1020–1024), and Dr. Reader’s declaration merely refers to them in passing.
See Pet. 12; Ex. 1008 ¶ 89 (citing Exs. 1020–1024 with short parenthetical
descriptions). The Petition similarly does nothing more than cite a portion
of the ’200 patent without explanation. See Pet. 12 (citing Ex. 1001, 2:55–
57). We, therefore, focus our review on Lee and Dr. Reader’s testimony.
Petitioners reproduce Figure 6 of Lee in a background section of the
Petition:

IPR2019-00833
Patent 7,974,200 B2
47
Pet. 22 (citing Ex. 1014, 751). Figure 6 of Lee provides a congestion control
example when available bandwidth has suddenly decreased. Ex. 1014, 751.
The Petition explains that Lee’s congestion control algorithm “reacts gently
to non-persistent congestion in order to keep the sending rate variation to a
minimum in steady state.” Pet. 22 (citing Ex. 1014, 750). The Petition
further explains that Lee’s algorithm “reacts aggressively, however, to
persistent congestion,” characterizing this response as “TCP-friendly.” Id. at
23 (citing Ex. 1014, 747). Dr. Reader also testifies that “Lee characterizes
the TCP transmission rate control technique as operating according to a
‘probe-lose-throttle’ principle.” Ex. 1008 ¶ 90 (citing Ex. 1014, 747).
The problem with Petitioners’ reliance on Lee to establish background
knowledge is that although Petitioners state that “a skilled worker would
recognize that, over the course of the streaming session, transmission rates
would rise,” Lee’s figure depicts probe-lose-throttle at a high level without
any particular time scale. See Pet. 27; Ex. 1014, 751. Lee’s vague figure
does not sufficiently illustrate how to compare the timing of Lee’s increasing
transmission rates with the teachings of Lippman, Chou, and Muroi; nor
does the Petition do so. See Pet. 27. Further, Petitioners do not explain how
one of ordinary skill in the art would know when Lee’s transmission rate
would reach its higher steady-state in the combination. Petitioners merely
assert that “[w]hen Chou’s teachings are applied, it would cause a skilled
worker to use Lippman’s server to switch to a higher rate coding and
transmit it at increasingly higher transmission rates according to the probe-
lose-throttle principle.” Id. at 53. From Lee’s Figure 6, it is unclear whether
Lee’s steady-state transmission rate would be reached during
Lippman/Chou’s transmission of lower-encoding rate packets or whether it
would be reached, as Petitioners claim, during Lippman/Chou’s transmission
IPR2019-00833
Patent 7,974,200 B2
48
of higher-encoding rate packets. In fact, as time goes on and congestion
rises, Lee Figure 6 shows a drop in transmission rate, which Petitioners do
not account for in their explanation. See Ex. 1014, 751. Thus, we find that
Lee does not adequately support Petitioners’ argument.
We next consider Dr. Reader’s testimony in greater detail. Dr. Reader
provides additional detail missing from the Petition, but his declaration
cannot make up for the Petition’s brevity when Petitioners cite
indiscriminately to four pages of his declaration in support of a single
proposition. See Pet. 53 (citing Ex. 1008 § VII.A.1). It is the Petition, not
the expert declaration, that must make Petitioners’ case. See Liberty Mutual
Ins. Co. v. Progressive Casualty Ins. Co., CBM2012-00003, Paper 8 at 10,
14 (PTAB Oct. 25, 2012) (the Board “will address only the basis, rationale,
and reasoning put forth by the Petitioner in the petition, and resolve all
vagueness and ambiguity in Petitioner’s arguments against the Petitioner”);
cf. 37 C.F.R. § 42.6(a)(3) (“Arguments must not be incorporated by
reference from one document into another document.”); Cisco Systems, Inc.
v. C-Cation Techs., LLC, IPR2014-00454, Paper 12 at 9 (PTAB Aug. 29,
2014) (“This practice of citing the Declaration to support conclusory
statements that are not otherwise supported in the Petition also amounts to
incorporation by reference.”).
In any event, Dr. Reader’s declaration does not adequately support or
clarify the Petition’s argument. Initially, Dr. Reader discusses the probe-
lose-throttle principle in the context of Lippman’s pre-roll period. Ex. 1008
¶ 158. Dr. Reader testifies that “a skilled worker would understand the
transmission rate during [Lippman’s] pre-roll period would be less than the
full 34 kbps transmission rate ascribed to the ‘56K Modem’ case.” Id.
(emphasis added). Dr. Reader further testifies that “[w]hen the start-up
IPR2019-00833
Patent 7,974,200 B2
49
phase is complete, . . . the streaming server may switch to the 34 kbps
coding . . . . The server may continue to increase its transmission rate
beyond the 34 kbps coding protocol, consistent with the TCP-friendly rate
control techniques.” Id. ¶ 159 (citing id. ¶¶ 86, 89) (emphasis added). In
other words, Dr. Reader testifies that Lippman would initiate a moderate
transmission rate during the pre-roll buffering phase but may continue
increasing transmission rate after switching to a higher encoding rate.
There are several problems with this testimony. First, Dr. Reader
relies only on a citation to his prior general discussion of probe-lose-throttle
techniques to support his opinion that the transmission rate may continue to
increase after Lippman’s coding rate switch. See id. ¶ 159 (citing id. ¶¶ 86,
89); see also id. ¶ 86 (citing Ex. 1001, 2:55–58), id. ¶ 89 (citing Exs. 1014,
1020–1024 without detailed discussion)). Dr. Reader does not explain why
transmission rates may continue to increase after Lippman’s coding rate
switch. In other words, Dr. Reader’s testimony on this point amounts to
unsupported expert opinion. See 37 C.F.R. § 42.65(a) (“Expert testimony
that does not disclose the underlying facts or data on which the opinion is
based is entitled to little or no weight.”); In re Am. Acad. of Sci. Tech Ctr.,
367 F.3d 1359, 1368 (Fed. Cir. 2004) (“[T]he Board is entitled to weigh the
declarations and conclude that the lack of factual corroboration warrants
discounting the opinions expressed in the declarations.”).
Second, Dr. Reader’s assertion that the transmission rate “may
continue to increase” after a coding rate switch is insufficient to show that a
skilled artisan would have understood the transmission rate would increase.
See Ex. 1008 ¶ 159 (emphasis added); cf. Personal Web Techs. LLC v.
Apple, Inc., 848 F.3d 987, 993 (Fed. Cir. 2017) (it is not enough to state that
“a skilled artisan, once presented with the two references, would have
IPR2019-00833
Patent 7,974,200 B2
50
understood that they could be combined”) (emphasis added); InTouch
Techs., Inc. v. VGO Commc’ns, Inc., 751 F.3d 1327, 1352 (Fed. Cir. 2014)
(“[Defendant’s] expert also succumbed to hindsight bias in her obviousness
analysis. [Her] testimony primarily consisted of conclusory references to
her belief that one of ordinary skill in the art could combine these references,
not that they would have been motivated to do so.”).
Finally, Dr. Reader’s testimony is unpersuasive because it relies on
Lippman’s pre-roll period to establish transmission rate timing, but
Petitioners propose eliminating Lippman’s pre-roll period in the
combination with Chou. Compare Ex. 1008 ¶ 158 (“[A] skilled worker
would understand the transmission rate during [Lippman’s] pre-roll period
would be less than the full 34 kbps transmission rate ascribed to the ‘56K
Modem’ case.”) (emphasis added) with Pet. 26–27 (“Lippman explains that
players collect a minimum amount of ‘pre-roll’ packets, then start playing”;
“Chou describes that start-up delays of 2–10 seconds are ‘annoying’ and
‘intolerable’”; “A skilled worker would be motivated to apply Chou’s
techniques in Lippman’s architecture to avoid these ‘annoying’ delays”; and
“playback [would] begin immediately.”) (emphasis added). Dr. Reader does
not explain how probe-lose-throttle timing would work once Lippman’s pre-
roll period is eliminated in the modification with Chou.
Accordingly, we give little weight to Dr. Reader’s testimony that one
of ordinary skill in the art would have understood that Lippman, Chou, and
Muroi in view of the probe-lose-throttle principle would have taught or
suggested the disputed limitation of claim 4. See 37 C.F.R. § 42.65(a).
In the Reply, Petitioners present additional arguments supporting their
probe-lose-throttle theory. See Reply 24–28. Petitioners expound on Lee’s
Figure 6, supplying the following new annotations:
IPR2019-00833
Patent 7,974,200 B2
51

Id. at 24–25. Figure 6 is annotated to show alleged probe, lose, and throttle
portions of a transmission. See id. Petitioners assert, “[i]n this technique, a
server transmits packets at ever-increasing transmission rates (the ‘probe’)
until a packet is lost (‘lose’), at which time the server drops the transmission
rate to a lower level (‘throttle’), and the process repeats.” Id. at 24. We find
that Petitioners’ newly annotated Figure 6 does not clarify how one of
ordinary skill in the art would have understood the timing of probe-lose-
throttle to apply to the combination of Lippman, Chou, and Muroi for the
reasons we describe above, including that the annotated Figure 6 still does
not include a time scale.
Petitioners also provide detailed discussion of the probe-lose-throttle
principle in the context of a new reference, Rejaie (Ex. 2006), cited in
Dr. Polish’s second declaration. Id. at 26–27; Ex. 2002 ¶ 27. Patent
Owner’s Sur-Reply objects to the Reply’s reliance on Rejaie and Lee as
presenting untimely arguments. Sur-Reply 21 (“Petitioners’ new probe-lose-
throttle discussion set forth for the first time in their Reply advances new
arguments based on citations to additional prior art references.”). We agree
with Patent Owner that Petitioners’ argument based on Rejaie is untimely for
reasons we explain below.
IPR2019-00833
Patent 7,974,200 B2
52
Our reviewing court has instructed that “[u]nlike district court
litigation—where parties have greater freedom to revise and develop their
arguments over time and in response to newly discovered material—the
expedited nature of IPRs bring with it an obligation for petitioners to make
their case in their petition.” Intelligent Bio-Systems, Inc. v. Illumina
Cambridge Ltd., 821 F.3d 1359, 1369 (Fed. Cir. 2016). Accordingly, “[i]t is
of the utmost importance that petitioners in the IPR proceedings adhere to
the requirement that the initial petition identify ‘with particularity’ the
“evidence that supports the grounds for the challenge to each claim.’” Id.
(quoting 35 U.S.C. § 312(a)(3)). In Illumina, a petitioner included a new
theory of invalidity in its reply brief, relying on new references not
submitted with its petition. Id. In those circumstances, the Federal Circuit
held that the Board did not err in refusing the reply brief as improper. Id. at
1370. Similarly here, Petitioners improperly include a new theory of
invalidity in the Reply based on a new reference, Rejaie, not in the Petition.
The fact that Dr. Polish introduced Rejaie in his second declaration for
one purpose does not permit Petitioners to use Rejaie for another, unrelated
purpose. See Ex. 2002 ¶ 27. Dr. Polish did not introduce Rejaie for the
purpose of addressing the probe-lose-throttle principle. Rather, Dr. Polish
introduced Rejaie to support his opinion that “a distributed, client-controlled
architecture like the one described in Muroi is not a guaranteed improvement
over a centralized, server-controlled architecture like the one described in
Lippman.” Id. ¶ 15. While Dr. Polish references Rejaie’s “congestion
control mechanisms,” Dr. Polish does so without discussing probe-lose-
throttle techniques and only to support his server-centric point. See id. ¶ 27
(“Because ‘congestion control mechanisms for streaming video playback are
critical’ . . ., Rejaie suggests to persons of skill in the art that decisions about
IPR2019-00833
Patent 7,974,200 B2
53
controlling the amount of traffic to insert into the network . . . are better left
to the server rather than the client.”). Dr. Polish’s discussion of Rejaie in the
context of server-centric architecture did not open the door for Petitioners to
use Rejaie to bolster an unrelated argument about probe-lose-throttle. See
37 C.F.R. § 42.23(b). While we asked the parties in our Institution Decision
to address in more detail whether Lippman teaches or suggests the disputed
limitation, we agree with Patent Owner that our Decision “did not invite
Petitioners to cite new evidence” for its probe-lose-throttle theory. See
Decision 30–31; Sur-Reply 21–22.
Thus, although we are mindful not to parse reply arguments with “too
fine of a filter,”12 we determine that Petitioners’ reliance on Rejaie
constitutes untimely new argument, not responsive to the Patent Owner
Response. Accordingly, we decline to consider it.13 As we determine the
Petitioners failed to meet their burden to show the unpatentability of claim 4,
we need not address the Sur-Reply’s claim 4 arguments on the merits. See
Sur-Reply 23–26.

12 See Ericsson Inc. v. Intellectual Ventures I LLC, 901 F.3d 1374, 1380
(Fed. Cir. 2018). There, the Board improperly excluded petitioner’s reply
argument, but it was “[u]ndoubtedly . . . a special case in which Petitioner,
Patent Owner, and the Board all initially applied the broadest reasonable
interpretation claim construction standard, and only after institution applied
Phillips instead.” Id. This case differs from Ericsson because no such
special claim construction issue exists and because unlike Ericsson,
Petitioners here rely on a new Reply reference. See id.
13 We consider any potential prejudice to Petitioners from declining to
consider Rejaie to be minimal. Counsel for Petitioners admitted during oral
argument that “I don’t think that Rejaie is critical in any way. . . . So, it’s
not so much that Rejaie is needed. It’s that Rejaie is just sort of icing on the
cake because they found a reference that corroborates our argument.”
Tr. 43:19–45:1.
IPR2019-00833
Patent 7,974,200 B2
54
For the foregoing reasons, and based on our review of the complete
record, we are not persuaded that Petitioners have established by a
preponderance of the evidence that claim 4 is rendered obvious by Lippman,
Chou, and Muroi, or those references in view of the background knowledge
of a person of ordinary skill in the art.
c) Independent Claims 7 and 10
Petitioners assert that independent claims 7 and 10 have nearly all of
the same features as claim 1 and identify any pertinent differences on
pages 57–70 of the Petition. Petitioners’ detailed analysis explaining where
the combination of Lippman, Chou, and Muroi teaches the limitations in
these claims is essentially identical to the analysis discussed above with
respect to claim 1. Pet. 29–42. Patent Owner relies on the same arguments
for claims 7 and 10 as it does for claim 1. PO Resp. 27–54.
For the reasons discussed above, and based on our review of the
complete record, we are persuaded that Petitioners have established by a
preponderance of the evidence that claims 7 and 10 are rendered obvious by
Lippman, Chou, and Muroi.
d) Dependent Claims 2 and 3
Petitioners provide a detailed analysis explaining where Lippman,
Chou, and Muroi disclose the limitations in dependent claims 2 and 3.
Pet. 42–44. Patent Owner does not offer separate arguments for these claims
or contest Petitioners’ analysis. For the reasons discussed above, and the
argument and evidence in the Petition (id.), we are persuaded that Petitioners
have established by a preponderance of the evidence that claims 2 and 3 are
rendered obvious by Lippman, Chou, and Muroi.
IPR2019-00833
Patent 7,974,200 B2
55
2. Challenges Applying Lippman, Chou, Muroi, and Lee
Petitioners contend that claims 5, 6, 11, and 12 are obvious over
Lippman, Chou, Muroi, and Lee under 35 U.S.C. § 103(a). Pet. 70–85.
Patent Owner does not challenge Petitioners’ arguments for these claims.14
Petitioners provide a detailed analysis explaining where Lippman,
Chou, Muroi, and Lee disclose the limitations in dependent claims 5, 6, 11,
and 12, which we find persuasive. Pet. 80–85. Petitioners also articulate a
detailed rationale to combine Lee with Lippman, Chou, and Muroi, which
we also find persuasive. Id. at 70–73. Based on our review of the complete
record, we are persuaded that Petitioners have established by a
preponderance of the evidence that claims 5, 6, 11, and 12 are rendered
obvious by Lippman, Chou, Muroi, and Lee.

14 Although the Preliminary Response challenged Petitioner’s arguments on
claim 5 (Prelim. Resp. 56–62), the Patent Owner’s failure to renew these
challenges in its Patent Owner Response constitutes waiver. See Paper 8, 7
(“Patent Owner is cautioned that any arguments for patentability not raised
in the response may be deemed waived.”).
IPR2019-00833
Patent 7,974,200 B2
56
IV. CONCLUSION15
In summary:

V. ORDERS
In consideration of the foregoing, it is hereby:
ORDERED that Petitioners have shown by a preponderance of the
evidence that claims 1–3, 5–7, and 10–12 of the ’200 patent are
unpatentable;
FURTHER ORDERED that Petitioners have not shown by a
preponderance of the evidence that claim 4 is unpatentable; and
FURTHER ORDERED that, because this is a Final Written Decision,
parties to the proceeding seeking judicial review of the decision must
comply with the notice and service requirements of 37 C.F.R. § 90.2.

15 Should Patent Owner wish to pursue amendment of the challenged claims
in a reissue or reexamination proceeding subsequent to the issuance of this
decision, we draw Patent Owner’s attention to the April 2019 Notice
Regarding Options for Amendments by Patent Owner Through Reissue or
Reexamination During a Pending AIA Trial Proceeding. See 84 Fed. Reg.
16,654 (Apr. 22, 2019). If Patent Owner chooses to file a reissue application
or a request for reexamination of the challenged patent, we remind Patent
Owner of its continuing obligation to notify the Board of any such related
matters in updated mandatory notices. See 37 C.F.R. § 42.8(a)(3), (b)(2).
Claims

35
U.S.C. §
Reference(s)/Basis Claims
Shown
Unpatentable
Claims
Not shown
Unpatentable
1–4, 7,
10 103(a)
Lippman, Chou,
Muroi
1–3, 7, 10 4
5, 6, 11,
12 103(a)
Lippman, Chou,
Muroi, Lee
5, 6, 11, 12
Overall
Outcome
1–3, 5–7, 10–
12
4
IPR2019-00833
Patent 7,974,200 B2
57
For PETITIONER:
Robert Hails
T Cy Walker
BAKER HOSTETLER LLC
rhails@bakerlaw.com
cwalker@bakerlaw.com

For PATENT OWNER:
Daniel Boehnen
Grantland Drutchas
Jeffrey Armstrong
George Lyons
MCDONNELL BOEHNEN HULBERT & BERGHOFF, LLP
boehnen@mbhb.com
drutchas@mbhb.com
armstrong@mbhb.com
lyons@mbhb.com