Trials@uspto.

gov Paper 36
Tel: 571-272-7822 Entered: May 14, 2019

UNITED STATES PATENT AND TRADEMARK OFFICE

_______________

BEFORE THE PATENT TRIAL AND APPEAL BOARD

_______________

RIOT GAMES, INC.,

Petitioner,

v.

PALTALK HOLDINGS, INC.,

Patent Owner.
_______________

Case IPR2018-001321
Patent 6,226,686 & 6,226,686 C12
_______________

Before THU A. DANG, KARL D. EASTHOM, and NEIL T. POWELL,

Administrative Patent Judges.

DANG, Administrative Patent Judge.

FINAL WRITTEN DECISION

Inter Partes Review
35 U.S.C. § 318(a) and 37 C.F.R. § 42.73

1
The panel joined Petitioner Valve Corp. and Case IPR2018-01243 to the
instant proceeding. See Paper 34.
2
The Petition challenges original claims and claims issued pursuant to an
ex parte reexamination.
IPR2018-00132
Patent 6,226,686

I. INTRODUCTION

A. Background
Riot Games Inc. (“Petitioner”) filed a Petition requesting an inter
partes review of claims 1, 3, 7, 12, 18, 22–27, 36, 41– 46, 55, and 58–63 of
U.S. Patent No. 6,226,686 (Ex. 1002, “the ’686 patent”). Paper 1 (“Pet.”).
PalTalk Holdings, Inc. (“Patent Owner”) filed a Preliminary Response.
Paper 6 (“Prelim. Resp.”). Pursuant to our prior authorization (Paper 8,
“Order”), Petitioner filed a Reply to the Patent Owner Preliminary Response
(Paper 9, “Reply to Prelim. Resp.”) as to the issue of Patent Owner’s claim
constructions, and Patent Owner filed a Preliminary Sur-Reply (Paper 10,
“Prelim. Sur-Reply”).
We instituted trial to determine whether claims 1, 3, 7, 12, 18, 22–27,
36, 41– 46, 55, and 58–63 are unpatentable under 35 U.S.C. § 103 based on
the combination of Aldred and RFC 1692 either alone or in combination with
Ulrich or Denzer. See Paper 11 (“Institution Decision” or “Inst. Dec.”).
After institution of trial, Patent Owner filed a Request for Rehearing. Paper
14 (“Reh’g. Req.”). We denied Patent Owner’s Request for Rehearing.
Paper 18 (“Rehearing Decision” or “Reh’g Dec.”).
Patent Owner then filed a Response. Paper 22 (“PO Resp.”).
Petitioner filed a Reply to Patent Owner’s Response. Paper 25 (“Pet.
Reply”). Pursuant to our prior authorization (Paper 26, “Order”), Patent
Owner filed a Sur-Reply to Petitioner’s Reply (Paper 30, “PO Sur-Reply”).
Oral argument was conducted on February 13, 2019.
We have jurisdiction under 35 U.S.C. § 6. This decision is a Final
Written Decision under 35 U.S.C. § 318(a) as to the patentability of

1
IPR2018-00132
Patent 6,226,686

claims 1, 3, 7, 12, 18, 22–27, 36, 41– 46, 55, and 58–63 of the ’686 patent.
For the reasons discussed below, we hold that Petitioner has demonstrated by
a preponderance of the evidence that claims 1, 3, 7, 12, 18, 22–27, 36, 41–
46, 55, and 58–63 of the ’686 patent are unpatentable under 35 U.S.C.
§ 103(a).

B. Related Proceedings
Petitioner states that the ’686 patent is related to the following U.S.
Patents: 5,822,523 (“the ’523 patent”) and 6,018,766. Pet. 1. According to
Petitioner, ex partes reexamination No. 90/011,036 (Ex. 1006) involved a
reexamination of the ’686 patent. Pet. 1.
A concurrent request for inter partes review, IPR2018-00131,
challenges claims of the ’686 patent. Pet. 1. Two other concurrent requests
for inter partes review, IPR2018-00129 and IPR2018-00130, challenge
claims of the ’523 patent. Pet. 1.
Petitioner also states that the following cases involve the ’523 and ’686
patents: PalTalk Holdings, Inc. v. Valve Corp.n, No. 16-cv-1239-JFB-SRF
(D. Del.) (filed Dec. 16, 2016); PalTalk Holdings, Inc. v. Riot Games, Inc.,
No. 1:16-cv-1240-JFB-SRF (D. Del.) (filed Dec. 16, 2016); PalTalk
Holdings, Inc. v. Sony Computer Entertainment America, Inc. et al., No.
2:09-cv-00274-DF-CE (E.D. Tex.) (filed Sept. 14, 2009); PalTalk Holdings,
Inc. v. Microsoft Corp., No, 2:06-cv-00367-DF (E.D. Tex.) (filed Sept. 12,
2006); and Mpath Interactive v. Lipstream Networks, Inc., et al., No. 3:99-cv-
04506-WHA (N.D. Cal.) (filed Oct. 7, 1999). Pet. 1–2.

C. The ’686 Patent

The ’686 patent issued on May 1, 2001, from an application filed
September 28, 1999, and claims priority to parent application

2
IPR2018-00132
Patent 6,226,686

No. 08/896,797, filed on July 18, 1997, now US 6,018,766, which in turn is a
continuation of application No. 08/595,323, filed on February 1, 1996, now
US 5,822,523. Ex. 1002, [45], [22], and [63].
The ’686 patent, titled “Server-Group Messaging System for
Interactive Applications,” describes a “method for deploying interactive
applications over a network containing host computers and group messaging
servers.” Id. at [54], [57]. Figure 5, reproduced below, illustrates a unicast
network over which the interactive applications may be deployed.

Figure 5 depicts a wide area network with hosts 58, 59, 60, and 61, and a
group messaging server (“GMS”) 62. Id. at 8:65–66. Host 58 has Transport
Level Protocol (TLP) address A and Upper Level Protocol (ULP) address H.
Id. at 8:66–67. Host 59 has TLP address C and ULP address J, host 60 as

3
IPR2018-00132
Patent 6,226,686

TLP address B and ULP address I, host 61 has TLP address D and ULP
address K, and GMS 62 has TLP address S. Id. at 8:67–9:2. “The network is
a conventional unicast network of network links 69, 70, 71, 72, 73, 74, 75,
76, and 77 and unicast routers 63, 64, 65, 66, 67, and 68.” Id. at 9:2–5. GMS
“62 receives messages from the hosts addressed to a message group and
sends the contents of the messages to the members of the message group.”
Id. at 9:5–8.
Figure 7, reproduced below, depicts ULP datagrams with payload
aggregations for implementing an interactive gaming application between the
four hosts in Figure 5.

Figure 7 shows GMS (“Group Server”) 62 receiving multiple messages 96,

97, 98, and 99 before sending them to hosts within message group G.

4
IPR2018-00132
Patent 6,226,686

Id. at 9, 18–20, 10:24–28. As shown in Figure 7, multiple messages 96, 97,

98, and 99, each respectively contain payload P1, P2, P3, and P4, to be
aggregated into a single larger message, 100, 101, 102, or 103. Id. Host 58
sends message 96 (shown in Figure 7 as “Host A sends”), host 60 sends
message 97 (shown in Figure 7 as “Host B sends”), host 59 sends
message 98 (shown in Figure 7 as “Host C sends”), and host 61 sends
message 99 (shown in Figure 7 as “Host D sends”), wherein each of the
messages from the hosts has destination TLP address S and ULP address G
for GMS 62. Id. at 10:28–32. After GMS 62 receives all four of these
messages, it creates four outbound messages 100, 101, 102, and 103. Id. at
10:33–34. Aggregated message 100 includes destination TLP address A and
ULP address H for host 58 and aggregated payload P2, P3, and P4,
respectively, from the messages from hosts 59, 60, and 61. Id. at 10:38–40.
Aggregated message 101 targets host 60, aggregated message 102 targets
host 59, and aggregated message 103 targets host 61. Id. at 10:41–42.
Figure 9, reproduced below, depicts a datagram format and address
format for ULP messages.

5
IPR2018-00132
Patent 6,226,686

Figure 9 shows a ULP datagram message comprising elements 123, 124, 125,
126, 127, 128, and 129. Id. at 13:62–64. Transport datagram TLP header
123 encapsulates the ULP datagram that includes ULP message type field
124, destination ULP address 125, address count field 126, auxiliary
destination address 127 and 128, and finally payload 129. Id. at 13:64–14:37.
Items 116, 117, 118, 119, 120, 121, and 122 define the payload format of the
ULP datagram. Id. at 14:38–39. Item 116 specifies the message count and
defines the number of payload elements contained in the payload. Items 117,
118, and 119 comprise the first payload element in the payload, and items
120, 121, and 112 comprise the last payload element in the payload. Id. at
14:39–48. In particular, item 117 specifies the ULP address of the source of
the first payload element, item 118 specifies the data length for the data in the
first payload element, and item 119 constitutes the actual data. Similarly,
item 120 specifies the ULP address of the source of the last payload element
N, item 121 specifies the data length for the data in the last payload element
N, and item 122 constitutes the actual data. Id.

D. The Challenged Claims

Of the challenged claims, claims 1, 3, 7, 12, and 18 are the independent
claims at issue. Claim 1 is illustrative of the challenged claims, and is
reproduced below:

1. A method for facilitating communications among a

plurality of host computers over a network to implement a shared,
interactive application, comprising the steps of:
(1) receiving a create message from one of the plurality of host
computers, wherein said create message specifies a message group to
be created;

6
IPR2018-00132
Patent 6,226,686

(2) receiving join messages from a first subset of the plurality of

host computers, wherein each of said join messages specifies said
message group;
(3) receiving host messages from a second subset of said first
subset of the plurality of host computers belonging to said message
group, wherein each of said messages contains a payload portion and a
portion that is used to identify said message group;
(4) aggregating said payload portions of said host messages
received from said second subset of the plurality of host computers to
create an aggregated payload;
(5) forming an aggregated message using said aggregated
payload; and
(6) transmitting said aggregated message to said first subset of
the plurality of host computers belonging to said message group;
wherein said aggregated message keeps the shared, interactive
application operating consistently on each of said first subset of the
plurality of host computers.
Ex. 1002, 27:50–28:8.
E. Instituted Grounds of Unpatentability
We instituted trial on the following specific grounds (Pet. 21, 51, and
66):

Reference Basis Claim(s) Challenged

1, 3, 7, 12, 18, 26, 27, 45, 46, 62,
Aldred,3 and RFC 16924 § 103
and 63
Aldred, RFC 1692, and Ulrich5 § 103 22–27, 41–46, and 58–63
Aldred, RFC 1692, and Denzer6 § 103 36, and 55

3
WO 94/11814 (May 26, 1994) (“Aldred”; Ex. 1009).
4
Request for Comments (RFC) 1692 (Aug. 1994) (“RFC 1692”; Ex. 1010).
5
US 5,466,200 (Nov. 14, 1995) (“Ulrich”; Ex. 1012).
6
US 5,307,413 (Apr. 26, 1994) (“Denzer”; Ex. 1014).

7
IPR2018-00132
Patent 6,226,686

Petitioner also relies on the testimonies of Dr. Steve R. White.7

Exs. 1007, 1053. Patent Owner relies on the testimony of Dr. Kevin C.
Almeroth. Ex. 2002.

II. ANALYSIS

A. Claim Construction
The parties agree that the ’686 patent expired. Pet. 5; PO Resp. 1.
Accordingly, we construe its challenged claims as they would be construed in
district court. See 37 C.F.R. § 42.100(b) (2017) (permitting a “district court-
type claim construction approach . . . if a party certifies that the involved
patent will expire within 18 months from the entry of the Notice of Filing
Date Accorded to Petition”).
In district court, claim terms carry their plain and ordinary meaning as
would be understood by a person of ordinary skill in the art at the time of the
invention and in the context of the entire patent disclosure. Phillips v. AWH
Corp., 415 F.3d 1303, 1313 (Fed. Cir. 2005) (en banc). “There are only two
exceptions to this general rule: 1) when a patentee sets out a definition and
acts as his own lexicographer, or 2) when the patentee disavows the full
scope of a claim term either in the specification or during prosecution.”
Thorner v. Sony Comput. Entm’t Am. LLC, 669 F.3d 1362, 1365 (Fed. Cir.
2012). Only terms in controversy must be construed and only to the extent
necessary to resolve the controversy. Vivid Techs., Inc. v. Am. Sci. & Eng’g,
Inc., 200 F.3d 795, 803 (Fed. Cir. 1999); Nidec Motor Corp. v. Zhongshan

7
Petitioner superfluously cites the “Knowledge of an Ordinary Artisan,”
which we do not repeat, as an obviousness determination takes into account
that knowledge.

8
IPR2018-00132
Patent 6,226,686

Broad Ocean Motor Co., 868 F.3d 1013, 1017 (Fed. Cir. 2017) (applying
Vivid Techs. in the context of an inter partes review).
Petitioner notes Patent Owner “advanced several constructions for . . .
claim elements” in prior district court litigation. See Pet. 5–6. Petitioner
generally contends the “precise scope” of the terms need not be determined,
provided the terms track “any interpretation consistent with their plain and
ordinary meaning in the context of the [’]686 [p]atent.” Id.
After determining via a teleconference with the parties that with
respect to three claim terms, Patent Owner’s proposed constructions in its
Preliminary Response differed from what Patent Owner provided in prior
district court litigation (Paper 8, 2–4), we authorized the filing of a
Preliminary Reply by Petitioner (Paper 9 (“Prelim. Reply”)) and a Sur-Reply
by Patent Owner (Paper 10 (“Prelim. Sur-Reply”)) to address three terms:
“aggregated message,” “aggregated payload,” and “payload portion.” In its
Patent Owner Response (Paper 22, 1–15), Patent Owner maintains the
constructions for “aggregated message” and “aggregated payload” it
proposed in its Preliminary Response. As discussed below and as set forth in
the Institution Decision, the constructions of the terms involve the
overlapping issue of a transport layer header.
1. “aggregated message”(claims 1, 3, 7, 12);
“server message” (claim 18)
Patent Owner contends “aggregated message” means “[o]ne or more
messages containing a single transport layer message header, destination
data, and data items from an aggregated payload.” PO Resp. 4. Patent
Owner relies on Figure 7 of the Specification as providing an example:
Each of the messages 100, 101, 102 and 103 received by a
host from a server includes the aggregated payloads (Pn1, Pn2,

9
IPR2018-00132
Patent 6,226,686

Pn3) in each message and a header portion consisting of a

transport layer protocol source address (S) of the server, a
transport layer protocol destination address (A, B, C or D) for
the destination host and a destination upper layer protocol (ULP)
address (H, I, J or K) for the destination host.

Id. at 6 (citing Ex. 1002, 8:1–10:67; Fig. 7).

Patent Owner contends Figure 7 discloses “only a single message
header consisting of the transport layer protocol source address, the transport
layer protocol destination address and the ULP address,” which is then
“combined with the aggregated payload.” Id.
Patent Owner then relies on Figure 9 of the Specification. Id.
Figure 9, annotated by Patent Owner, follows:

As annotated and described by Patent Owner, Figure 9 shows “an aggregated

message” which includes “a message header (blue)[,] a payload (green)[,
with] multiple payload elements (red) included as part of an aggregated

10
IPR2018-00132
Patent 6,226,686

payload.” Id. at 7 (citing Ex. 1002, 14:37–52).8 Although Patent Owner

concedes that “the payload 129 does include multiple Source ULP addresses
117, 120,” Patent Owner contends that “the Source ULP addresses 117, 120
are not transport layer headers” because a “ULP source address is part of a
layer above the transport layer (the Session Layer).” Id. at 9 (citing Ex. 2002
¶¶ 48–49).
Further, Patent Owner contends that the Specification “supports Patent
Owner’s construction.” Id. In particular, the Specification states: “The GMS
control function 136 will use the destination ULP host address to look up the
TLP address of the host from the host address map 137,” and “[t]his will be
used to create a TLP header for the message 123.” Id. (citing Ex. 1002,
23:20–23 (emphasis included)). Thus, according to Patent Owner, “[t]here is
. . . no indication in the ’686 Patent that multiple TLP headers are included
within the aggregated message,” but instead, “[a] single transport layer
header is used because all aggregated payloads are being transmitted to a

8
Contrary to Patent Owner’s argument, the ’686 patent does not describe
data elements 124–128 as part of transport layer header 123. Rather, it
indicates that transport header 123 encapsulates those elements and payload
portion 129. See Ex. 1002, 13:59–66 (“The ULP can be implemented as a
datagram protocol by encapsulating addresses, message type information and
the message payload within a datagram of the underlying network transport
protocol. The general form of the ULP datagram message format is shown in
FIG. 9 as elements 123, 124, 125, 126, 127, 128 and 129. The transport
header 123 is the datagram header of the TLP that is encapsulating the ULP
datagram. The ULP message type field 124 . . . must be present in a ULP
datagram.” (emphases added)); Ex. 1053 ¶¶ 6–8 (describing encapsulated
payload and address portions as typical under the OSI, TCP, and IP
frameworks and citing Dr. Almeroth’s testimony from another proceeding).
In any case, the outcome here does not depend on what the disclosed
transport header includes in this particular disclosed example of Figure 9.

11
IPR2018-00132
Patent 6,226,686

same destination host running a same application as other hosts.” Id. at 11

(citing Ex. 2002 ¶¶ 50–52).
Petitioner contends that the ordinary meaning of “aggregated message”
or “server message” does not “require excluding any headers,” but rather, the
term “message” is “used by the ’686 patent in [its] conventional sense.” Pet.
Reply 13 (citing Ex. 1053, ¶¶ 5–6; Ex. 1002, 1:28–55). Petitioner contends,
similarly, “[t]he claims provide sufficient guidance on the meaning of . . .
‘aggregated message’/‘server message,’” but they “do not support excluding
any ‘transport layer’ headers.” Id. at 14. Petitioner points out that Patent
Owner’s “district court construction of ‘aggregating/ aggregated’ likewise
includes no ‘transport layer’ header requirement.” Id. 13–14 (citing Ex.
1055, 2–3), 17 (asserting that in prior district court proceedings, Patent
Owner “never advanced a ‘transport layer message header’ requirement until
after these proceedings were filed” (citing Ex. 1005, 396–97; Ex. 1006, 234–
36; Ex. 1052, 108:8–24; Ex. 1054, and Ex. A, 1, 3).9

9
Petitioner alleges the ’686 patent creates a distinction between layers and
levels. See Pet. Reply 18. The ’686 patent references to “Level” protocols in
Upper Level Protocol (ULP) or Transport Level Protocol (TLP), respectively,
as associated with a “Session Layer protocol on top of the Transport Layer”
of the network in the “OSI reference model.” See Ex. 1002, 12:46–50;
accord id. at 8:38–43. The ’686 patent also refers to “the OSI reference
model for layers of network protocols.” Id. at 3:45–46. “On top of IP [at
layer 3] are the layer 4 transport protocols TCP and [“User Datagram
Protocol”] UDP.” UDP does not guarantee “in-order delivery” of application
datagrams of a data stream, whereas TCP divides the stream into packets to
ensure “reliable, in-order delivery.” See id. at 3:46–51. Our claim
construction and holding does not turn on any alleged distinction between
level and layer, but we agree with Petitioner that the ’686 patent discusses
TLP as either IP or TCP/IP. See Pet. Reply 18–19.

12
IPR2018-00132
Patent 6,226,686

Petitioner contends that the Specification supports Petitioner’s position

of not excluding any “transport layer” headers. Pet. Reply 14. According to
Petitioner, the ’686 patent “explains the Internet Protocol (IP) and
conventional networking use of the [Open Systems Interconnection] OSI
reference model for layers of network protocols,” wherein “OSI network
layers are hierarchical—the packet for each layer (containing a header and
payload) encapsulates the packets for the layers above.” Id. According to
Petitioner, in OSI network layers, “an IP packet payload may be an entire
TCP packet including a TCP header and payload.” Prelim. Reply 4 (citing
Ex. 1011 (RFC 791), 1).
Petitioner provides the testimony in previous proceedings by
Dr. Almeroth, Patent Owner’s declarant, for explanation of encapsulation
using the OSI model. See Pet. Reply 14–15. As an example, Petitioner
provides the following diagram by Dr. Almeroth:

Pet. Reply 15 (reproducing the above figure from Ex. 1056 ¶ 68). The above
figure represents encapsulation of higher layers, including Transmission

13
IPR2018-00132
Patent 6,226,686

Control Protocol (“TCP”) segments and headers, as data forming an IP

datagram. Dr. Almeroth explains as follows:
This process of adding a layer header to the data from the
preceding layer is sometimes referred to as “encapsulation”
because the data and layer header is treated as the data for the
immediately following layer, which, in turn, adds its own layer
header to the data from the preceding layer. Each layer is
generally not aware of which portion of the data from the
preceding layer constitutes the layer header or the user data.

Ex. 1056 ¶ 68.

In summary, according to Dr. Almeroth, encapsulation using the OSI
model involves treating upper level headers as data, and thus, Petitioner
contends, “‘aggregated message’/‘server message’ could have multiple TCP
headers.” Pet. Reply 15 (citing 1053 ¶¶ 7–8).
Further, Petitioner contends that Patent Owner impermissibly relies on
a single embodiment in the ’686 patent “where the server removes Transport
Level Protocol (TLP) headers from received messages” to support its
“transport layer” header requirement, although “the claims of the patent are
not ‘construed as being limited to that embodiment.” Pet. Reply 16 (citing
PO Resp. 4–11, 13–14; Phillips v. AWH Corp., 415 F.3d 1303, 1323 (Fed.
Cir. 2005)). In particular, Petitioner contends that the ’686 patent supports
more than a single embodiment, thereby impacting the breadth of an
“aggregated message” (and the related term “aggregated payload”). Id.
We note Patent Owner advanced similar arguments prior to institution.
Here, Patent Owner concedes the claims encompass encapsulated headers, as
used in the OSI model. See PO Resp. 8 (“Patent Owner’s construction
position is not that the term ‘aggregated message’ does not encompass

14
IPR2018-00132
Patent 6,226,686

encapsulated headers.”) (citing Inst. Dec. 13)).10 According to Patent Owner,

“Patent Owner’s construction is that an ‘aggregated message’ or ‘server
message’ includes only a single transport layer message header.” PO
Resp. 8–9. Nevertheless, as explained further below and above, and as Patent
Owner concedes, the ’686 patent supports encapsulating header information
as data, and thus, as Petitioner contends, “‘aggregated message’/‘server
message’ could have multiple TCP headers.” Pet. Reply 15.
In response to Patent Owner’s similar arguments prior to institution,
the panel determined “on this preliminary record, that the Specification and
evidence support an ‘aggregated message’ as including an aggregated
payload and at least one header in addition to any that may happen to be
within the aggregated payload.” Inst. Dec. 15 (citing Prelim. Reply 1–5;
Ex. 1016, 93; Ex. 1002, Fig. 7). In particular, the panel determined the
following:
headers, such as headers 117 and 118, or 120 and 121, appear in
each payload. See Ex. 1002, 23:11–12 (“Each payload
[includes] a ULP source address, a data length and the data to be
sent.”). Even though an embodiment strips out a TLP header
from a “message,” it looks up a TLP header of the host from
“host address map 137.” Id. at 23:20–22. The specification
consistently shows that a payload, even within an aggregated
payload, may contain header fields. See, e.g., id. at Fig. 9.

Inst. Dec. 12.

10
Patent Owner’s concession responds to the panel’s preliminary
determination in the Institution Decision stating “Patent Owner does not
dispute, in a clear fashion, Petitioner’s contention that the claims may
encompass encapsulated headers.” Inst. Dec. 13.

15
IPR2018-00132
Patent 6,226,686

Here, Patent Owner does not dispute this preliminary finding but rather
argues that there is “no indication in the ’686 Patent that multiple TLP
headers are included within the aggregated message.” PO Resp. 11.
However, the record supports the preliminary finding, and the parties agree,
that the ’686 patent discloses using TLP headers or a “datagram protocol” to
encapsulate messages and/or payloads that include headers (e.g., address
information, message type), as discussed above. See Ex. 1002, 13:59–62,
14:38–62, 26:28–45. As such, the ’686 patent supports including any type of
a header, including TLP headers or other headers in the OSI model, as part of
the data portion of encapsulated messages.
As Patent Owner also points out, the Institution Decision also includes
the following preliminary finding: “The specification describes any data
reduction as significant only for small packet sizes, and generally attributes
data reductions due to message aggregation.” See PO Resp.12 (citing Inst.
Dec. 14; Ex. 1002, 24:23–28). As we noted in the Institution Decision, “the
challenged claims do not limit the payload size and generally allow for
different header types according to the Specification,” wherein ‘the
Specification generally describes savings based on aggregation for all packet
sizes based on “greatly reduc[ing] the total message rate received by the
hosts,’” and “[t]he Specification therefore does not limit an aggregated
payload, as claimed, from including headers in general.” Inst. Dec. 14;
Ex. 1002, 24:12–15, 24:38–47. We note Patent Owner does not urge a
packet size limitation in its claim construction.
Patent Owner responds to this preliminary finding by arguing as
follows:

16
IPR2018-00132
Patent 6,226,686

The ‘686 Patent however clearly discusses significant data

reduction by eliminating transport headers from payloads. The
‘686 Patent states “[a]ggregation will also reduce the total data
rate to the hosts since aggregation eliminates the need for
separate message headers for each payload item. The savings
will be significant for small payload items since there will be
only one message header comprising fields 123, 124 and 125
for multiple payload items.” Ex. 1002, 24:23–28 (emphasis
added). The ‘686 Patent also states that an aggregated message
is “longer and contains multiple payloads, but this is a
significant improvement over received multiple messages with
the wasted overhead of multiple message headers and
message processing time.” Ex. 1002, 10:44–47.

Id. at 12.
Patent Owner (id.) and Dr. Almeroth (Ex. 2002 ¶¶ 53–54) focus on
reduced data rate, but the Specification also describes savings based on
aggregation for all packet sizes based on “greatly reduc[ing] the total
message rate received by the hosts,” because “[a] single message to a host
will be able to carry multiple payload items received from the other hosts
during the aggregation period.” Ex. 1002, 24:12–15 (emphasis added). This
shows that savings in message rate occurs regardless of whether the data
packet portion contains encapsulated header information, because no wasted
overhead occurs in treating the encapsulated header data as data. So this
message rate savings still occurs even if the encapsulated portion of the
packet includes TCP header information, because the system processes that
encapsulated header portion as data, as Dr. Almeroth explains as noted
above. See Ex. 1056 ¶ 68; Ex. 1058 ¶ 56. The ’686 patent supports this
finding as it recognizes that “[a]ggregation will be very effective in collecting
together all of the messages from all of the other hosts into a single message

17
IPR2018-00132
Patent 6,226,686

for each member of the group,” and “[t]his reduces processing . . . since a
single message will be received rather than many separate messages.”
Ex. 1002, 24:18–23. In other words, the reduced message rate benefit that
accrues for a single message occurs regardless of the size of packets (each
which may or may not include headers) aggregated in the single message.
See id. The Specification, therefore, does not limit an aggregated message or
server message, as claimed, from including encapsulated headers as data in a
single aggregated message (which occurs in the OSI model), including
transport layer headers encapsulated within the payload.11
As Petitioner points out, the ’686 patent supports more than a single
embodiment. Pet. Reply 16. The Specification refers to a preferred
embodiment as specifying “the TLP protocol is TCP/IP,” and it states that for
aggregated messages, “the [encapsulated] payload will still contain the source
host ULP addresses in each [of] the payload items.” Ex. 1002, 26:28–50. In
general, however, the Specification supports many types of packets, further
showing that the broad claims must not be limited to stripping TCP or TLP
headers from a payload: “The wide area network used to transport the ULP
protocol need not be the Internet or based on IP. Other networks with some
means for wide area packet or datagram transport are possible including
ATM networks or a digital cable television network.”

11
The Federal Circuit instructs that simply describing alternative features
without articulating advantages or disadvantages of each feature cannot
support a negative limitation. Inphi Corp. v. Netlist, Inc., 805 F.3d 1350,
1356–57 (Fed. Cir. 2015). To the extent that excluding multiple TLP headers
involves the advantage of data reduction, including other header types within
the scope of the claim defeats any advantage of excluding a specific type
from that scope.

18
IPR2018-00132
Patent 6,226,686

Id. at 27:38–43. Consistent with the ’686 patent, a packet message includes
at least one header, and packet bodies may contain encapsulated packets each
with their own headers and bodies. See Prelim. Reply 4; Ex. 1016, 93;
Ex. 1002, 3:28–56, Fig. 7, Fig. 9; Ex. 1011, 1; Ex. 1056 ¶ 68; Ex. 1058 ¶ 56;
Ex. 1046 (PC NETWORKING HANDBOOK (1996)).12
On this record, the ’686 Specification supports Petitioner’s argument
that the claim term “aggregated message” or “server message,” consistent
with its ordinary and plain meaning, includes a message having an
aggregated payload and at least one header in addition to any additional
headers that may happen to be within the aggregated payload. Here, Patent
Owner’s past claim construction positions support this determination by
showing, at the least, prior to this inter partes trial, how Patent Owner viewed
the meaning of this claim term. See Ex. 1005, 396–97; Ex. 1006, 234–36;
Ex. 1052, 108:8–24; Ex. 1054, and Ex. B, 1.13
2. “aggregated payload” (claims 1, 7, 12);
“aggregating said payload portions” (claim 3);

12
“A packet that contains data and delivery information is a datagram.”
Ex. 1046, 178. “Packets have two parts: the header and the body.”
Id. at 179. “The header carries information such as the source and
destination of a packet.” Id. “The body is the raw data carried by a packet
or, in many cases, another type of (encapsulate) packet that contains its own
header and body.” Id. (emphasis added).
13
Petitioner notes that Patent Owner did not alter its original claim
construction positions during district court litigation even up to about two
and a half weeks prior to filing its Preliminary Response here on
February 15, 2018, but altered its position to include the transport later
requirements after filing the Preliminary Response. See Pet. Reply 17 (citing
Ex. 1054, A-1; Ex. 1055, 2–4).

19
IPR2018-00132
Patent 6,226,686

“aggregating said payload portion with the

payload portion of a second host message”
(claim 18)

Patent Owner contends “aggregated payload” should be construed as

“[a] collection of two or more data items that does not include transport layer
headers.” PO Resp. 13. To support its construction, Patent Owner contends
payload portions of messages, such as the messages 96, 97, 98,
and 99 in FIG. 7, received by the group messaging server have
TLP headers removed and are aggregated into an aggregated
payload. The aggregated payload is included in a single
aggregated message with a single transport layer message
header. As explained above in Section II.A., the specification of
the ‘686 Patent describes that transport layer headers are
removed from messages sent to the group messaging server.

Id. at 14 (citing Ex. 1002, 20:14–30; Ex. 2002 ¶ 56).

Petitioner contends that “[the claims of] the ’686 patent do not limit the
content of ‘payloads’ and thus does not limit what the claimed ‘aggregated
payload’ . . . can comprise.” See Pet. Reply 13. Petitioner notes that, similar
with “aggregated message” discussed above in Section II(A)(1), in prior
litigation, Patent Owner also submitted a construction for “aggregating . . .
said payload portions” without submitting the negative limitation regarding
the “transport layer header” requirements. See Pet. Prelim. Reply 2 (citing
Ex. 1016, 93, and 121–22). Petitioner contends, similar to the arguments
concerning “aggregated message” in Section II(A)(1), “[t]he claims provide
sufficient guidance on the meaning of ‘aggregated payload,’” but they “do
not support excluding any ‘transport layer’ headers.” Id. at 14.
According to Petitioner, the Specification supports Petitioner’s position
of not excluding any “transport layer” headers. Pet. Reply 14. In particular,

20
IPR2018-00132
Patent 6,226,686

as an example, the Specification shows that a payload may include more than
“actual data” 119, for example, “a triplet of source ULP address, data length
and data.” Ex. 1002, 14:38–43 (discussing items 117, 118, and 119). It also
shows that a payload maybe “encapsulat[ed]” as a “datagram protocol.”
Id. at 13:59–62.
As Petitioner contends above in Section II (A)(1), “OSI network layers
are hierarchical—the packet for each layer (containing a header and payload)
encapsulates the packets for the layers above.” Pet. Reply 14. According to
Petitioner’s declarant, Dr. Almeroth, encapsulation using the OSI model
involves treating upper level headers as data, wherein “[t]he ‘payload
portion’ of an IP packet could thus comprise a TCP header.” Id. at 15 (citing
Ex.1053 ¶¶ 7–8).
Petitioner also relies on similar teachings in RFC 791, which states that
the IP module that a TCP module calls could take “a TCP segment (including
the TCP header and user data) as the data portion of an [IP datagram].” Pet.
Reply 14 (citing Ex. 1011, 1).
For additional reasons similar to those explained above in construing
“aggregated message” or “server message,” an “aggregated payload,”
consistent with its ordinary and plain meaning, encompasses a collection of
two or more data items that may include headers transported as data. Patent
Owner’s past claim construction positions support this determination by
showing, at the least, prior to this inter partes trial, how Patent Owner viewed
the meaning of this term. See Ex. 1005, 396–97; Ex. 1006, 234–36;
Ex. 1052, 108:8–24; Ex. 1054, and Ex. B, 1.

21
IPR2018-00132
Patent 6,226,686

B. Asserted Obviousness over Aldred, and RFC 1692; Aldred, RFC

1692 and Ulrich; or Aldred, RFC 1692 and Denzer
Petitioner alleges that claims 1, 3, 7, 12, 18, 26, 27, 45, 46, 62, and 63
would have been obvious over Aldred and RFC 1692 as viewed by the
knowledge of an ordinary artisan. Pet. 21–51. Petitioner also alleges that
claims 22–27, 41–46, and 58–63 would have been obvious over the
combination of Aldred, RFC 1692 in further view of Ulrich; and that claims
36 and 55 would have been obvious over the combination of Aldred,
RFC 1692 in further view of Denzer. Pet. 51–69. Patent Owner raise several
arguments in response. Prelim. Resp. 13–50. Brief overviews of Aldred,
RFC 1692, Ulrich and Denzer follow, followed by an analysis of the parties’
contentions regarding the challenged claims.
1. Principles of Law
A claim is unpatentable under 35 U.S.C. § 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 evidence, objective indicia of non-
obviousness (i.e., secondary considerations). See Graham v. John Deere Co.,
383 U.S. 1, 17–18 (1966). We analyze this asserted ground based on
obviousness with the principles identified above in mind.

22
IPR2018-00132
Patent 6,226,686

2. Level of Skill in the Art

The level of skill in the art is a factual determination that provides a
primary guarantee of objectivity in an obviousness analysis. Al-Site Corp. v.
VSI Int’l Inc., 174 F.3d 1308, 1324 (Fed. Cir. 1999) (citing Graham, 383 U.S.
at 17–18; Ryko Mfg. Co. v. Nu-Star, Inc., 950 F.2d 714, 718 (Fed. Cir.
1991)). Relying on the declaration of Dr. White, Petitioner asserts that a
person of ordinary skill in the art at the time of the invention would have to
be an individual with “at least a master’s degree (or equivalent course work)
in computer science, computer engineering, or physics, and at least two years
of experience related to networked interactive applications,” or a person with
“at least a bachelor’s degree in computer science, computer engineering, or
physics” who has “approximately four years’ experience in network[ing]
interactive applications, or the equivalent, which would include experience in
network programming.” Pet. 4 (citing Ex. 1007 ¶¶ 42–43). Patent Owner
does not assess the level of ordinary skill in the art.
We adopt the Petitioner’s, and Dr. White’s assessment of the level of
ordinary skill in the art because it is consistent with the ’686 patent and the
asserted prior art, and apply it to our obviousness evaluation below. We also
observe that the prior art of record reflects the level of ordinary skill in the
art. See Okajima v. Bourdeau, 261 F.3d 1350, 1355 (Fed. Cir. 2001); In re
GPAC Inc., 57 F.3d 1573, 1579 (Fed. Cir. 1995); In re Oelrich, 579 F.2d 86,
91 (CCPA 1978).
3. Aldred
Aldred, titled “Collaborative Working in a Network,” published as an
International Application on May 26, 1994 from an application filed on
November 10, 1993. Ex. 1009, at [54], [43], and [22]. Aldred relates to a

23
IPR2018-00132
Patent 6,226,686

“programmable workstation for collaborative working in a network,” which

includes a “collaborative application support subsystem for interfacing with
application programs,” wherein, “[t]he subsystem is responsive to
predetermined application program calls to create a logical network model of
a collaborative environment” that comprises a “sharing set[] of application
programs, which share data and resources across nodes and logical dedicated
data channels connecting members of the sharing set.”
Ex. 1009, at [57].
Figures 3 and 4, reproduced below, show a logical network model that
comprises sharing sets of application programs between various nodes.

24
IPR2018-00132
Patent 6,226,686

Figure 3 depicts aware applications sharing data and resources between

nodes A, B, C, D, and E. Id. at 5–6. Figure 4 depicts the two sharing sets
resulting from the sharing of the aware applications in Figure 3. Id.
An aware application initiates a share request, naming an application
sharing set, a target application and a destination node. Id. at 5. Support
software passes the request to a call manager at the sending node, which then
transfers it to the call manager at the destination node. Id. The sharing
mechanism can be cascaded; for example, two sharing applications can
initiate a share with a third application naming the same sharing set, with the
result that all three applications then share with each other. Id. Applications
in a sharing set establish data communication links with each other known as
channels. Id. at 6. As shown in Figures 3 and 4, various channels link aware
applications at nodes A, B, C, and E, and one channel links aware
applications at nodes D and E. See Figure 3 and 4. In particular, various
channels link aware application 1 at node A, aware application 2 at node B,
aware applications 3 and 4 at node C, and aware application 8 at node E,
which all belong to the same sharing set. One channel links aware
application 7 at node D and aware application 9 at node E, which belong to
the same sharing set. Id.
A serializing channel set feature combines data packets from different
channels, and deliver serialized packets to each application such that each
receiving port receives the same sequence of data. Id. at 7.14 Through this
synchronizing feature, “data packets on separate channels are tied together in

14
Aldred refers to “serialisation,” but unless quoting Aldred or the parties, we
refer to “serialization” throughout this Institution Decision.

25
IPR2018-00132
Patent 6,226,686

time (for example, voice with video), but delivered through the individual
ports belonging to the channels.” Id.
Figure 6, reproduced below, provides an example of a shared drawing
board using serializing channel set 59 consisting of channels 57 and 58:

With respect to Figure 6,

[t]wo identical applications, A and B (50 and 52), allow their
users to draw on a single shared surface. . . . [wherein] all []
drawing orders at [one node must be sent to the other node] in
such a way that the sequence processed at [one node is] identical
[to the one at the other.] This is accomplished by each [node]
transmitting [its] own data both to each other and to [itself], over
two channels 57 and 58 which are members of a common
serialising channel set 59.

Id. at 7.
Serialization can be implemented at a single central point with all data
being sent there for serialization and subsequent distribution. Id. at 9. Figure
19, reproduced below, shows the serializing process:

26
IPR2018-00132
Patent 6,226,686

Figure 19 depicts a collaborative group with established

communication between members, applications A, B, and C, through various
channels. Id. at 50. Each member application A, B, and C can receive data
packets from any of the other members that “arrive at all the receiving ports
in the identical sequence.” Id.
With respect to each application, logical channels consist of a
uni-directional data pipe with a sending port at one end and a receiving port
at the other. See id. With respect to the group, channels include one sending
port and many receiving ports. Id. (discussing Fig. 15).
To join, a new member participant must “know[] the application name
or an existing member and uses the same channel set name.” Id. at 50. Thus,
participants easily and regularly can join or leave the group. Id. at 47.
4. RFC 1692
RFC 1692, titled, “Transport Multiplexing Protocol (TMux),”
published August 1994, is a document specifying an Internet standards track
protocol for the Internet community, and requests discussion and suggestions
for improvements. Ex. 1010, at 1. RFC 1692 discloses that TMux is a

27
IPR2018-00132
Patent 6,226,686

protocol which allows multiple short transport segment, independent of

application type, to be combined between a server and host pair. Id. at
Abstract. RFC 1692 recognizes that “the network and host load could be
greatly reduced if traffic from multiple users, destined for the same host,
could be sent in the same packet.” Id. at 2. Thus, “TMux is designed to
improve network utilization and reduce the interrupt load on hosts which
conduct multiple sessions involving many short packets,” by “multiplexing
transport traffic onto a single IP datagram,” thereby “resulting in fewer,
larger packets.” Id.
RFC 1692 explains that “it is the overhead of processing a packet
which consumes a host’s resources, not the processing of the data.”
Id. “[T]he work a host must do to process a 1-octet packet is very nearly as
much as the work it must do to process a 1500-octet packet.” Id.
“[C]ommunication load is not measured only in bits per seconds but also in
packets per seconds,” and “[t]he proposed multiplexing is aimed at
alleviating this situation” of the “latter” as “the true performance limit” “in
many situations.” Id. The method presents demultiplexed segments of the
single IP datagram to the transport layer “as if they had been received in the
usual IP/transport packaging.” Id. at 2–3. The “[t]he transport layer is,
therefore, unaware of the special encapsulation” employed in multiplexing “a
set of transport segments into the same IP datagram.” Id. at 3.
Thus, RFC 1692 discloses achieving multiplexing by combining the
individual segments into a single message having an IP header followed by
all the segments, each being preceded by a “4 octet TMux mini header.”
Id. at 3. In other words, RFC 1692 discloses achieving multiplexing by
combining the individual packet bodies of different messages into a single

28
IPR2018-00132
Patent 6,226,686

message having an IP header followed by encapsulated segments, such that

each message includes a “TMux mini- header which specifies the segment
length and the actual segment transport protocol.” Id.
RFC 1692 depicts a TMux message as follows:

Id. In the TMux message represented above, “TM hdr” represents the mini-
header, and “IP hdr” represents the “Internet Protocol (IP) header.” See id.
RFC 1692 teaches that the disclosed multiplexing scheme involves
removing the same IP hdr for each segment: “TMux first removes the IP
header (if present) and adds a TMux mini-header and the segment to the
Multiplexed Message under construction for the host specified by the
destination address of the segment.” Id. at 6.
Only segments with the same Internet Protocol (IP)
header, (with the possible exception of the protocol and
checksum fields) may be combined. For example, the segment
(H1, B1) and the segment (H2, B2), where Hi and Bi are the
headers and the bodies of the segment, respectively, may be
combined (multiplexed) only if H=H1=H2. The combined
TMux message is either (H, B1, B2) or (H, B2, B1).

Id. at 3 (emphasis added).

5. Ulrich
Ulrich, titled, “Interactive Exercise Apparatus,” published as a U.S.
patent on November 14, 1995, from an application filed on February 1, 1994,
which in turn is a continuation-in-part of Serial No. 12,305, filed on February
2, 1993, now abandoned. Ex. 1012, at [54], [45], [22], [63]. Ulrich discloses
the interconnection of two or more exercise machines via computer
networking such that users of the exercise machines can interact with each

29
IPR2018-00132
Patent 6,226,686

other as teammates or competitors in a variety of athletic events. Id. at 2:8–

15. The processor runs many different programs to provide a variety of
simulated environments, including programs providing new worlds for the
user to explore or travel across the solar system. Id. at 2:41–46. A simulated
environment can be multi-dimensional to appear more realistic. Id. at 2:46–
48. Computers can communicate and share information and allow users to
navigate freely in the same simulated environment. Id. at 7:35–42.
Figure 19, reproduced below, shows a central hub that controls
communications between two or more exercise apparatuses.

Figure 19 shows a central hub that controls communications between

exercise apparatuses (nodes) by receiving information from all nodes and
directing information to all of, or a subset of all of the nodes. Id. at 3:45–49.
The central hub processor 104 ensures that each apparatus (node) in the
network receives updates about other users in the same general area of the
simulated environment. Id. at 8:64–9:10.
6. Denzer
Denzer, titled, “Method and Apparatus for Adding Data Compression
and Other Services in a Computer Network,” published as a U.S. patent on
April 26, 1994, from an application filed on July 19, 1991. Ex. 1014, at [54],

30
IPR2018-00132
Patent 6,226,686

[45], [22]. Denzer relates to data compression in the context of networked

computers. Id. at 1:7–9. In particular, Denzer discloses a host-to-host
interconnection of computers having Transmission Control Protocol/Internet
Protocol (TCP/IP) network capabilities for communicating across a Wide
Area Network. Id. at 5:22–26. When an application seeks to communicate
with a remote application, a request and data are passed to the host TCP/IP
layers, and a packet switch driver examines all such data and intercepts
selected data for routing to the local compress process. Id. at 5:56–66. Data
is selected for compression based upon operating parameters established.
Id. at 5:67–6:3. The compressed data is transmitted to the destination
computer for decompression and processing. Id. at 7:60–8:2.
7. Prior Art Printed Publication Status of RFC
1692
Patent Owner does not challenge our initial determination in the
Institution Decision that RFC 1692 (dated August 1994) was available to
persons of ordinary skill interested in computer networking and security
sufficiently to be deemed “publicly accessible” at the relevant time. See Inst.
Dec. 25–30; SRI Int’l, Inc. v. Internet Security Sys., Inc., 511 F.3d 1186, 1194
(Fed. 21 Cir. 2008). For the reasons discussed in the Institution Decision, we
maintain our preliminary determination that RFC 1692 qualifies as prior art
printed publications.
8. Analysis
a. Motivation to combine Aldred and RFC 1692
Petitioner contends “Aldred discloses the serialising and transmission
of data streams by a CSP [‘central serialisation point’],” where “[d]ata is sent
over channels in packets,” and where “each packet is added to a queue and

31
IPR2018-00132
Patent 6,226,686

then serially transmitted to each member of the Sharing Set associated with
the serialised channel set.” Pet. 34 (citing Ex. 1009, 9, 16, 51, and
Fig. 22; Ex. 1007 ¶ 142). Although Petitioner acknowledges that “Aldred
does not, however, explicitly disclose the claimed ‘aggregating,’” Petitioner
contends, “RFC 1692 renders the claimed ‘aggregating’ obvious.” Id.
According to Petitioner, RFC 1692 explains that “network and host load
could be greatly reduced if traffic from multiple users, destined for the same
host, could be sent in the same packet.” Id. (citing Ex. 1010, 2). That is,
“TMux is designed to improve network utilization and reduce the interrupt
load on hosts which conduct multiple session involving many short packets . .
. by multiplexing transport traffic onto a single IP datagram [2], thereby
resulting in fewer, larger packets.” Id.
Petitioner contends that “Aldred and RFC 1692 are both analogous art
to the 686 patent,” being all “from the fields of ‘computer network systems,’”
wherein “[i]t would have been obvious to an Ordinary Artisan in 1995 to
modify Aldred’s CSP to communicate with other nodes via RFC 1692’s
TMux protocol so as to ‘aggregat[e] said payload portions’ as claimed.”
Id. at 35 (citing Ex. 1007 ¶ 145). According to Petitioner, “Aldred also
supports ‘TCP/IP’” (id. (citing Ex. 1009, Fig. 10)), while “RFC 1692
discusses that although ‘[t]he use of the TMux protocol’ in ‘other situation[s]
may require some modification,’ it ‘may be applicable to other situations
where small packets are generated.’” Id. at 36 (citing Ex. 1010, 1–2).
Thus, “[a]n Ordinary Artisan would have understood that Aldred’s
system could likewise produce small packets that would benefit from RFC
1692’s multiplexing.” Id. (citing Ex. 1007 ¶ 146). Because “RFC 1692
explains that ‘network and host load could be greatly reduced if traffic from

32
IPR2018-00132
Patent 6,226,686

multiple users, destined for the same host, could be sent in the same packet,’”
the Ordinary Artisan “would have therefore been motivated to use TMux in
Aldred’s scheme for these benefits” because “TMux is an extension of the
Internet Protocol (IP)” which “Aldred itself states ‘has many benefits.’”
Id. at 36–37 (citing Ex. 1010, 2, 6, 10; Ex. 1007 ¶¶ 147–148; Ex. 1009, 4).
That is, “[a]s combined with RFC 1692, Aldred’s messages would be
multiplexed via TMux (‘aggregat[ed]’) into a single packet (‘aggregated
payload’) prior to transmission from the CSP to each member of the
serialising channel set associated with the Sharing Set.” Id. at 38 (citing
Ex. 1007 ¶¶ 150–151).
Patent Owner contends that “Petitioner’s alleged motivation for this
combination is based on its small packet centric argument that RFC 1692
provides benefits for systems generating a number of small packets.” PO
Resp. 17 (citing Pet. 36). Patent Owner contends that, however, “Petitioner
has failed to consider several key factors,” including as follows:

(1) the requirement for “order” in the serialization operation of

“all” packets in Aldred, and that order is not addressed in RFC
1692;
(2) the possibility of large packets in the serialization process of
Aldred;
(3) how using the TMux protocol disclosed in RFC 1692 could
disrupt the required order of the serialisation operation of Aldred
in light of the disclosure in RFC 1692 that large packets should
not be multiplexed and immediately transmitted; and
(4) why a POSITA would turn to RFC 1692 when RFC 1692
would introduce additional latency into the system of Aldred and
when Aldred already discusses alternative bandwidth solutions.

Id. at 18–19 (emphasis omitted).

33
IPR2018-00132
Patent 6,226,686

In particular, Patent Owner contends that the CSP of Aldred “has a

very specific function – ordering messages and transmitting the messages in
that particular order” (id. at 20 (citing Ex. 1009, 51)), and that “Petitioner
failed to consider the effects of large packets with respect to the serialization
process of Aldred.” Id. at 21 (emphasis omitted). According to Patent
Owner, “Petitioner has also failed to consider the effect of larger packets on
the packet order of the CSP of Aldred if the TMux protocol is included.”
Id. at 23.
Furthermore, Patent Owner contends that “Petitioner has also failed to
consider why a POSITA would turn to RFC 1692 when Aldred already
discusses alternative bandwidth saving solutions.” Id. at 29 (emphasis
omitted).
Based on the record before us, we are unpersuaded by Patent Owner’s
contention that “Petitioner’s analysis has provided an insufficient bases for
the combination of Aldred and RFC 1692.” PO Resp. 15, see also
id. at 50–51.
As an initial matter, we agree with Petitioner that Patent Owner
“unjustifiably assumes that every embodiment of Aldred would require the
use of ‘large’ TCP segments,” but “neither the challenged claims nor Aldred
require ‘large’ packets,” wherein “both encompass scenarios where only
small packets are generated.” Pet. Reply 7. According to Petitioner, Patent
Owner’s “‘large’ packet scenario does not show nonobviousness when the
claim and Aldred’s disclosure encompass systems that use only small
packets.” Id. at 8 (citing Apple Inc. v VirnetX Inc., IPR2014-00237,
paper 41, 40 (PTAB May 11, 2015)). That is, although Patent Owner
contends that “Petitioner’s alleged motivation for this combination is based

34
IPR2018-00132
Patent 6,226,686

on its small packet centric argument that RFC 1692 provides benefits for
systems generating a number of small packets” (PO Resp. 17 (citing
Pet. 36)), nothing in the challenged claims preclude a scenario in which small
packets are generated. We agree with Petitioner’s contention, to which
Patent Owner does not contest, that, in scenarios where only small packets
are generated, “[a]n Ordinary Artisan would have understood that Aldred’s
system could likewise produce small packets that would benefit from
RFC 1692’s multiplexing” (Pet. 36 (citing Ex. 1007 ¶ 146)) because the
Ordinary Artisan would have realized that “[a]s combined with RFC 1692,
Aldred’s messages would be multiplexed via TMux (‘aggregat[ed]’) into a
single packet (‘aggregated payload’) prior to transmission from the CSP to
each member of the serialising channel set associated with the Sharing Set.”
Pet. 38 (citing Ex. 1007 ¶¶ 150–151).
Here, according to Petitioner’s declarant, Dr. White, “[o]ne of ordinary
skill in the art would have understood that drawing orders and other events
used to keep data consistent between applications [in Aldred], such as user
input, would result in messages significantly smaller than the IP protocol
supports, such that RFC 1692’s methodology would improve Aldred’s
performance by reducing the number of packets.” Ex. 1007 ¶ 146. Although
Patent Owner provides testimony of its declarant, Dr. Almeroth’s testimony
tracks Patent Owner’s arguments but does not contradict Dr. White’s
testimony that Aldred discloses small packet applications (id.), because Dr.
Almeroth only points out that Aldred discloses some examples (e.g., video)
of “data packets much larger than the drawing orders of Aldred [that]
Petitioner uses as an example.” See Ex. 2002 ¶ 70.

35
IPR2018-00132
Patent 6,226,686

Nevertheless, as to Patent Owner’s contention that Petitioner has failed

to consider “the effect of larger packets on the packet order of the CSP of
Aldred if the TMux protocol is included” (PO Resp. 23), we are persuaded by
Petitioner’s argument that “RFC 1692 discusses adding segments to a
multiplexed message in the order they are received.” Pet. Reply 4 (citing
1010 ¶¶ 6–7). In particular, Petitioner argues that “RFC 1692 expressly
discloses that large TCP segments can be added to the end of any pending
multiplexed packet, such that all segments would be sent in the order
received.” Pet. Reply 3. According to Petitioner, “RFC 1692 explains that
when a segment that would not normally be multiplexed – such as a large
segment – is received and a multiplexed message is under construction, ‘the
extra segment can be added to the TMux message under construction, and
this complete message should be sent immediately . . . .’” Id. (citing
Ex. 1010, 8–9). That is, RFC 1692 discloses that “large and small TCP
segments can be sent in-order by the multiplexing process.” Id. at 4 (citing
Ex. 1053 ¶¶ 19–23).
In its Sur-Reply, Patent Owner does not dispute Petitioner’s point and
its declarant’s, Dr. White’s, testimony that the logical channels in Aldred
guarantee order. See generally Sur-Reply; Reply 5; Ex.1053 ¶ 25. Rather,
Patent Owner summarizes its position by, again, relying on its large packet
argument. Sur-Reply 3 (arguing “smaller items would be removed from the
serialization queue and added to the TMux buffer, and larger segments would
be transmitted immediately”); see also id. at 3–4 (similar large packet
arguments).
However, in Aldred, after demultiplexing, to the extent TMux does not
maintain packet order, the higher level channel logical channel layer

36
IPR2018-00132
Patent 6,226,686

maintains packet order. See Ex. 1009, 6 (defining a channel “by the sending
application” and “with application specified transmission characteristics,”
and stating “a receiving port . . . receives data packets from the channel in the
order in which they were sent”).
Furthermore, as Petitioner points out, by contending that combining
Aldred and RFC 1692 would create “out-of-order” TCP segments, Patent
Owner’s contentions are premised on the assumption that “out-of-order TCP
segments would go uncorrected by the combined system,” whereas “Aldred’s
scheme maintains the order of serialized updates regardless of the order of IP
packets transmitted by RFC 1692 through both Aldred’s channel mechanism
and the use of TCP.” Pet. Reply 5.
We are also unpersuaded by Patent Owner’s contention that a POSITA
would not have turned to the teachings of RFC 1692 because “Aldred already
discusses alternative bandwidth saving solutions.” PO Resp. 29. As
Petitioner contends, “the techniques by RFC 1692 were complementary to
the techniques discussed in Aldred,” wherein “Aldred’s disclosure of
‘alternative bandwidth solutions’ does not constitute a ‘teaching away’”
because it “does not criticize, discredit, or otherwise discourage the use of
another.” Pet. Reply 9 (citing In re Fulton, 391 F.3d 1195. 1200-1 (Fed Cir.
2004)). Citing to its declarant, Petitioner contends “[a]s Dr. White explained,
when there is a high-volume of small packets and high data redundancy,
TMux and compression are ‘both obvious things to do’ and ‘[y]ou can use
them both at the same time. They’re not exclusive.’” Id. (citing Ex. 2004,
54:21–55:20; Ex. 1053 ¶ 33). We agree with Petitioner that “‘[a] person of
ordinary skill is also a person of ordinary creativity, not an automaton.” Id.
(citing KSR at 419). That is, “obviousness requires only a showing ‘that a

37
IPR2018-00132
Patent 6,226,686

skilled artisan would have been motivated to combine the teachings of the
prior art references to achieve the claimed invention, and that the skilled
artisan would have had a reasonable expectation of success from doing so.’”
Id. at 10 (citing Bristol-Myers Squibb Co. v. Teva Pharms. USA, Inc., 752
F.3d 967, 973 (Fed. Cir. 2014)).
b. Independent Claims 1, 3, 7, 12, and 18 over Aldred
and RFC 1692

Petitioner primarily relies on Aldred to teach and suggest all the

limitations of independent claims 1, 3, 7, 12, and 18. Pet. 21–34, 41–48. To
the extent Aldred does not teach all the limitations of independent claims 1,
3, 7, 12, and 18, Petitioner relies on either RFC 1692 or knowledge of the
ordinary artisan to teach certain limitations. Id. at 34–41, 49.
In particular, Petitioner contends that “Aldred’s serialisation logic
transmits the same sequence of messages to each member of a Sharing Set in
the same order to maintain consistency between the applications,” such that
“the users . . . [would] see identical results,” and thus, “Aldred therefore
discloses that these messages contain a ‘payload portion’ in the form of the
information (e.g., an event or drawing order) in the packet used to maintain
consistency.” Id. at 29–30 (citing Ex. 1009, 5–7, 9, 16, 51, Figs. 19, 22;
Ex. 1007 ¶ 130). Petitioner also contends that although “Aldred
consequently does not explicitly disclose that ‘each of said [host] messages
contains . . . a portion that is used to identify said [] message group’ . . .
Aldred inherently discloses these feature.” Id. at 30. Citing its declarant,
Petitioner contends, “[a]s Dr. White explains, messages communicated
between Aldred’s nodes in [scenarios with users A, B, C, and D, each on a
different node, participating in a variety of different and Sharing Sets] ‘must

38
IPR2018-00132
Patent 6,226,686

necessarily identify the specific application sharing set to which they

pertain.’” Id. at 30–31 (citing Ex. 1007 ¶ 134).
Petitioner then contends that “RFC 1692 renders the claimed
‘aggregating’ obvious,” because “RFC 1692 discloses an improvement on the
Internet Protocol called the Transport Multiplexing Protocol (TMux), which
‘allows multiple short transport segments, independent of application type, to
be combined between a server and host pair.’” Id. at 33 (citing
Ex. 1010, 1). According to Petitioner, “TMux is designed to improve
network utilization and reduce the interrupt load on hosts which conduct
multiple sessions involving many short packets . . . by multiplexing transport
traffic onto a single IP datagram . . . thereby resulting in fewer, larger
packets.” Id. at 34 (citing Ex. 1010, 2).
In response, Patent Owner contends that “Petitioner has failed to
demonstrate that RFC 1692 discloses or suggests ‘aggregating said payload
portions of said host messages . . . to create an aggregated payload’ [as
recited in claims 1 and 3, and similarly recited in claims 7, 12, and 18].” PO
Resp. 33 (emphasis omitted). Patent Owner also contends that “Petitioner
has failed to demonstrate that RFC 1692 discloses or suggests ‘forming an
aggregated message using said aggregated payload’ [as recited in claims 1
and 12, and similarly recited in claims 3, 7, and 18].” Id. at 43.
We focus on the following limitations in the contested claims:

39
IPR2018-00132
Patent 6,226,686

i. “aggregating said payload portions of said

host messages received from said second subset of the
plurality of host computers to create an aggregated
payload” and “forming an aggregated message using
said aggregated payload ” (claims 1 and 3)
Petitioner relies on Aldred to teach sending data in packets over
channels, “where each packet is added to a queue and then serially
transmitted to each member of the Sharing Set associated with the serialized
channel set.” Pet. 34 (citing Ex. 1009, 9, 51, Fig. 22; Ex. 1007 ¶ 142).
Petitioner then relies on the disclosure of RFC 1692 to teach or suggest the
claimed “aggregating” obvious. Id. at 34–38 (citing Ex. 1002, 1:10–16; 826–
51; Ex. 1009, 2–4, 7, 9, 28–30, Fig. 10; Ex. 1010, 1–3, 6, 10; Ex. 1007
¶¶ 143–151).
In particular, Petitioner relies on RFC 1692 for teaching, “an
improvement on the Internet Protocol called the Transport Multiplexing
Protocol (TMux),” which “allows multiple short transport segments,
independent of application type, to be combined between a server and host
pair,” wherein “network and host load could be greatly reduced if traffic from
multiple users, destined for the same host, could be sent in the same packet.”
Id. at 34 (citing Ex. 1010 at 1–2). That is, “TMux is designed to improve
network utilization and reduce the interrupt load on hosts which conduct
multiple sessions involving many short packets . . . by multiplexing transport
traffic onto a single IP datagram,” thereby “resulting in fewer, larger
packets.” Id (citing Ex. 1010, 2). According to Petitioner, in TMux, “[t]he
multiplexed packet comprises an IP header followed by one or more TMux
mini-headers that specify respective transport segments (i.e., the original
payloads).” Id. at 34–35 (citing Ex, 1010, 3; Ex. 1007 ¶ 143).

40
IPR2018-00132
Patent 6,226,686

Accordingly, Petitioner relies on RFC 1692 to also teach and suggest

forming and creating an “aggregated message”/“server message,” as recited
in claims 1, 3, 7, 12, and 18. Id. at 39, 49. Petitioner contends, “RFC 1692’s
TMux functionality would construct a multiplexed message[sic] until a timer
expires, at which point the multiplexed is complete.” Id. at 39. According to
Petitioner, “[a] completed TMux packet comprises an IP header (‘IP hdr’)
followed by one or more TMux mini-headers (‘TM hdr’) that specify
following transport segments (‘Tport segment’).” Id. (citing Ex. 1010, 3).
In response, Patent Owner asserts that “[t]he ’686 Patent requires that
an aggregated payload is a collection of two or more data items that does not
include transport layer headers,” whereas “RFC 1692 instead discloses
combining separate messages together that retain transport layer headers as
aggregated packets rather than an ‘aggregated payload.’” PO Resp. 33
(emphasis omitted). That is, “RFC 1692 only discloses combining separate
messages together as a TMuxed message” (id.), whereas “the ’686 Patent
describes the formed aggregated message as comprising a message that has
only a single message header that is combined with aggregated payload
items.” Id. at 43.
Patent Owner acknowledges that each packet described in Figure 9 of
the ’686 patent, which Patent Owner refers to as a “payload,” includes more
than “actual data,” i.e., each payload includes “the source ULP address 117
of the transmitted payload element, the data length 118 of the payload
element and the actual data 119.” Id. at 45–46 (citing Ex. 1002, 13:1–14:69,
Fig. 9; Ex. 2002 ¶¶ 116–120). However, Patent Owner describes the TMux
protocol of RFC 1692 as describing “a message of combined [encapsulated]
messages”:

41
IPR2018-00132
Patent 6,226,686

The Tport segments comprise individual messages which may

be separately transported to a destination after they are broken
up from a received combination of messages as described with
respect to the TMux protocol. Id. After breaking them up, they
still have data, addresses, and transport headers encapsulated
therein, such that they can be transported and processed in
accordance with the transport layer.

Id. at 47 (emphasis added).

Accordingly, Patent Owner contends that “the TMux protocol of

RFC 1692, rather than forming a message containing an aggregated payload
as recited . . ., instead describes a message including combined messages.”
Id. That is, contrary to the message in RFC 1692, “[a] formed aggregated
message or server message of the ’686 Patent comprises a single transport
layer message header and an aggregated payload, generated by an
aggregating step; an aggregated message of the ’686 Patent is not comprised
simply of combined messages as disclosed in RFC 1692.” Id. at 50 (citing
Ex. 2002 ¶ 130).
Patent Owner’s argument that “[t]he ’686 Patent requires that an
aggregated payload is a collection of two or more data items that does not
include transport layer headers” (PO Resp. 33) does not undermine
Petitioner’s showing, because we do not adopt Patent Owner’s proposed
claim construction. See supra Section II (A)(1)–(2). In line with our claim
construction, the claim term “aggregated payload,” consistent with its
ordinary and plain meaning in light of the Specification, encompasses “a
collection of two or more data items that may include headers transported as
data.” Supra Section II (A)(2).
As Patent Owner’s declarant, Dr. Almeroth, testified in a previous
proceeding, the “process of adding a layer header to the data from the

42
IPR2018-00132
Patent 6,226,686

preceding layer is sometimes referred to as ‘encapsulation’ because the data

and layer header is treated as the data for the immediately following layer,
which, in turn, adds its own layer header to the data from the preceding
layer.” Ex. 1056 ¶ 68 (emphasis added). Here, Patent Owner does not
dispute that its disclosed header information (e.g., addresses) may be
encapsulated. See PO Resp. 8 (“Patent Owner’s construction position is not
that the term ‘aggregated message’ does not encompass encapsulated
headers”). Patent Owner’s own description of the ’686 patent shows that the
patent describes combining different “payloads,” each of which include ULP
address and data length header information, similar to the process of TMux,
which according to Patent Owner, includes “data, addresses, and transport
headers encapsulated therein.” See PO Resp. 47 (emphasis added). In other
words, “aggregate message” may constitute or comprise packets with
additional header information “encapsulated therein,” as Patent Owner
recognizes. See id. at 46 (quoting Ex. 1010, 2–3); Ex. 1010, 2–3 (describing
“special encapsulation”).
According to RFC 1692, and as quoted by Patent Owner, the TMux
message involves “special encapsulation” of the underlying transport
segments. PO Resp. 36 (quoting Ex. 1010, 2–3). That is, Patent Owner
agrees that RFC 1692’s TMux process involves encapsulation. See PO
Resp. 47 (arguing the transport segments “still have data, addresses, and
transport headers encapsulated therein, such that they can be transported and
processed in accordance with the transport layer”).
In RFC 1692, the TMux message includes a single IP header (IP hdr)
“equal to H,” so the process strips out the IP header H for at least some of the
segments (H, B1), (H, B2), (H, Bi) prior to combining them, and then

43
IPR2018-00132
Patent 6,226,686

combines (i.e., multiplexes) the two segments into aggregated packets having
the TMux structure “(H, B1, B2) or (H, B2, B1),” “where Hi and Bi are the
headers and the bodies of the segment, respectively.” Ex. 1010, 3. In other
words, “TMux removes the IP header (if present) and adds a TMux mini-
header and the segment to the Multiplexed Message under construction.”
Id. at 6. Thus, the TMux process involves at least two steps, similar to the
two disclosed steps of the ’686 patent (as characterized by Patent Owner (PO
Resp. 38–39)), and it involves encapsulation in which the system treats the
bodies B1 and B2 (which include mini-headers) as data relative to the single
IP header H and IP layer. See Ex. 1010, 6.
As Petitioner points out, the ’686 patent states “the TLP protocol is
TCP/IP” wherein the ’686 patent “refers to a ‘TCP/IP header’ as a single
header” (Pet. Reply 18–19 (citing Ex. 1002, 26:28–29)), and thus, “removing
a TLP header encompasses removing the IP header alone.” Id. at 19.
Similarly, in RFC 1692, the TMux process “first removes the IP header (if
present). See Ex. 1010, 6.
Therefore, even if we adopt Patent Owner’s construction, according to
Petitioner, “[a]s explained in the Petition and by Dr. White, Aldred in view of
RFC 1692 results in an ‘aggregated payload’ without any IP headers and an
‘aggregated message’ with a single IP header.” Pet. Reply 19 (citing
Pet. 36–38; Ex.1007, ¶¶ 68–76, 143; Ex.1053 ¶ 34).
In its Sur-Reply, addressing Petitioner’s argument that “Aldred
encourages multiplexing at lower layers—the very technique to which
RFC 1692 is directed’ (Pet. Reply 9), Patent Owner contends Petitioner takes
Aldred’s “statement . . . out of context.” Sur-Reply 5 (citing Ex. 1009, 5).
According to Patent Owner, “[m]ultiplexing in Aldred refers to the

44
IPR2018-00132
Patent 6,226,686

‘separation of data traffic,’ such that ‘voice, video and data traffic . . . can be
sent over multiple channels’ so that ‘data components are presented
individually.’” Id. (citing Ex. 1009, 30). Therefore, according to Patent
Owner, “Aldred does not encourage the same technique to which RFC 1692
is directed – combining multiple messages into a single message for
transmission.” Id.
However, Patent Owner’s argument (id.) mischaracterizes Aldred’s
teachings. Aldred describes “separation of data traffic into logical . . . flows
of data” so that “[d]ata multiplexing can be implemented in different ways
depending on the underlying transport mechanism,” including, as an
example, by “multiplexing all the data in the support layer.” See
Ex. 1009, 30. Contrary to Patent Owner’s arguments, an artisan of ordinary
skill would have understood that multiplexing in Aldred and RFC 1692
involves combining data units or packets––the opposite of separating. See
Ex. 1010, 3 (“The TMux Protocol is defined to allow the combining of
transmission units of different higher level protocols in one transmission unit
of a lower level protocol.” (emphasis added)).
Also, in Aldred, even if data separation occurs prior to multiplexing,
this supports Petitioner’s small packet theory, in which the claims embrace
small packets disclosed by Aldred and multiplexed by RFC 1692, because the
flow of small packet data from a chalkboard application in Aldred would be
separated from, for example, larger video packets. See Ex. 1053 ¶ 26 (“An
Ordinary Artisan would understand the ‘chalkboard’ application can send
only small messages and the ‘file transfer’ or ‘voice/video link’ applications
can send only larger messages”).

45
IPR2018-00132
Patent 6,226,686

By contending that “RFC 1692 only discloses combining separate

messages together as a TMuxed message,” and “does not disclose
aggregating payload portions to create an aggregated payload” (PO
Resp. 33), Patent Owner appears to attempt to distinguish combining
messages from combining packet payloads by emphasizing the combining of
“messages” in the TMux process. However, Patent Owner does not set forth
a material difference between combining messages that include payloads and
combining payloads. See PO Resp. 33, 42.
Nevertheless, Aldred’s messages, like those of the TMux protocol,
come in packet form, and as described above, the ’686 patent’s disclosed
system combines more than packet “actual data.” See Ex. 1002, 14:41–47
(“A single payload element consists of a triplet of source ULP address [117],
data length [118], and data [119],” where “item 119 is the actual data”);
Ex. 1009, 7 (describing “data packets”), 64 (disclosing TCP/IP); Ex. 1010, 16
(“Data is sent over channels by applications in packets; at the physical level
the unit of data transmission is a frame.”) According to RFC 1692, “[t]he
TMux protocol is intended to optimize the transmission of large numbers of
small data packets.” Ex. 1010, 1 (emphasis added). “It does this by
multiplexing transport traffic onto a single IP datagram [2], thereby resulting
in fewer, larger packets.” Id. at 2 (emphasis added).
In addition, the ’686 patent generally refers to aggregating the
“contents of messages.” Ex. 1002, [57] (“Rather than simply forward each
message to its targeted hosts, the group messaging server aggregates the
contents of each of messages received during a specified time period and
then sends an aggregated message to the targeted hosts.” (emphasis added)).
The ’686 patent also states “[a] key concept in the present invention is the

46
IPR2018-00132
Patent 6,226,686

aggregation of multiple messages in a message queue into a single ULP

receive message to a host that contains multiple payload items in the
payload.” Id. at 23:50–53 (emphasis added).
Here, claim 1 of the ’686 patent recites “aggregating . . . said payload
portions” and “forming an aggregated message using said aggregated
payload,” but the Specification does not restrict a payload from containing
multiple items, including address information See Ex. 1002, [57], 14:41–47,
27:65–28:2. Neither does it restrict the process from combining message
portions that contain payloads to create an aggregated message of aggregated
payloads. Id. As noted, the ’686 patent generally allows “aggregate[ing] the
contents of . . . messages,” wherein aggregating messages also includes
aggregating payload items. See id. [57]. By stripping out the IP header of
each packet segment, RFC 1692’s TMux involves aggregating payload items
as called for by the challenged claims in light of the ’686 Specification. See
Ex. 1010, 6 (“TMux first removes the IP header (if present)”).
Based on the foregoing, and after consideration of Patent Owner’s
remaining arguments addressed below, Petitioner shows that the combination
of Aldred and RFC 1692 would have suggested the claimed aggregating and
forming steps, including the steps of aggregating packet portions using a
well-known TMux protocol in order to allow hosts to form a message of
several packet messages with a single IP header instead of individually
processing many message packets individually header by header. On this
record, we are persuaded Petitioner has demonstrated by a preponderance of
the evidence that the combination of Aldred and RFC 1692 teaches or
suggests these claimed limitations of claims 1, 3, 7, 12, and 18.
ii. Remaining limitations

47
IPR2018-00132
Patent 6,226,686

Petitioner provides explanations and supporting evidence regarding the

remaining limitations in independent claims 1, 3, 7, 12, and 18. See
generally Pet. 21–49. We have reviewed Petitioner’s explanations and
supporting evidence regarding these remaining limitations, as well as
Petitioner’s assertion that a person of ordinary skill in the art would have had
a sufficient reason to combine or modify the teachings of the references, and
we find them sufficiently persuasive. Id. We adopt Petitioner’s showing as
our own.
Patent Owner does not address separately Petitioner’s showing as to
these added limitations. See generally Pet. Resp. Based on the foregoing and
a review of the record, Petitioner demonstrates that the combination of
Aldred and RFC 1692 would have rendered obvious these remaining
limitations of independent claims 1, 3, 7, 12, and 18.
c. Claims 26, 27, 45, 46, 62, and 63 over Aldred
and RFC 1692

Petitioner provides explanations and supporting evidence regarding

dependent claims 26, 27, 45, 46, 62, and 63. See generally Pet. 65–66. We
have reviewed Petitioner’s explanations and supporting evidence regarding
these remaining limitations, as well as Petitioner’s assertion that a person of
ordinary skill in the art would have had a sufficient reason to combine or
modify the teachings of the references. Id. We adopt Petitioner’s showing as
our own.
Patent Owner does not address separately Petitioner’s showing as to
these claims but merely contends that “Petitioner has failed to prove”
obviousness of the claims “[f]or at least” the reasons “discussed herein.” See
Pet. Resp. 50–51. Based on the foregoing and a review of the record,

48
IPR2018-00132
Patent 6,226,686

Petitioner also demonstrates that the combination of Aldred and RFC 1692
would have rendered obvious claims 26, 27, 45, 46, 62, and 63.
d. Claims 22–27, 41–46, and 58–63 over Aldred,
RFC 1692, and Ulrich

Petitioner relies on Ulrich for suggesting the steps of “performing, by

said server, echo suppression” (claims 22, 41, and 58), “changing
membership of said message group” (claims 24–26, 43–45, and 60–62),
“wherein said plurality of host computers belonging to said message group
correspond to players that are in close proximity to one another within a
three-dimensional space of a computer game” (claims 23, 42, and 59), and
“wherein said application is a game” (claims 27, 46, and 63). See Pet. 51–66.
In particular, Petitioner asserts that Ulrich discloses “a networked
interactive software system for playing games in a virtual world that supports
‘a variety of simulated environments.’” Pet. 52 (citing Ex. 1012, 1:63–2:7,
2:41–55). According to Petitioner, “echo suppression ‘means that the host
will not receive a copy of its own message to the group,’” and since Ulrich
“discloses directing ‘updates about other user’ to each apparatus,” Ulrich
“suggests performing ‘echo suppression.’” Id. at 61–62 (citing
Ex. 1002, 22:66–23:7; Ex. 1012, 9:5–10). Further, Petitioner contends that
Ulrich discloses a “network topology involving a central hub that ‘controls
communication between two or more exercise apparatus (‘nodes’) by
receiving information from all nodes and directing information to all of, or to
a subset of all of, the nodes,’” wherein “[t]his ‘large-scale network’ hub
receives all location updates from the participating machines and transmits
updates to other machines ‘in the same general area of the simulated
environment.’” Id. at 53 (citing Ex. 1012, 3:45–49, 8:64–9:10, Fig. 8;

49
IPR2018-00132
Patent 6,226,686

Ex. 2002 ¶ 242).

According to Petitioner, a POSITA would have found it obvious to
combine the CSP of Aldred to “(1) implement Ulrich’s game simulation
system and (2) distribute location updates from a CSP only to ‘users in the
same general area of the simulated environment.’” Id. (citing Ex. 1007
¶¶ 247–256; Ex. 1012, 9:5–10). Citing its declarant, Dr. White, Petitioner
contends, combining the echo suppression of Ulrich with the CSP of Aldred
“would have been the product of ordinary skill and common sense, and
would have been obvious to try,” wherein echo suppression would “reduce
the amount of information transmitted from the hub to the apparatuses.”
Id. at 63 (citing Ex. 1007 ¶¶ 280–281).
Petitioner also contends that “[b]ecause ‘travel in the simulated
environment is substantially restricted,’ . . . the choices a user makes in their
travel will change which ‘general area of the simulated environment’ they are
in,” which would then “change which set of other users is within ‘the same
general area’ as that newly located user,” which would “‘chang[e]
membership of said first message group’ (the group of hosts that receive that
update)” as claimed. Pet. 64 (citing Ex. 1012, 2:41–55, 9:5–10; Ex. 1007
¶ 259).
With respect to the “echo suppression” limitation, Patent Owner
contends, “[t]he described function of the CSP of Aldred would have to be
substantially altered to force the CSP to include echo suppression [of
Ulrich]” because “Aldred clearly discloses that the CSP should send a node’s
own data back to that node.” PO Resp. 53 (citing Ex. 2002 ¶¶ 167–168).
Further, because “Ulrich does not disclose that data must be received in any
specific order, . . . [a] POSITA would thus be discouraged from attempting to

50
IPR2018-00132
Patent 6,226,686

combine the CSP of Aldred, which requires a specified order, with the game
of Ulrich.” Id.
With respect to the “plurality of host computers belonging to said
message group correspond to players that are in close proximity to one
another” limitation, Patent owner contends “in such a combination as
proposed by Petitioner, the members of a sharing set would not correspond to
players that are in close proximity,” but rather “the members of the sharing
set would include all players in the game.” PO Resp. 55–56 (citing Ex. 2002
¶¶ 174–175). In particular, Patent Owner contends, in a combination of
Aldred and Ulrich, “all players/nodes in Aldred would issue ‘shareapp’
requests and join a single sharing set so that each player/node can potentially
receive all updates for the game,” and thus, for this combination to work with
the disclosure of Aldred, “all the players would still need to be in the sharing
set,” and thus, “there could be application/nodes in the sharing set that do not
correspond to players in close proximity to one another.”
Id. at 56 (citing Ex. 2002 ¶ 176).
Further, Patent Owner contends that the combination of Aldred,
RFC 1692 and Ulrich “merely provides that a CSP selectively transmits
communications to certain applications/nodes in a sharing set based on player
proximity in a 3D environment,” but “[m]embership of the sharing set would
not change.” Id. at 58 (citing Ex. 2002 ¶ 181).
Based on the record before us, we are unpersuaded by Patent Owner’s
contentions.
With respect to the “echo suppression” limitation, Patent Owner’s
argument that “[t]he described function of the CSP of Aldred would have to
be substantially altered to force the CSP to include echo suppression” (PO

51
IPR2018-00132
Patent 6,226,686

Resp. 53) fails to address Petitioner’s reliance on Aldred’s other channel

arrangements where “the transmitter does not automatically receive a copy of
an update that it sends to a Sharing Set.” Pet. 62 (citing Ex. 1009, Figs. 15,
16, 18). That is, according to Petitioner, “Aldred already supports echo
suppression via merged channels.” Pet. Reply 24. Further, as Petitioner
points out, Patent Owner also fails to address Petitioner’s proffered network
improvement rationale for providing Ulrich’s echo suppression in Aldred’s
system––i.e., improving network performance by limiting the transmitted
data via echo suppression. See Reply 24–25 (noting Patent Owner
acknowledges, but does not address, the Petition’s showing “that echo
suppression could reduce the amount of data” (citing Resp. 52–53; Pet. 63;
Ex.1007 ¶¶ 254, 280; Ex.1057, 5)).
Here, Petitioner persuasively describes myriad similarities between
Ulrich’s hub and Aldred’s CSP processes and topologies. See Pet. 59 (citing
Ex. 1009, 7–9; Ex. 1012, 8:64–9:10). Citing the testimony of Dr. White,
Petitioner explains persuasively that modifying Aldred’s similar CSP
functionality to include Ulrich’s echo suppression “would . . . have been the
product of ordinary skill,” and would also arrange prior art elements (i.e.,
Aldred’s CSP (as combined) and Ulrich’s hub message reduction technique),
each performing their known function, to yield no more than one would
expect from such an arrangement: Aldred’s CSP passing updates along only
to those members of the Sharing Set “in the same general area of the
simulated environment.” See Pet. 61 (citing Ex. 1007 ¶ 255). As Petitioner
also explains, even if some alteration must occur, “the degree of alteration,
without more, is not relevant to obviousness, In re Mouttet, 686 F.3d 1322,
1332–33 (Fed. Cir. 2012), and Aldred already supports echo suppression.”

52
IPR2018-00132
Patent 6,226,686

Pet. Reply 24 (quoting Pet. 62 (citing Ex.1009, Figs. 15, 16, 18)). Petitioner
persuasively adds that using Aldred’s central point in a merged (echo
suppression) mode would have been obvious, because Ulrich’s “direct hub
embodiment is a central point” and involves “a star topology, which was a
well-known way of connecting network nodes” to “provide anonymity and
coordination benefits.” Id. (citing Ex.1007 ¶¶ 252–57, 280; Ex.1012, 5–10;
Ex.1030, Fig. 22, 32:47–60; Pet 59–60).
Patent Owner also argues that combining Aldred with Ulrich “would
create unnecessary complexity” and “latency” due to serialization. See
Resp. 53–54. These arguments ignore Petitioner’s demonstration of the
obviousness involving “using a central point without serialization,” in other
words, Petitioner relies on a merge channel mode in Aldred, as explained
above. See Pet. Reply 25 (citing Pet. 62–64; Ex. 1007 ¶¶ 280–81).
With respect to “plurality of host computers belonging to said message
group correspond to players that are in close proximity to one another”
limitation, Patent Owner’s arguments do not undermine Petitioner’s showing.
Here, Petitioner proposes that the combination suggests distributing location
updates “via a central point only to users in the same general area of the
simulated environment”––i.e., users in “said first message group.” See
Reply 26 (citing Pet. 51–61; Ex.1007 ¶¶ 252–57). In other words, as
Petitioner explains, “Ulrich’s subset of Aldred’s Sharing Set is a ‘message
group’” as claimed. See id. (citing Ex. 1052, 100:3–12, 105:19–106:24). As
Petitioner also explains, “[t]his mapping is . . . consistent with the ’686
patent’s usage of ‘message group.’” Id. at 27 (citing Ex. 1002, 10:5–18,
11:20–29; 1052, 108:25–111:4). As an example in the section cited by
Petitioner, the ’686 patent describes the following scenario:

53
IPR2018-00132
Patent 6,226,686

Consider a computer game for multiple players that supports

hundreds of players that are spread throughout a three
dimensional space created by the game. At any time only a few
players will be able to see and effect one another in the game
since other players will be in other areas that are out of sight. . .
. It is only necessary to send data between the players that are
in close proximity to one another.

Ex. 1002, 10:6–16 (emphasis added). This description in the ’686 patent
closely tracks Petitioner’s proposed combination involving distributed
location updates to users in the same general area of the simulated
environment as Ulrich teaches.
Although Patent Owner contends that in Ulrich, “[m]embership of the
sharing set would not change.” (PO Resp. 58), this argument attempts to
constrict the claimed membership to membership in a static sharing set of
Aldred. However, the claims at issue (see, for example, claim 24) allow
membership to change based on virtual activities or location changes. As
Petitioner explains, “the set of computers that receive Ulrich’s location-based
updates (i.e., the ‘membership’ of that ‘message group’) changes based on
player activity.” Pet. Reply 27 (citing Pet. 64–65). As Petitioner also notes,
Aldred’s system similarly allows membership to change, further suggesting
the combination as suggested by Ulrich’s location-based updates provided
only to users who change membership by exiting and entering the same
general virtual area. See Pet. Reply 26 (citing Ex.1009, 49–51; Ex.1007
¶¶ 66, 105).
Based on the foregoing discussion, Petitioner shows persuasively that
the combination of Aldred, RFC 1692, and Ulrich render obvious claims 22–
27, 41–46, and 58–63.
e. Claims 36 and 55 over Aldred, RFC 1692 and Denzer

54
IPR2018-00132
Patent 6,226,686

Petitioner provides explanations and supporting evidence regarding

dependent claims 36 and 55. See generally Pet. 66–69. We have reviewed
Petitioner’s explanations and supporting evidence regarding these remaining
limitations, as well as Petitioner’s assertion that a person of ordinary skill in
the art would have had a sufficient reason to combine or modify the teachings
of the references. Id. We adopt Petitioner’s showing as our own.
Patent Owner does not address separately Petitioner’s showing as to
these claims but merely repeats that “Petitioner has failed to show that Aldred
and RFC 1692 disclose or suggest all the elements of claims 1, 3, 7, 12, and
18” from which claims 36 and 55 depend, and that “a POSITA would not
have been motivated to combine Aldred and RFC 1692.” See Pet. Resp. 59–
60. Based on the foregoing discussion, Petitioner shows persuasively that the
combination of Aldred and RFC 1692, in further view of Denzer, renders
obvious claims 36 and 55.

III. CONCLUSION

For the foregoing reasons, we are persuaded on the record at hand that
Petitioner has demonstrated by a preponderance of the evidence that claims 1,
3, 7, 12, 18, 22–27, 36, 41–46, 55, and 58–63 are unpatentable under
35 U.S.C. § 103 as obvious over Aldred and RFC 1692; over Aldred,
RFC 1692 and Ulrich; or over Aldred, RFC 1692 and Denzer.

IV. ORDER

For the reasons given, it is

ORDERED that Petitioner has shown by a preponderance of the
evidence that claims 1, 3, 7, 12, 18, 22–27, 36, 41–46, 55, and 58–63 of the

55
IPR2018-00132
Patent 6,226,686

’686 patent are unpatentable under 35 U.S.C. § 103 as obvious over Aldred
and RFC 1692; Aldred, RFC 1692 and Ulrich; or Aldred, RFC 1692 and
Denzer; and
FURTHER ORDERED that this is a Final Written Decision under
35 U.S.C. § 318(a), and that parties to the proceeding seeking judicial review
of the decision under 35 U.S.C. § 319 must comply with the notice and
service requirements of 37 C.F.R. § 90.2.

PETITIONER:

Joseph A. Micallef
Samuel A. Dillon
SIDLEY AUSTIN LLP
[email protected]
[email protected]

PATENT OWNER:

Gregory M. Howison
Keith D. Harden
Brian D. Walker
MUNCK, WILSON, MANDALA, LLP
[email protected]
[email protected]
[email protected]

56