Jenam Tech, LLC

Patent Trials and Appeals Board

Aug 27, 2021

IPR2020-00845 (P.T.A.B. Aug. 27, 2021)

Trials@uspto.gov Paper 33
571-272-7822 Date: August 27, 2021

UNITED STATES PATENT AND TRADEMARK OFFICE

BEFORE THE PATENT TRIAL AND APPEAL BOARD

GOOGLE LLC, LG ELECTRONICS, INC., and
LG ELECTRONICS U.S.A., INC.,
Petitioner,
v.
JENAM TECH, LLC,
Patent Owner.

IPR2020-00845
Patent 9,923,995 B1

Before DANIEL J. GALLIGAN, SCOTT B. HOWARD, and
JASON M. REPKO, Administrative Patent Judges.
GALLIGAN, Administrative Patent Judge.

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

IPR2020-00845
Patent 9,923,995 B1
2
I. INTRODUCTION
In this inter partes review, Google LLC, LG Electronics, Inc., and LG
Electronics U.S.A., Inc. (“Petitioner”) challenge the patentability of
claims 1–24 of U.S. Patent No. 9,923,995 B1 (“the ’995 patent,” Ex. 1001),
which is assigned to Jenam Tech, LLC (“Patent Owner”).
We have jurisdiction under 35 U.S.C. § 6. This Final Written
Decision, issued pursuant to 35 U.S.C. § 318(a), addresses issues and
arguments raised during the trial in this inter partes review. For the reasons
discussed below, we determine that Petitioner has proven by a
preponderance of the evidence that claims 1–24 of the ’995 patent are
unpatentable. See 35 U.S.C. § 316(e) (2018) (“In an inter partes review
instituted under this chapter, the petitioner shall have the burden of proving a
proposition of unpatentability by a preponderance of the evidence.”).
A. Procedural History
Petitioner requested inter partes review of claims 1–24 of the ’995
patent on the following grounds:
Claim(s) Challenged 35 U.S.C. §1 Reference(s)/Basis
1–24 103 Eggert2
1–24 103 Eggert, Hankinson3

1 The Leahy-Smith America Invents Act (“AIA”) included revisions to
35 U.S.C. §§ 102 and 103 that became effective March 16, 2013. The
application for the ’995 patent was filed on September 3, 2017. Although
the ’995 patent claims priority to applications filed before March 16, 2013,
Patent Owner has not shown that the written description of the earlier
applications supports the challenged claims. See Dynamic Drinkware, LLC
v. Nat’l Graphics, Inc., 800 F.3d 1375, 1378–80 (Fed. Cir. 2015); Tech.
Licensing Corp. v. Videotek, Inc., 545 F.3d 1316, 1327 (Fed. Cir. 2008).
Therefore, we apply the post-AIA versions of 35 U.S.C. §§ 102 and 103.
2 L. Eggert, “TCP Abort Timeout Option,” Apr. 14, 2004 (Ex. 1004).
3 Hankinson, EP 1 242 882 B1, published Apr. 20, 2005 (Ex. 1005).
IPR2020-00845
Patent 9,923,995 B1
3
Claim(s) Challenged 35 U.S.C. §1 Reference(s)/Basis
16 103 Eggert, Hankinson, RFC 11224
Paper 3 (“Pet.”), 11. Patent Owner filed a Preliminary Response. Paper 8.
With our authorization, Petitioner filed a Reply to the Preliminary Response
(Paper 13) addressing 35 U.S.C. § 325(d). We instituted trial on all grounds
of unpatentability. Paper 16 (“Inst. Dec.”), 27.
During the trial, Patent Owner filed a Response (Paper 18, “PO
Resp.”), Petitioner filed a Reply (Paper 23, “Pet. Reply”), and Patent Owner
filed a Sur-reply (Paper 27, “PO Sur-reply”).
An oral hearing was held on July 8, 2021, a transcript of which
appears in the record. Paper 32 (“Tr.”).
B. Related Matters
As required by 37 C.F.R. § 42.8(b)(2), the parties identify various
related matters. Pet. 77; Paper 5, 1. A different petitioner (Unified Patents,
LLC) challenged claims of the ’995 patent in IPR2020-00742, but that case
was terminated due to settlement after institution of trial. Unified Patents
LLC v. Jenam Tech LLC, IPR2020-00742, Paper 31 (PTAB June 29, 2021).
C. Real Parties in Interest
Petitioner identifies the following real parties in interest: LG
Electronics, Inc., LG Electronics U.S.A., Inc., Google LLC, Samsung
Electronics Co., Ltd., and Samsung Electronics America, Inc. Pet. 76.
Patent Owner identifies itself as the real party in interest. Paper 5, 1.

4 RFC 1122, “Requirements for Internet Hosts - - Communication Layers,”
Oct. 1989 (Ex. 1007).
IPR2020-00845
Patent 9,923,995 B1
4
D. The ’995 Patent and Illustrative Claim
The ’995 patent states that some transmission control protocol (TCP)
implementations support a keep-alive option, but the ’995 patent states that
“[e]ach node supports or does not support keep-alive for a TCP connection
based on each node’s requirements without consideration for the other node
in the TCP connection.” Ex. 1001, 1:43–45, 1:53–56. According to the
’995 patent, “[t]o date no mechanism to allow two TCP connection
endpoints to cooperate in supporting the keep-alive option has been
proposed or implemented.” Ex. 1001, 2:14–17. Thus, the ’995 patent states
that “there exists a need for methods, systems, and computer program
products for sharing information for detecting an idle TCP connection.”
Ex. 1001, 2:21–23.
Claims 1 and 11 are illustrative and are reproduced below.
1. An apparatus comprising:
a non-transitory memory storing instructions; and
one or more processors in communication with the non-
transitory memory, wherein the one or more processors execute
the instructions for:
receiving, by a second node from a first node, a
transmission control protocol (TCP)-variant packet in advance of
a TCP-variant connection being established;
detecting an idle time period parameter field in the TCP-
variant packet;
identifying metadata in the idle time period parameter
field for an idle time period and, during which, no packet is
communicated in the TCP-variant connection to keep the TCP-
variant connection active; and
modifying, by the second node and based on the metadata,
a timeout attribute associated with the TCP-variant connection.

IPR2020-00845
Patent 9,923,995 B1
5
11. The apparatus of claim 1 wherein the one or more
processors execute the instructions for:
detecting the idle time period based on the timeout
attribute; and
in response to detecting the idle time period, deactivating
the TCP-variant connection by releasing a resource allocated for
the TCP-variant connection by one of the first and second nodes
without signaling another one of the first and second nodes that
is related to the detection of the idle time period.

II. ANALYSIS
A. Level of Ordinary Skill in the Art
Petitioner contends that a person of ordinary skill in the art
would have an undergraduate degree in electrical engineering,
computer engineering, computer science or a related field along
with at least two years of work experience in the field of
networking, or in the alternative, equivalent experiences, such as
6 years of work or research experience in the field of networking.
Pet. 10 (citing Ex. 1003 ¶¶ 29–32). Patent Owner largely agrees with
Petitioner’s assessment:
Patent Owner does not contest Petitioners’ assertion that a person
of ordinary skill in the art (“POSITA”) would have had (i) an
undergraduate degree in electrical engineering, computer
engineering, computer science or a related field and (ii) at least
two years of work experience in the field of networking. A
higher level of education (e.g., a master’s degree) may make up
for less work experience, and additional work experience (e.g.,
5-6 years) may make up for less education.
PO Resp. 9–10 (citing Ex. 1003, 7–85).

5 Patent Owner cites pages 7–8 of the declaration of Mr. Bradner, who is
Petitioner’s declarant. Paragraphs 29–32, which Petitioner cites as support
for its proposal, appear at pages 7–8.
IPR2020-00845
Patent 9,923,995 B1
6
Thus, the parties are in general agreement as to the level of ordinary
skill in the art, and both cite the same testimony of Petitioner’s declarant in
support. We note that Petitioner’s alternative qualifications specify the field
of networking, whereas Patent Owner’s do not, although we view this as
implied by Patent Owner’s argument. Because Petitioner more precisely
specifies the level of skill and because Petitioner’s proposed level of skill is
supported by evidence cited by both parties, we adopt Petitioner’s
description, with the exception of the open-ended language “at least.” See
Ex. 1003 ¶¶ 30–32.
B. Claim Construction
We interpret claim terms using “the same claim construction standard
that would be used to construe the claim in a civil action under 35 U.S.C.
282(b), including construing the claim in accordance with the ordinary and
customary meaning of such claim as understood by one of ordinary skill in
the art and the prosecution history pertaining to the patent.” 37 C.F.R.
§ 42.100(b) (2019).
1. Idle Time Period
The focus of this trial is whether or not the asserted prior art teaches
an “idle time period,” as recited in claims 1 and 15. Claim 1 recites
“identifying metadata in the idle time period parameter field for an idle time
period and, during which, no packet is communicated in the TCP-variant
connection to keep the TCP-variant connection active,” and claim 15 recites
“receiving idle information for detecting an idle time period, during which,
no packet is communicated in a transmission control protocol (TCP)-variant
connection to keep the TCP-variant connection active” (emphases added).
Patent Owner provides the following table setting forth its proposed
constructions for the claim recitations italicized above:
IPR2020-00845
Patent 9,923,995 B1
7
Claim Language Patent Owner’s Proposed
Construction
“idle time period and, during which,
no packet is communicated in the
TCP-variant connection to keep the
TCP-variant connection active”
(Claim 1)
“idle time period and, during which,
no packet of any kind is sent or
received by the apparatus in the
TCP-variant connection to keep the
TCP-variant connection active with
or without a condition and via any
mechanism”
“an idle time period, during which,
no packet is communicated in a
transmission control protocol
(TCP)-variant connection to keep
the TCP-variant connection active”
(Claim 15)
“an idle time period, during which,
no packet of any kind is sent or
received by the apparatus in a
transmission control protocol
(TCP)-variant connection to keep
the TCP-variant connection active
with or without a condition and via
any mechanism”

PO Resp. 38; see also PO Resp. 10–38 (argument concerning these
constructions).
As set forth in the table above, Patent Owner’s proposed constructions
are based on the following three terms or phrases in the recited subject
matter: “no packet,” “communicated,” and “to keep the TCP-variant
connection active.” PO Resp. 10–38. First, as to the language “no packet,”
Patent Owner argues that “[t]here are no qualifiers, adjectives, etc. stated in
connection with the term ‘packet’ and none should be inserted through
erroneous construction or interpretation.” PO Resp. 11–12 (citing Ex. 2020
¶¶ 33–45). We disagree. Although the phrase “no packet” is absolute, the
remainder of the claim does include a significant qualifier by reciting that no
packet is communicated “to keep the TCP-variant connection active.” Thus,
we have no issue with interpreting the words “no packet” in isolation to
mean “no packet of any kind,” but we cannot ignore the phrase “to keep the
IPR2020-00845
Patent 9,923,995 B1
8
TCP-variant connection active,” which is at the center of the parties’ dispute,
as discussed below. Thus, the phrase “no packet of any kind” is still
qualified by other language in the claim.
Second, Patent Owner argues that the term “communicated” means
“sent or received by the apparatus.” PO Resp. 15–20. Patent Owner
proposes this construction “to clarify that the claims, when taking into
consideration the effect of the negative ‘no packet’ limitation, properly
exclude the possible meanings of ‘communicated’ including, but not limited
to ‘only receiving,’ ‘only sending,’ and ‘receiving and sending.’”
PO Resp. 19–20. According to Patent Owner, “under no circumstance[]
should the claims be misinterpreted to require that any packet ‘sent’ by the
apparatus must also be ‘received’ by another node (i.e. not ‘only sen[t]’ per
claim 30).” PO Resp. 20. Petitioner argues that Patent Owner’s proposed
construction is improper but argues that the asserted prior art teaches the
subject matter under Patent Owner’s interpretation in any event. Pet.
Reply 2–7. As explained below, we agree with Petitioner that the asserted
prior art teaches “communicated” under Patent Owner’s proposed
construction. Therefore, we apply Patent Owner’s construction of
“communicated,” and we need not resolve any disputes between the parties
over the scope of this term. See, e.g., Nidec Motor Corp. v. Zhongshan
Broad Ocean Motor Co., 868 F.3d 1013, 1017 (Fed. Cir. 2017) (“[W]e need
only construe terms ‘that are in controversy, and only to the extent necessary
to resolve the controversy’ . . . .” (quoting Vivid Techs., Inc. v. Am. Sci. &
Eng’g, Inc., 200 F.3d 795, 803 (Fed. Cir. 1999))).
Third, Patent Owner argues that the phrase “to keep the TCP-variant
connection active” means “to keep the TCP-variant connection active with
IPR2020-00845
Patent 9,923,995 B1
9
or without a condition and via any mechanism.” PO Resp. 20–38. Patent
Owner explains that this means keeping the connection active
regardless as to what condition, timer, period, counter, or other
mechanism exists to keep the connection active. Put differently,
if any packet is sent or received by the apparatus for the purpose
of keeping the connection active regardless of how it is designed
to keep the connection active, this is outside the scope of the
claimed idle time period.
PO Resp. 21 (citing Ex. 2020 ¶¶ 54–65). Thus, Patent Owner’s construction
asks us to look at the purpose for which a packet is communicated. As
explained below, we find that the asserted prior art teaches, in at least one
scenario, an absence of packets that are communicated for the purpose of
keeping the connection active, and, therefore, we apply Patent Owner’s
construction of the phrase “to keep the TCP-variant connection active” and
need not resolve any disputes regarding the scope of this phrase.
2. Remaining Terms
For purposes of this Decision, we need not construe expressly any
other claim terms. See Nidec, 868 F.3d at 1017.
C. Principles of Law
A patent claim is unpatentable under 35 U.S.C. § 103 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 ordinary skill in the art; and (4) any secondary
IPR2020-00845
Patent 9,923,995 B1
10
considerations, if in evidence.6 Graham v. John Deere Co. of Kan. City, 383
U.S. 1, 17–18 (1966).
D. Alleged Obviousness over Eggert
(Claims 1–24)
Petitioner asserts that claims 1–24 of the ’995 patent are unpatentable
under 35 U.S.C. § 103 as obvious over the teachings of Eggert. Pet. 18–61.
Patent Owner opposes. PO Resp. 41–55. For the reasons explained below,
we determine that Petitioner has proven by a preponderance of the evidence
that claims 1–24 are unpatentable as obvious over Eggert.
1. Eggert
Eggert “specifies a new TCP option - the Abort Timeout Option - that
allows conforming hosts to negotiate per-connection abort timeouts. This
allows mobile hosts to maintain TCP connections across disconnected
periods that are longer than their system’s default abort timeout.”
Ex. 1004, 3. Petitioner argues that Eggert was published on April 14, 2004,
which is more than one year before the earliest priority date of the ’995
patent, and therefore qualifies as prior art under 35 U.S.C. § 102. Pet. 12–14
(citing Ex. 1003 ¶¶ 53–54, 67, 73–80, 114–120–126, 128–138; Ex. 1028;
Ex. 1029; Ex. 1031, 4–10). Patent Owner does not challenge the prior art
status of Eggert. See generally PO Resp. We are persuaded that Petitioner
has shown that Eggert qualifies as prior art under 35 U.S.C. § 102. See
Pet. 12–14.
2. Independent Claim 1
Independent claim 1 is directed to “[a]n apparatus comprising: a
non-transitory memory storing instructions; and one or more processors in

6 Patent Owner does not present any objective evidence of nonobviousness
(i.e., secondary considerations) as to any of the challenged claims.
IPR2020-00845
Patent 9,923,995 B1
11
communication with the non-transitory memory, wherein the one or more
processors execute the instructions for” performing four operations listed in
claim 1. Petitioner argues that Eggert’s reference to Solaris would have
been understood by a person of ordinary skill in the art as describing a
Solaris computer with the Solaris Operating System and a TCP
implementation. Pet. 18 (citing Ex. 1004, 37, § 2.1; Ex. 1003 ¶ 163).
Petitioner further contends that it would have been obvious to a person of
ordinary skill in the art that a hard disk (the recited “non-transitory
memory”) in the Solaris computer stores instructions and that a CPU (the
recited “one or more processors”) would execute instructions that are stored
in the hard disk. Pet. 18–20 (citing Ex. 1004, 3, 5–8 (§ 2.1); Ex. 1003
¶¶ 163–189).
Patent Owner does not dispute Petitioner’s contentions for the claimed
subject matter quoted above, but Patent Owner argues that Eggert does not
teach the “idle time period” subject matter recited in one of the four
operations. See PO Resp. 41–54. We address Patent Owner’s “idle time
period” arguments below.
We are persuaded by Petitioner’s contentions and evidence,
summarized above, and we conclude that Eggert’s disclosure of a Solaris
computer implementation renders obvious “[a]n apparatus comprising: a
non-transitory memory storing instructions; and one or more processors in

7 The Petition cites the page numbers that appear as part of Eggert itself,
which are in the upper right corner of every other page. Pet. 14 n.4
(“Petitioners cite to the page numbers indicated in Eggert.”). We cite the
exhibit page numbers in the lower right-hand corner of Exhibit 1004.
IPR2020-00845
Patent 9,923,995 B1
12
communication with the non-transitory memory, wherein the one or more
processors execute the instructions.”8
a) Receiving a TCP-variant packet
The first operation recited in claim 1 is “receiving, by a second node
from a first node, a transmission control protocol (TCP)-variant packet in
advance of a TCP-variant connection being established” (referred to herein
as “the receiving operation”). Petitioner argues that Eggert discloses that a
responder (“second node”) receives a synchronization (SYN) packet that
contains an abort timeout option (ATO) during a three-way handshake (i.e.,
before a connection is established). Pet. 20 (citing Ex. 1004, 5; Ex. 1003
¶¶ 190–223). Petitioner argues that “a modified TCP SYN segment
(SYN+ATO) that contains an ATO field is a TCP-variant packet, since it
uses the same format as set forth in the TCP standard but adds the ATO
field.” Pet. 24 (citing Ex. 1003 ¶¶ 196, 451) (footnote omitted); see also
Pet. 22 (arguing that Eggert’s disclosure that the initiator can receive a
packet containing the ATO from the responder also teaches “receiving, by a
second node from a first node, a transmission control protocol (TCP)-variant
packet” before a connection is established). Petitioner also argues that
“[c]onnections established by exchanging ATOs are TCP-variant
connections because they are established in accordance with a TCP-variant
protocol by exchanging TCP-variant packets (SYN+ATO,
SYN/ACK+ATO) during the three-way handshake.” Pet. 26 (citing
Ex. 1003 ¶¶ 217–223).9

8 Because we find that Petitioner has demonstrated obviousness of the
subject matter recited in the preambles of the challenged claims, we need not
decide whether the preambles are limiting.
9 ACK refers to acknowledgment. See PO Resp. 12; Ex. 1003 ¶ 58.
IPR2020-00845
Patent 9,923,995 B1
13
During the trial, Patent Owner did not dispute Petitioner’s contentions
for this subject matter. See generally PO Resp. 41–54 (addressing “idle time
period” subject matter); see also Paper 17, 8 (“Patent Owner is cautioned
that any arguments not raised in the response may be deemed waived.”); In
re NuVasive, Inc., 842 F.3d 1376, 1380–81 (Fed. Cir. 2016) (explaining that
the patent owner waived an issue presented in its preliminary response that it
failed to renew in its response during trial); Consolidated Trial Practice
Guide 52 (Nov. 2019) (“Once a trial is instituted, the Board may decline to
consider arguments set forth in a preliminary response unless they are raised
in the patent owner response.”).
We are persuaded by Petitioner’s showing for the receiving operation.
We find that Eggert teaches TCP-variant connections and packets because it
discloses a feature to add to TCP—the abort timeout option. Ex. 1004, 3
(“This document specifies a new TCP option - the Abort Timeout Option -
that allows conforming hosts to negotiate per-connection abort timeouts.”).
Eggert states that “[a] TCP implementation that does not support the TCP
Abort Timeout Option SHOULD silently ignore it.” Ex. 1004, 7, quoted in
Pet. 23. “Thus,” Petitioner correctly asserts, “Eggert discloses adding a new
TCP option to the traditional TCP implementation.” See Pet. 23 (citing
Ex. 1003 ¶¶ 193–195, 214–223).
We also find that Eggert teaches receiving a TCP-variant packet “in
advance of a TCP-variant connection being established,” as recited in
claim 1, because the packets used to negotiate the ATO are sent as part of
the three-way handshake to establish the connection between nodes. For
example, Eggert states the following: “This specification allows both the
initiator of a TCP connection (i.e., the node sending the SYN) as well as the
responder of a TCP connection (i.e., the node receiving the SYN) to initiate
IPR2020-00845
Patent 9,923,995 B1
14
an abort timeout negotiation during the connection’s three-way handshake.”
Ex. 1004, 5, quoted in Pet. 20.
Therefore, based on Petitioner’s persuasive contentions and evidence,
discussed above, we are persuaded that Petitioner has shown that Eggert
teaches “receiving, by a second node from a first node, a transmission
control protocol (TCP)-variant packet in advance of a TCP-variant
connection being established.”
b) Detecting an idle time period parameter and identifying metadata
The second and third operations recited in claim 1 are “detecting an
idle time period parameter field in the TCP-variant packet” and “identifying
metadata in the idle time period parameter field for an idle time period and,
during which, no packet is communicated in the TCP-variant connection to
keep the TCP-variant connection active.” Petitioner argues that Eggert
discloses detecting the abort timeout field and identifying metadata in that
field (the value of the timeout) in the three-way handshake of Figure 2.
Pet. 26–30 (discussing exchange of ATO values in Eggert’s Figure 2).
Figure 2 of Eggert is reproduced below.

Ex. 1004, 7. Figure 2 of Eggert, reproduced above, “illustrates the two
allowed exchanges” in an abort timeout negotiation. Ex. 1004, 5.
IPR2020-00845
Patent 9,923,995 B1
15
Petitioner argues that the abort timeout value contained in the ATO
field that is exchanged during Eggert’s three-way handshake teaches “an idle
time period parameter field,” as recited in claim 1. Pet. 26–29 (citing
Ex. 1004, 3, 5, Fig. 2; Ex. 1003 ¶¶ 224–232). Petitioner argues that
“Eggert’s description of the abort timeout as the duration of time during
which a node can wait to receive an ACK is a period ‘during which no
packet is communicated in the TCP-variant connection to keep the
TCP-variant connection active.’” Pet. 30–31 (citing Ex. 1003 ¶¶ 247–256).
Petitioner argues that, in Eggert, the ACK is the packet that is communicated
to keep the connection active because Eggert discloses that “the TCP
specification includes a ‘user timeout’ that defines the maximum amount of
time that segments may remain unacknowledged before TCP will abort the
connection.” Pet. 30 (quoting Ex. 1004, 3). Thus, according to Petitioner,
when no ACK is received within a certain period of time after sending a
packet, the connection will end. See Pet. 30.
Petitioner also argues that Eggert’s disclosure that the ATO “allows
mobile hosts to maintain TCP connections across disconnected periods that
are longer than their system’s default abort timeout” (Ex. 1004, 3) “parallels
the ’995 Patent, which indicates that the idle time period may be associated
with physical disconnection of a network medium or simply a dead
connection.” Pet. 31 (citing Ex. 1004, 3, 9; Ex. 1001, 2:17–20, 11:53–62,
21:11–18; Ex. 1003 ¶¶ 247–248); see also Ex. 1004, 9 (“Long abort timeout
values allow hosts to tolerate extended periods of disconnection.”).
According to Petitioner, “no packets are communicated to keep the
connection active because they cannot get through the interruption or the
disconnection in the physical medium.” Pet. 31 (citing Ex. 1003 ¶¶ 247–
248; Ex. 1001, 2:17–20, 11:53–62, 21:11–18).
IPR2020-00845
Patent 9,923,995 B1
16
The parties’ dispute centers on whether Eggert teaches a period during
which “no packet is communicated in the TCP-variant connection to keep
the TCP-variant connection active,” which we refer to as the “idle time
period” or “ITP” subject matter. Patent Owner argues that Eggert does not
teach the ITP subject matter because there are packets communicated to
keep the connection active during the period identified by Petitioner as the
idle time period. PO Resp. 41–54. Patent Owner provides the data flow
diagram reproduced below to illustrate this point.

PO Resp. 45; see also Ex. 2020 ¶¶ 80–85 (testimony of Patent Owner’s
declarant, Dr. Michael Smith, regarding this data flow diagram). The
diagram above illustrates a data flow sequence between two nodes in which
data are sent from each of nodes 1 and 2 and an ACK is sent from node 2 to
node 1. PO Resp. 45. With reference to this diagram, Patent Owner argues
the following:
As shown, at step 1., the 1st Node sends Data_1, and starts the
“Abort Timeout” period (or allows an existing abort timeout
period to continue for Data_1). (Ex. 2020 at 78-82.) At step 2.,
the 1st Node receives Data_2 sent from the 2nd Node as
illustrated, such that, at step 3., the 1st Node sends an ACK for
IPR2020-00845
Patent 9,923,995 B1
17
Data_2 during the “Abort Timeout” period of the 1st Node for
Data_1 (i.e. before the expiration thereof). (Ex. 2020 at 78-82.)
The “Abort Timeout”- period of the 1st Node for Data_1 is thus
detected with at least one packet (e.g. the sent ACK at step 3.)
being communicated in the connection to keep the connection
active (e.g. to avoid an “ended” connection). (Ex. 2020
at 78-82.)
PO Resp. 44–45.
Petitioner responds by arguing that Eggert does not require Patent
Owner’s proposed data flow and produces the following data flow diagram
to illustrate its point.

Pet. Reply 16–18. The diagram above is an altered version of Patent
Owner’s proposed data flow (PO Resp. 45) in which Petitioner removes
steps 2 and 3. Pet. Reply 17. In annotations in the diagram above, Petitioner
notes that if the second node does not send data at step 2, then the first node
does not send an acknowledgement at step 3. Pet. Reply 17. Thus,
according to Petitioner, “a timeout will occur if no data packet is
communicated during the idle time period that requires acknowledgement,
IPR2020-00845
Patent 9,923,995 B1
18
or if such a data packet is so communicated, no ACK packet is
communicated during that idle time period.” Pet. Reply 16.
Patent Owner responds that Petitioner is focusing on ACKs as “the
only packets capable of keeping Eggert’s connection active, which is not the
case.” PO Sur-reply 20. In particular, Patent Owner argues that, if node 1
sends data and does not receive an ACK, then node 1 will retransmit the data
on the expiration of a retransmission timer. PO Sur-reply 20–25. According
to Patent Owner, “[l]ike ACKs, the retransmitted-packets are to keep the
connection active by eliciting a received ACK that was missing for an initial
transmission.” PO Sur-reply 25. Patent Owner contends that, “since the
retransmitted-packets elicit a received ACK (that Petitioners’ admit
maintains the connection), the ACK-eliciting retransmitted-packets are also
to keep the connection active. The connection will prematurely close
without them.” PO Sur-reply 25. Thus, during the oral argument, Patent
Owner’s counsel argued that “[t]his really can be resolved by answering a
very simple question. Are the retransmission packets communicated in
Eggert to keep the connection active?” Tr. 27:16–19.
For the reasons explained below, we find that retransmission packets
that may be communicated in Eggert are not packets “to keep the TCP-
variant connection active” and, therefore, that Eggert teaches that “no packet
is communicated in the TCP-variant connection to keep the TCP-variant
connection active,” as recited in claim 1.
There is no dispute that in Eggert, when data are first sent, the ATO
timer starts, and when the ATO timer expires, the connection is terminated.
See PO Resp. 39 (“Eggert’s ATO utilizes the TCP specification in which
there is a ‘user timeout’ that defines the amount of time that packets may
remain unacknowledged before the connection is terminated. (Ex. 1004
IPR2020-00845
Patent 9,923,995 B1
19
at 3.) The failure to receive an ACK packet triggers termination of the
connection.”); see also Ex. 2020 ¶ 30 (stating the same). The retransmission
timer of RFC 793 cited by Patent Owner is a separate timer that governs
retransmission of data if an ACK is not received for the data within a
particular timeout interval. See PO Sur-reply 22 (citing Ex. 1006, 8, 14).
RFC 793 explains that, “[w]hen the TCP transmits a segment containing
data, it puts a copy on a retransmission queue and starts a timer; when the
acknowledgment for that data is received, the segment is deleted from the
queue.” Ex. 1006, 14. However, “[i]f the acknowledgment is not received
before the timer runs out, the segment is retransmitted.” Ex. 1006, 14.
Eggert does not disclose, and Patent Owner does not argue, that
retransmitting data resets the original timeout timer that started upon sending
the data the first time. Thus, despite retransmitting the data, the connection
will terminate on the expiration of the ATO timer that started when the first
data were transmitted.
As discussed above in § II.B.1, Patent Owner argues that the language
“to keep the TCP-variant connection active” describes “a purpose” of the
packet. See PO Resp. 25. If the purpose of retransmitting the data were to
keep the connection active, then the device would be programmed to reset
the ATO timer upon retransmission to avoid termination of the connection.
The fact that this is not done supports a finding that the purpose of these
packets is not to keep the connection active. Indeed, during oral argument,
Patent Owner’s counsel stated that “the programming [of a computer] is
done in a specific way to try to achieve a specific result.” Tr. 46:16–18.
Thus, if the purpose of particular programming were to keep the connection
active, the device would be so programmed by resetting the ATO timer.
Furthermore, Patent Owner’s position is that, “since the retransmitted-
IPR2020-00845
Patent 9,923,995 B1
20
packets elicit a received ACK (that Petitioners’ admit maintains the
connection), the ACK-eliciting retransmitted-packets are also to keep the
connection active. The connection will prematurely close without them.”
PO Sur-reply 25. But a retransmitted packet will only elicit an ACK if it is
received at the other node. Patent Owner’s declarant, Dr. Smith, testifies
that a sent packet may not be received “due to a network condition.”
Ex. 2020 ¶ 106. Indeed, Eggert’s ATO “allows mobile hosts to maintain
TCP connections across disconnected periods that are longer than their
system’s default abort timeout.” Ex. 1004, 3. The possibility that a network
condition such as a disconnection would prevent delivery of an
ACK-eliciting packet is all the more reason that the ATO timer would be
reset upon retransmission if the purpose of the retransmission packet indeed
were to keep the connection active.
For all of these reasons, we find that retransmission packets in Eggert
are not communicated “to keep the TCP-variant connection active.” In view
of this finding, we turn back to the following data flow posited by Petitioner:

IPR2020-00845
Patent 9,923,995 B1
21
Pet. Reply 17. The diagram above is an altered version of Patent Owner’s
proposed data flow (PO Resp. 45) in which Petitioner removes steps 2 and 3.
Pet. Reply 17. In annotations in the diagram above, Petitioner notes that if
the second node does not send data at step 2, then the first node does not
send an acknowledgement at step 3. Pet. Reply 17.
In the above scenario, the first node sends “Data_1” at step one and no
ACK is received. According to Dr. Smith, there can be situations “when a
sent packet is not received (e.g. due to a network condition)” or “when no
ACK is sent for a received data packet (e.g. due to an error condition in the
receiving node).” Ex. 2020 ¶¶ 106, 108, 110, 112, 114, 116, 119, 122, 125.
As discussed above, Eggert expressly contemplates network disconnections.
Ex. 1004, 3. If the second node does not receive “Data_1” because of a
disconnection, then the second node would not attempt to send an ACK as it
would only acknowledge received packets. Even if “Data_1” were received
at the second node, the second node may not send an ACK “due to an error
condition in the receiving node,” as stated by Dr. Smith. Ex. 2020 ¶¶ 106,
108, 110, 112, 114, 116, 119, 122, 125. In these cases, no ACK is sent or
received, consistent with Patent Owner’s proposed construction of
“communicated.” See PO Resp. 15–20. The ATO timer continues to run,
and in the scenario above, neither node sends additional data. The only
other transmission, then, that might occur before the ATO timer expires and
the connection is terminated is a retransmission of “Data_1.”10 But as

10 Depending on the negotiated ATO value, a retransmission might not even
happen because Eggert “does not restrict the range of timeout values used
with the TCP Abort Timeout Option. The 32-bit value in the option can
express abort timeouts from zero seconds to over 136 years.” Ex. 1004, 9.
Eggert then explains that “[v]ery short timeout values can affect TCP
retransmissions over high-delay paths” and that, “[a]lthough the TCP Abort
IPR2020-00845
Patent 9,923,995 B1
22
discussed above, this retransmission is not “to keep the TCP-variant
connection active,” as recited in claim 1. If an ACK is not received by the
first node within the ATO period, the connection will terminate. Therefore,
Eggert teaches, in the above scenario, “an idle time period and, during
which, no packet is communicated in the TCP-variant connection to keep the
TCP-variant connection active.”
Based on the foregoing, we disagree with Patent Owner’s assertion
that “the ITP limitation is not disclosed in any scenario involving any of the
references.” See PO Sur-reply 26. Rather, for the reasons given above, we
are persuaded by Petitioner’s contention that, under at least certain operating
conditions, Eggert teaches the recited ITP subject matter, which is sufficient
to show obviousness. See Unwired Planet, LLC v. Google Inc., 841 F.3d
995, 1002 (Fed. Cir. 2016) (“[C]ombinations of prior art that sometimes
meet the claim elements are sufficient to show obviousness.”); Hewlett–
Packard Co. v. Mustek Sys., Inc., 340 F.3d 1314, 1326 (Fed. Cir. 2003)
(“[A] prior art product that sometimes, but not always, embodies a claimed
method nonetheless teaches that aspect of the invention.”).
We also are persuaded by Petitioner’s contention that Eggert teaches
detecting the abort timeout field and identifying metadata in that field (the
value of the timeout) in the three-way handshake of Figure 2. See Pet. 26–
30 (discussing exchange of ATO values in Eggert’s Figure 2). For example,
in discussing the negotiation of the abort timeout option, Eggert states that

Timeout Option allows negotiation of shorter timeouts, applications
requesting such short timeouts should consider these effects.” Ex. 1004, 9.
Thus, Eggert’s ATO timer may be programmed to be shorter than the
retransmission timer, which Eggert specifically mentions is a factor that
should be considered in selecting short timeout values.
IPR2020-00845
Patent 9,923,995 B1
23
“[t]he timeout value included in the option specifies the proposed abort
timeout for the connection.” Ex. 1004, 5, quoted in Pet. 26. Eggert further
discloses that, “[u]pon receipt of a segment with the Abort Timeout Option,
the receiving host decides whether to accept, shorten, or reject its peer’s
proposed abort timeout.” Ex. 1004, 5, quoted in Pet. 26. Thus, we find that
Eggert teaches the “detecting” and “identifying” operations recited in
claim 1.
For the reasons discussed above and given in the Petition, we are
persuaded that Petitioner has shown, by a preponderance of the evidence,
that Eggert teaches “detecting an idle time period parameter field in the
TCP-variant packet” and “identifying metadata in the idle time period
parameter field for an idle time period and, during which, no packet is
communicated in the TCP-variant connection to keep the TCP-variant
connection active.”
c) Modifying a timeout attribute
The fourth operation recited in claim 1 is “modifying, by the second
node and based on the metadata, a timeout attribute associated with the
TCP-variant connection.” Petitioner argues that Eggert’s disclosure of
shortening the proposed abort timeout value during the three-way handshake
negotiation teaches this subject matter. Pet. 32–35 (citing Ex. 1004, 5–7
(§ 2.1); Ex. 1003 ¶¶ 258–269).
During the trial, Patent Owner did not dispute Petitioner’s contentions
for this subject matter. See generally PO Resp. 41–54 (addressing “idle time
period” subject matter).
We are persuaded by Petitioner’s contentions that Eggert teaches this
“modifying” operation. In particular, Eggert discloses that, “[u]pon receipt
of a segment with the Abort Timeout Option, the receiving host decides
IPR2020-00845
Patent 9,923,995 B1
24
whether to accept, shorten, or reject its peer’s proposed abort timeout.”
Ex. 1004, 5. Thus, we find that Eggert’s disclosure of shortening the
proposed abort timeout teaches “modifying, by the second node and based
on the metadata, a timeout attribute associated with the TCP-variant
connection,” as recited in claim 1.
d) Determination of unpatentability
For the reasons discussed above and based on the full trial record, we
conclude that Petitioner has demonstrated by a preponderance of the
evidence that the subject matter recited in claim 1 would have been obvious
over the teachings of Eggert.
3. Independent Claim 15
Independent claim 15 recites “[a]n apparatus comprising: a
non-transitory memory storing instructions; and one or more processors in
communication with the non-transitory memory, wherein the one or more
processors execute the instructions.” Petitioner refers to its contentions for
the same subject matter recited in claim 1. Pet. 51. For the reasons
explained above in § II.D.2, we are persuaded that Eggert renders this
subject matter obvious.
Claim 15 recites three operations, the first one being “receiving idle
information for detecting an idle time period, during which, no packet is
communicated in a transmission control protocol (TCP)-variant connection
to keep the TCP-variant connection active.” Thus, claim 15 also recites the
ITP subject matter. Petitioner argues that Eggert’s disclosure of the
exchange of ATO information during the three-way handshake teaches this
subject matter, similar to its contentions for the ITP limitation in claim 1,
discussed above in § II.D.2.b. Pet. 51–54. Patent Owner argues claims 1
and 15 together, focusing on the ITP subject matter that is common to both.
IPR2020-00845
Patent 9,923,995 B1
25
PO Resp. 41–48. As discussed above in § II.D.2.b, we are persuaded by
Petitioner’s contentions that the ITP subject matter would have been obvious
over the teachings of Eggert. For the reasons given in the Petition at
pages 51–54 and for those discussed above in § II.D.2.b, we are persuaded
that Eggert’s disclosure renders obvious “receiving idle information for
detecting an idle time period, during which, no packet is communicated in a
transmission control protocol (TCP)-variant connection to keep the TCP-
variant connection active.”
The second operation recited in claim 15 is “generating a TCP-variant
packet including an idle time period parameter field identifying metadata for
the idle time period based on the idle information.” Petitioner argues that
Eggert discloses generating packets having the ATO value, thereby teaching
this subject matter. Pet. 54–56 (citing Ex. 1004, 3, 5, 9, Figs. 1, 2; Ex. 1003
¶¶ 444–445).
During the trial, Patent Owner did not dispute Petitioner’s contentions
for this subject matter. See generally PO Resp. 41–48 (addressing “idle time
period” subject matter of claims 1 and 15).
We are persuaded by Petitioner’s contentions that Eggert teaches this
“generating” operation. For example, Eggert discloses that the ATO value is
included in any segment (packet) that has “a SYN flag, i.e., either the initial
SYN or the SYN-ACK.” Ex. 1004, 5. Eggert’s Figure 2, cited by Petitioner
(Pet. 54–55), illustrates “Allowed TCP Abort Timeout Option (ATO)
Exchanges” and shows the ATO included in SYN and SYN/ACK packets.
Ex. 1004, 7. For the reasons discussed above in § II.D.2.a, we find that
these packets containing the ATO value are TCP-variant packets. See
Pet. 56. Thus, we find that Eggert teaches “generating a TCP-variant packet
IPR2020-00845
Patent 9,923,995 B1
26
including an idle time period parameter field identifying metadata for the
idle time period based on the idle information.”
The third operation recited in claim 15 is
sending, from a first node to a second node, the TCP-variant
packet in advance of the TCP-variant connection being
established to provide the metadata for the idle time period to the
second node, for use by the second node in modifying, based on
the metadata, a timeout attribute associated with the TCP-variant
connection.
Petitioner contends that Eggert’s disclosure of exchanging packets with the
ATO value between nodes and accepting or shortening the abort timeout
attribute based on the exchanged ATO value teaches this subject matter.
Pet. 56–58 (citing Ex. 1004, 5, 7; Ex. 1003 ¶¶ 446–462).
During the trial, Patent Owner did not dispute Petitioner’s contentions
for this subject matter. See generally PO Resp. 41–48 (addressing “idle time
period” subject matter of claims 1 and 15).
We are persuaded by Petitioner’s contentions that Eggert teaches this
“sending” operation. In particular, Eggert discloses that, “[u]pon receipt of a
segment with the Abort Timeout Option, the receiving host decides whether
to accept, shorten, or reject its peer’s proposed abort timeout.” Ex. 1004, 5;
see also § II.D.2.c above (addressing “modifying, by the second node and
based on the metadata, a timeout attribute associated with the TCP-variant
connection,” as recited in claim 1). Thus, we find that Eggert’s disclosure of
exchanging proposed abort timeout values and shortening a proposed abort
timeout teaches the “sending” operation recited in claim 15.
For the reasons discussed above and based on the full trial record, we
conclude that Petitioner has demonstrated by a preponderance of the
IPR2020-00845
Patent 9,923,995 B1
27
evidence that the subject matter recited in claim 15 would have been obvious
over the teachings of Eggert.
4. Dependent Claims 11 and 21
Dependent claims 11 and 21 recite some similar subject matter, but
claim 11 includes an additional recitation as to which Patent Owner
specifically challenges Petitioner’s showing. We discuss the overlapping
subject matter of these claims in subsection a) and the additional recitation
of claim 11 in subsection b) below.
a) Detecting the idle time period and deactivating the connection
Claim 11 depends from claim 1 and recites “wherein the one or more
processors execute the instructions for: detecting the idle time period based
on the timeout attribute; and in response to detecting the idle time period,
deactivating the TCP-variant connection by releasing a resource allocated
for the TCP-variant connection by one of the first and second nodes.”
Claim 21 depends from claim 15 and recites “wherein the one or more
processors execute the instructions for: detecting the idle time period based
on the idle information; and in response to detecting the idle time period,
deactivating the TCP-variant connection.”
Petitioner argues that “Eggert discloses that the host will use the
negotiated abort timeout, rather than the default user timeout, to detect
whether the idle time period, i.e., how long segments remain
unacknowledged, has expired.” Pet. 42 (citing Ex. 1004, 3). Petitioner
contends that a person of ordinary skill in the art “would find it obvious to
implement Eggert with a node that would execute instructions to measure
the amount of time that segments remain unacknowledged and detect when
the time elapsed reaches or exceeds the negotiated abort timeout.” Pet. 43
(citing Ex. 1003 ¶¶ 510–512). Petitioner argues that Eggert teaches
IPR2020-00845
Patent 9,923,995 B1
28
deactivating the connection by releasing a resource with its disclosure that
“[i]f a disconnection lasts longer than the user timeout, the TCP connection
will abort.” Pet. 43 (quoting Ex. 1004, 3).
Patent Owner does not dispute Petitioner’s contentions for this subject
matter. See PO Resp. 48–54 (addressing different recitation of claim 11).
We are persuaded by Petitioner’s contentions that it would have been
obvious for a node to detect the time period defined by the negotiated ATO
because Eggert specifies that the “connection will abort” when the
“disconnection lasts longer than the user timeout.” Ex. 1004, 3. Based on
Petitioner’s persuasive contentions and evidence, therefore, we are
persuaded that the above-quoted subject matter of claims 11 and 21 would
have been obvious to a person of ordinary skill in the art based on Eggert’s
disclosure.
b) Deactivating without signaling
Claim 11 further specifies that the deactivating occurs “without
signaling another one of the first and second nodes that is related to the
detection of the idle time period.” Petitioner argues that “[t]he ATO of
Eggert does not specify any mechanism to signal another node that an idle
time period has been detected.” Pet. 43. Petitioner further argues that a
person of ordinary skill in the art would not implement Eggert to signal the
other node because “the Eggert system assumes a disconnection occurred so
the signal may not reach the other node.” Pet. 44 (citing Ex. 1003 ¶¶ 375–
377; Ex. 1006, 36–38, 4011).

11 The Petition cites the page numbers that appear as part of RFC 793 itself,
which appear in brackets at the bottom of every page. We cite the exhibit
page numbers in the lower right-hand corner of Exhibit 1006.
IPR2020-00845
Patent 9,923,995 B1
29
Patent Owner responds that Eggert’s disclosure that the “connection
will abort” (Ex. 1004, 3) means that a signal is sent to the other side because
RFC 793 defines the “Abort” command as including the sending of a
RESET message. PO Resp. 48–50 (citing Ex. 1006, 54; Ex. 2020 ¶¶ 88–
104). Patent Owner also cites disclosures in RFC 793 about the sending of
RESET messages in certain circumstances. PO Resp. 51–52 (citing
Ex. 1006, 37–39, Fig. 10; Ex. 2020 ¶¶ 90–104).
For the reasons explained below, we are persuaded by Petitioner’s
contentions that the subject matter reciting “without signaling” would have
been obvious in view of Eggert. Eggert discloses the following: “The TCP
specification includes a ‘user timeout’ that defines the maximum amount of
time that segments may remain unacknowledged before TCP will abort the
connection. If a disconnection lasts longer than the user timeout, the TCP
connection will abort.” Ex. 1004, 3 (citing RFC 793 (Ex. 1006 in the
record)); see also Ex. 2020 ¶ 30 (“Eggert’s ATO utilizes the TCP
specification in which there is a ‘user timeout’ that defines the amount of
time that packets may remain unacknowledged before the connection is
terminated. (Ex. 1004 at 3.) The failure to receive an ACK packet triggers
termination of the connection. (Id.)”). Although Eggert uses the term
“abort,” the TCP Specification (RFC 793) provides the following description
of a “USER TIMEOUT”: “For any state if the user timeout expires, flush all
queues, signal the user ‘error: connection aborted due to user timeout’ in
general and for any outstanding calls, delete the TCB, enter the CLOSED
state and return.” Ex. 1006, 81, quoted in Pet. 44. RFC 793 further explains
that “CLOSED is fictional because it represents the state when there is no
[Transmission Control Block (TCB)], and therefore, no connection.”
Ex. 1006, 25; see also Ex. 1006, 26 (“CLOSED - represents no connection
IPR2020-00845
Patent 9,923,995 B1
30
state at all.”). Based on these disclosures, we agree with Petitioner that “the
RESET message is not sent when the user timeout expires.” See Pet. 44
(citing Ex. 1006, 77).
Although “a special RESET message” is sent as part of the ABORT
command, this is a TCP user command or user call, as Petitioner correctly
points out. See Pet. Reply 24–25 (citing Ex. 1006, 48, 54, 56). RFC 793
states that “[t]he activity of the TCP can be characterized as responding to
events,” which “can be cast into three categories: user calls, arriving
segments, and timeouts.” Ex. 1006, 56, cited in Pet. Reply 25. RFC 793
categorizes “ABORT” as a “User Call[]” event and “USER TIMEOUT” as a
“Timeout[]” event. Ex. 1006, 56, cited in Pet. Reply 25. Thus, we find that
Eggert’s disclosure of the “user timeout” in the “TCP Specification” (RFC
793) refers to the timeout event in RFC 793, in which no reset message is
sent. Ex. 1004, 81 (“For any state if the user timeout expires, flush all
queues, signal the user ‘error: connection aborted due to user timeout’ in
general and for any outstanding calls, delete the TCB, enter the CLOSED
state and return.”).
Based on the foregoing, we find that Eggert teaches deactivating the
connection “without signaling another one of the first and second nodes that
is related to the detection of the idle time period,” as recited in claim 11.
c) Determination of unpatentability
For the reasons discussed above and based on the full trial record, we
conclude that Petitioner has demonstrated by a preponderance of the
evidence that the subject matter recited in claims 11 and 21 would have been
obvious over the teachings of Eggert.
IPR2020-00845
Patent 9,923,995 B1
31
5. Dependent Claims 2–10, 12–14, 16–20, and 22–24
Petitioner contends that dependent claims 2–10, 12–14, 16–20, and
22–24 are unpatentable as obvious over the teachings of Eggert. Pet. 35–42,
45–51, 58–61. For these dependent claims, Patent Owner refers to its
arguments regarding the ITP subject matter of the independent claims and
does not set forth additional arguments for the particular subject matter
recited in these claims. See PO Resp. 55. As explained below, we are
persuaded by Petitioner’s contentions for dependent claims 2–10, 12–14,
16–20, and 22–24.
Claim 2 recites, “The apparatus of claim 1 wherein the apparatus is
configured such that the timeout attribute is an attribute of a keep-alive.”
Citing Eggert’s disclosure that the ATO “allows mobile hosts to maintain
TCP connections across disconnected periods” (Ex. 1004, 3), Petitioner
argues that “the initiator and responder in Eggert keep the TCP connection
alive based on information contained in the ATO.” Pet. 35–36 (citing
Ex. 1004, Abstract, 3; Ex. 1003 ¶¶ 271–277).
Claim 3 recites, “The apparatus of claim 1 wherein at least one of: the
second node includes a server . . . or the second node includes a client,” and
each alternative recites an operation for which the server or client is
configured. For example, the first alternative recites “the server being
configured to: in response to the receiving, send, by the second node to the
first node, another TCP-variant packet in advance of the TCP-variant
connection being established, the another TCP-variant packet including
other metadata for the idle time period.” Petitioner argues that Eggert’s
disclosure of the responder sending a SYN/ACK+ATO segment teaches this
subject matter. Pet. 36–37 (citing Ex. 1004, 5–7, 11–13, Fig. 2; Ex. 1003
¶¶ 279–291). Petitioner also argues that Eggert teaches the second
IPR2020-00845
Patent 9,923,995 B1
32
alternative recited in claim 3. Pet. 37–38 (citing Ex. 1004, 1–5, Fig. 2;
Ex. 1003 ¶¶ 192, 212, 221, 292–298).
Claim 4 depends from claim 3 and recites “wherein, regardless as to
whether the apparatus is the server or the client, the metadata is the same as
the other metadata.” Claim 5 depends from claim 3 and recites “wherein,
regardless as to whether the apparatus is the server or the client, the
metadata is different from the other metadata.” Petitioner argues that Eggert
teaches the subject matter in both of these claims because the receiving host
can either accept the ATO value it receives (“same” as recited in claim 4) or
shorten it (“different” as recited in claim 5). Pet. 38–39 (citing Ex. 1004, 5–
11; Ex. 1003 ¶¶ 299–316).
Claim 6 depends from claim 1 and recites that “the timeout attribute is
specified in at least one of a number of seconds or minutes,” and claim 19
depends from claim 15 and recites that “the timeout attribute is specified in a
number of seconds.” Petitioner cites Eggert’s disclosure that “the Abort
Timeout Option specifies abort timeouts as 32-bit values with a granularity
of seconds.” Pet. 39 (quoting Ex. 1004, 11) (citing Ex. 1003 ¶¶ 317–322).
Claim 7 depends from claim 1 and recites that “the apparatus is
configured such that the timeout attribute is used to keep the TCP-variant
connection open when inactive, and to prevent one or more other nodes from
closing the TCP-variant connection when inactive.” Citing Eggert’s
disclosure that the ATO allows nodes “to maintain TCP connections across
disconnected periods that are longer than their system’s default abort
timeout” (Ex. 1004, 3), Petitioner argues that Eggert’s ATO keeps the
connection open when inactive and prevents closure of the connection,
thereby teaching the subject matter of claim 7. Pet. 39–40 (citing Ex. 1004,
Abstract, 3, 9; Ex. 1003 ¶¶ 327–332).
IPR2020-00845
Patent 9,923,995 B1
33
Claim 8 depends from claim 1 and recites “wherein the apparatus is
configured such that the metadata is used as input of an algorithm for
determining a duration of time specified by the timeout attribute.” Petitioner
argues that “[i]n Eggert, the receiving host performs an algorithm to decide
whether to shorten, accept, or reject the proposed ATO,” and Petitioner cites
several disclosures in Eggert as examples of algorithms for determining the
timeout duration. Pet. 40 (citing Ex. 1004, 5, 9–13; Ex. 1003 ¶¶ 333–344).
For example, Eggert discloses that “implementations can require prior peer
authentication . . . before accepting long abort timeouts for the peer’s
connections.” Ex. 1004, 13.
Claim 9 depends from claim 8 and recites that “the apparatus is
configured such that the algorithm is determined based on at least one
particular attribute.” Petitioner argues that “the algorithm may be
determined based on attributes, including a security feature or a confirmation
of prior peer authentication.” Pet. 41 (citing Ex. 1004, 13). As discussed in
the preceding paragraph for claim 8, Eggert discloses an algorithm involving
“prior peer authentication.” Ex. 1004, 13.
Claim 10 depends from claim 1 and recites that “the apparatus is
configured such that the modification of the timeout attribute results from a
negotiation between the first node and the second node via a negotiation
protocol of a TCP-variant protocol.” Claim 20 depends from claim 15 and
recites that “the apparatus is configured such that the modification of the
timeout attribute results from a negotiation between the first node and the
second node.” Petitioner argues that this subject matter is taught by Eggert
because the “abort timeout value is negotiated during the connection’s three-
way handshake.” Pet. 41–42 (citing Ex. 1004, 5, 9, Fig. 2; Ex. 1003 ¶¶ 355–
362, 500–504).
IPR2020-00845
Patent 9,923,995 B1
34
Claim 12 depends from claim 1 and recites “wherein the apparatus is
configured such that at least one of the detecting or the identifying is
performed at a TCP-variant layer other than a TCP layer, where the TCP-
variant layer is above an Internet Protocol (IP) layer and below a hypertext
transfer protocol (HTTP) application layer.” Claim 22 depends from
claim 15 and recites “wherein the apparatus is configured such that at least
the generating is performed at a TCP-variant layer other than a TCP layer,
where the TCP-variant layer is above an Internet Protocol (IP) layer and
below a hypertext transfer protocol (HTTP) application layer.” Citing the
testimony of its declarant, Mr. Scott Bradner, Petitioner argues that “HTTP
applications are application layer protocols that run above transport layer
protocols (e.g., TCP), which run above the network layer (e.g., IP layer).”
Pet. 46 (citing Ex. 1003 ¶¶ 390–392, 522–523). According to Petitioner,
therefore, Eggert’s TCP-variant protocol would be implemented in a TCP-
variant layer, which a person of ordinary skill in the art would have
implemented “between the application layer (e.g., HTTP) and the IP layer,
just like the original TCP layer.” Pet. 45–46 (citing Ex. 1004, 5; Ex. 1003
¶¶ 381–392, 516–523).
Claim 13 depends from claim 1 and recites
wherein the one or more processors execute a network
application that is configured to perform a 3-way TCP handshake
for establishing a TCP connection that is different than the TCP-
variant connection, and wherein the network application is
configured to establish the TCP-variant connection instead of the
TCP connection in order to permit negotiation, between the first
node and the second node, of the timeout attribute, where the
timeout attribute is not available when establishing the TCP
connection, but is available when establishing the TCP-variant
connection so that the TCP-variant connection is capable of
IPR2020-00845
Patent 9,923,995 B1
35
being at least partially closed when inactive based on the timeout
attribute.
Claim 23 depends from claim 15 and recites similar subject matter.
With regard to these claims, Petitioner argues that Eggert’s ATO is
optional and that a person of ordinary skill in the art “would have found it
obvious to implement Eggert with a host, for example (e.g., a server), that
supports both the standard TCP and Eggert’s TCP modification.” Pet. 46–
47 (citing Ex. 1004, 7–13; Ex. 1003 ¶¶ 396–403, 527–528). For example,
Eggert discloses that “[a] TCP implementation that does not support the TCP
Abort Timeout Option SHOULD silently ignore it,” which “causes the
connection to use the default abort timeout, thus ensuring interoperability.”
Ex. 1004, 7. Petitioner further argues that “a host that wishes to negotiate
the timeout must resort to exchange Eggert’s ATO parameters during the
three-way handshake,” thereby teaching negotiation of the timeout attribute.
Pet. 48 (citing Ex. 1004, 3–5; Ex. 1003 ¶¶ 404–409). Petitioner argues that
“[t]he negotiated abort timeout attribute of Eggert is not available when
establishing the TCP connection but is available when establishing the
Eggert TCP-variant connection.” Pet. 49 (citing Ex. 1004, 9–13). Finally,
Petitioner argues that “Eggert’s TCP-variant connection is fully closed,
meeting the limitation of being ‘at least partially closed,’ when the host
determines that the connection is inactive based on the abort timeout.”
Pet. 50 (citing Ex. 1003 ¶¶ 417–418, 539).
Claim 14 depends from claim 1 and recites “wherein the apparatus is
configured such that the TCP-variant packet and the metadata included
therewith are received by the second node from the first node, without any
prior signaling from the second node to the first node.” Claim 17 depends
from claim 15 and recites “wherein the apparatus is configured such that the
IPR2020-00845
Patent 9,923,995 B1
36
TCP-variant packet and the metadata included therewith are sent by the first
node to the second node, without any prior signaling received by the first
node from the second node.” Petitioner argues that Eggert teaches this
subject matter because “[t]he messages exchanged in the Eggert three-way
handshake are the first messages received by any node to establish the
Eggert TCP-variant connection.” Pet. 51 (citing Ex. 1004, Fig. 2; Ex. 1003
¶¶ 423–425); see Pet. 59 (addressing claim 17). In support, Petitioner
produces annotated excerpts from Eggert’s Figure 2 (not reproduced here)
noting that “SYN+ATO” is the first signal with no prior signaling. Pet. 51,
59.
Claim 16 depends from claim 15 and recites “wherein the apparatus is
configured such that modifying the timeout attribute reduces a number of
keep-alive signals that are required to be communicated.” Petitioner cites
keep-alive messages in RFC 1122, which Eggert cites as a normative
reference (Ex. 1004, 15), and argues that “[o]ne obvious implementation of
Eggert is employing keep-alive packets per RFC 1122.” Pet. 58 (citing
Ex. 1004, 3, 15; Ex. 1003 ¶¶ 151, 476; Ex. 1007). According to Petitioner,
“[s]ince the keep-alive packets are sent at a regular interval, shortening the
abort timeout will reduce the number of keep-alive signals that are
transmitted when there is a disconnection.” Pet. 58–59.
Claim 18 depends from claim 15 and recites “wherein the metadata
includes a first value and the timeout attribute is capable of being modified,
by the second node, to include a second value that is different than the first
value of the metadata.” Petitioner argues that, in Eggert, the metadata
includes the ATO value and that this value can be shortened, i.e., “modified”
during the three-way handshake negotiation. Pet. 59–60 (citing Ex. 1004, 5,
11, Figs. 1, 2; Ex. 1003 ¶¶ 490–494). Eggert discloses that, “[u]pon receipt
IPR2020-00845
Patent 9,923,995 B1
37
of a segment with the Abort Timeout Option, the receiving host decides
whether to accept, shorten, or reject its peer’s proposed abort timeout.”
Ex. 1004, 5 (emphasis added).
Claim 24 depends from claim 15 and recites “wherein the apparatus is
configured such that the TCP-variant packet and the metadata included
therewith are sent by the first node to the second node, in response to
receiving, by the first node from the second node, another TCP-variant
packet with other metadata.” Petitioner argues that Eggert’s
“SYN/ACK+ATO segment from the responder (first node) is a TCP-variant
packet that contains in the ATO sub-field metadata for an abort timeout to be
used for the connection and is a response to a first SYN+ATO segment
received from the initiator (second node).” Pet. 61 (citing Ex. 1003 ¶¶ 541–
546).
As noted at the beginning of this section, Patent Owner does not
separately address the additional subject matter recited in these dependent
claims. See generally PO Resp.
Having considered the full record developed during trial, we are
persuaded by Petitioner’s contentions and evidence, summarized above, and
we determine that Petitioner has proven by a preponderance of the evidence
that dependent claims 2–10, 12–14, 16–20, and 22–24 are unpatentable as
obvious over the teachings of Eggert.
E. Remaining Grounds
As noted above in § I.A, Petitioner presents separate challenges to the
claims in this proceeding. Because we determine that all challenged claims
are unpatentable as discussed above, we need not separately assess the
patentability of these claims based on the remaining grounds in this
proceeding. 35 U.S.C. § 318(a) (“If an inter partes review is instituted and
IPR2020-00845
Patent 9,923,995 B1
38
not dismissed under this chapter, the Patent Trial and Appeal Board shall
issue a final written decision with respect to the patentability of any patent
claim challenged by the petitioner and any new claim added under section
316(d).”); Bos. Sci. Scimed, Inc. v. Cook Grp. Inc., 809 F. App’x 984, 990
(Fed. Cir. 2020) (nonprecedential) (“We agree that the Board need not
address issues that are not necessary to the resolution of the proceeding.”).

III. CONCLUSION12
For the reasons discussed above, we determine that Petitioner has
proven, by a preponderance of the evidence, that the challenged claims are
unpatentable, as summarized in the following table:

12 Should Patent Owner wish to pursue amendment of the challenged claims
in a reissue or reexamination proceeding subsequent to the issuance of this
decision, we draw Patent Owner’s attention to the April 2019 Notice
Regarding Options for Amendments by Patent Owner Through Reissue or
Reexamination During a Pending AIA Trial Proceeding. See 84 Fed. Reg.
16,654 (Apr. 22, 2019). If Patent Owner chooses to file a reissue application
or a request for reexamination of the challenged patent, we remind Patent
Owner of its continuing obligation to notify the Board of any such related
matters in updated mandatory notices. See 37 C.F.R. § 42.8(a)(3), (b)(2).
13 As explained above, because we determine that the challenged claims are
unpatentable on another ground, we decline to address the remaining
grounds.
Claim(s) 35 U.S.C. § Reference(s)/Basis
Claims
Shown
Unpatentable
Claims
Not Shown
Unpatentable
1–24 103 Eggert 1–24
1–24 103 Eggert,
Hankinson13

16 103 Eggert, Hankinson,
RFC 1122

Overall
Outcome
1–24
IPR2020-00845
Patent 9,923,995 B1
39

IV. ORDER
Accordingly, it is
ORDERED that claims 1–24 of the ’995 patent have been shown to be
unpatentable; and
FURTHER ORDERED that, because this is a Final Written Decision,
parties to the proceeding seeking judicial review of the Decision must
comply with the notice and service requirements of 37 C.F.R. § 90.2.

IPR2020-00845
Patent 9,923,995 B1
40
FOR PETITIONER:
Naveen Modi
Joseph E. Palys
Quadeer A. Ahmed
Jason Heidemann
Ph-google-jenam-IPR@paulhastings.com
naveenmodi@paulhastings.com
josephpalys@paulhastings.com
quadeerahmed@paulhastings.com
jasonheidemann@paulhastings.com

FOR PATENT OWNER:
Timothy Devlin
Derek Dahlgren
DEVLIN LAW FIRM LLC
td-ptab@devlinlawfirm.com
ddahlgren@devlinlawfirm.com