Ex Parte Kennedy

Board of Patent Appeals and Interferences

Jun 22, 2012

10989067 (B.P.A.I. Jun. 22, 2012)

UNITED STATES PATENT AND TRADEMARK OFFICE
________________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
________________

Ex parte ROBERT A. KENNEDY
________________

Appeal 2010-002365
Application 10/989,067
Technology Center 2400
________________

Before SCOTT R. BOALICK, JOHN A. JEFFERY,
and STANLEY M. WEINBERG, Administrative Patent Judges.

WEINBERG, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2010-002365
Application 10/989,067

2
SUMMARY
Appellant appeals under 35 U.S.C. §§ 6(b) and 134 from the
Examiner’s rejection of claims 1-3, 5-14, 16-24, and 26-31. We have
jurisdiction under 35 U.S.C. § 6(b).
We affirm.

STATEMENT OF CASE
Appellant’s invention relates generally to the field of communication
networks and more particularly to mobile ad hoc wireless networks
(MANETs) and related methods. A MANET includes a number of
geographically-distributed, potentially mobile nodes that are wirelessly
connected by one or more radio frequency channels. The most distinctive
feature of MANETs is the lack of any fixed infrastructure. The network is
formed only of mobile nodes, and is created on the fly as the nodes transmit
to or receive from other nodes. Such a network would be useful, for
example, in a hostile environment where a fixed communication
infrastructure is unreliable or unavailable, such as in a battle field or in a
natural disaster area struck by earthquake or hurricane. See generally Spec.
¶¶ 0001-03, 0005.
More particularly, the present invention provides predictive mobile ad
hoc networking in which the network predicts and provides a response for
future needed network services, resources, and configurations. Spec.
¶ 0008.
Claim 1 is representative with key disputed limitations emphasized:
1. A method for operating a mobile ad hoc network (MANET)
comprising a plurality of mobile nodes and a plurality of wireless
Appeal 2010-002365
Application 10/989,067

3
communication links connecting the mobile nodes together, the method
comprising:
predicting future-needed network services, network
resources and network configurations in the MANET using fuzzy
logic and a knowledge base of fuzzy if-then rules;
predicting a network response, based upon the predicted
future-needed network services, network resources and network
configurations; and
adjusting the MANET based upon the predicted network
response before the future-needed network services, network
resources and network configurations are actually needed in the
MANET.
The Examiner relies upon the following as evidence of
unpatentability:
Chung-Ju US 6,067,287 May 23, 2000
Haas US 2003/0005149 A1 Jan. 2, 2003
Bellur US 2003/0120809 A1 June 26, 2003
McKinnin, III US 6,917,628 B2 July 12, 2005

Tonjes US 2006/0268873 A1 Nov. 30, 2006

William Su et al., Mobility Prediction in Wireless Networks
MILCOM 2000. 21st CENTURY MILITARY COMMUNICATIONS
CONFERENCE PROCEEDINGS 491(2000).

Appeal 2010-002365
Application 10/989,067

4
THE REJECTIONS
Claims 1, 2, 5, 6, 9, and 10 stand rejected under 35 U.S.C. § 103(a) as
unpatentable over Su in view of Chung-Ju. Ans. 3-8.1
Claim 3 stands rejected under 35 U.S.C. § 103(a) as unpatentable over
Su in view of Chung-Ju and further in view of Tonjes. Ans. 8-9.
Claims 7 and 8 stand rejected under 35 U.S.C. § 103(a) as
unpatentable over Su in view of Chung-Ju and further in view of McKinnin,
III. Ans. 9.
Claim 11 stands rejected under 35 U.S.C. § 103(a) as unpatentable
over Su in view of Chung-Ju and further in view of Bellur. Ans. 10.
Claims 12, 13, 18, 19, 21-23, 28, 29, and 31 stand rejected as
unpatentable over Su in view of Chung-Ju and further in view of Haas. Ans.
10-18.
Claims 14 and 24 stand rejected under 35 U.S.C. § 103(a) as
unpatentable over Su in view of Chung-Ju and further in view of Haas and
Tonjes. Ans. 18-20.
Claims 16, 17, 26, and 27 stand rejected under 35 U.S.C. § 103(a) as
unpatentable over Su in view of Chung-Ju and further in view of Haas and
McKinnin. Ans. 20-21.
Claims 20 and 30 stand rejected under 35 U.S.C. § 103(a) as
unpatentable over Su in view of Chung-Ju, Haas and further in view of
Bellur. Ans. 22-23.

1 Throughout this opinion, we refer to (1) the Final Rejection mailed August
7, 2008; (2) the Appeal Brief filed April 30, 2009; (3) the Examiner’s
Answer mailed July 27, 2009; and (4) the Reply Brief filed September 23,
2009, which incorrectly states that it was responding to an Examiner’s
Answer mailed on April 6, 2009. Reply Br. 1.
Appeal 2010-002365
Application 10/989,067

5
The Examiner finds that Su discloses every feature of claim 1 except
for using fuzzy logic and a knowledge of fuzzy if-then rules, which the
Examiner finds are taught by Chung-Ju. Ans. 4-5. The Examiner concludes
that it would have been obvious to combine the invention of Chung-Ju with
the teachings of Su because Chung-Ju provides a method where fuzzy logic
can be implemented in decision making and the implementation of Chung-
Ju’s teachings into Su’s method would provide more robust and accurate
network predictions and network adjustments as mobile nodes move into,
away, and within an ad hoc network. Final Rejection 3-4; Ans. 4-5.
Appellant first contends that the combination rejection is erroneous
because the Examiner relies upon Appellant’s application as a hindsight
roadmap to modify the prior art, and because the fuzzy logic admission
control scheme of Chung-Ju “has nothing to do with the prediction of
anything let alone prediction of future-needed network services, network
resources and/or network configurations” (hereinafter “the three claimed
prediction requirements”). App. Br. 8, 1st and 3rd full paragraphs (emphasis
omitted).
Appellant also contends that even if the combination were somehow
proper, the combination could not result in the claimed invention because (1)
the combination of Su and Chung-Ju would result in a fuzzy logic admission
control scheme being used in Su’s network; and (2) neither reference
discloses the three claimed prediction requirements using fuzzy logic and a
knowledge base of fuzzy if-then rules. App. Br. 8, 2nd full paragraph; Reply
Br. 4-5.
Appellant also contends that the Examiner impermissibly uses a single
portion of Su to show that Su teaches both of the separately claimed
Appeal 2010-002365
Application 10/989,067

6
predictions of network services and network resources. Reply Br. 3,
comparing Ans. 24, ll. 4-12 with Ans. 24, l.14-Ans. 25 l. 3.

ISSUE
Under § 103(a), has the Examiner erred by (1) finding that Su teaches
all of the limitations of claim 1, except for using fuzzy logic and a
knowledge base of fuzzy if-then rules and (2) concluding that it would have
been obvious in view of Chung-Ju to use fuzzy logic and a knowledge of
fuzzy if-then rules in the Su method?

FINDINGS OF FACT (FF)
Appellant’s Invention
1. Appellant’s invention provides a method and related
structure that predicts future-needed network services, network resources
and network configurations (the three claimed prediction requirements) in a
MANET and predicts a network response based upon the three claimed
prediction requirements. The MANET is adjusted based upon the predicted
appropriate network response before the three claimed prediction needs are
actually needed in the MANET. Spec. ¶¶ 0008-09.
2. Network services can be either routing or mobility or
both. Spec. ¶ 0014.
3. Network resources include at least one of power and
frequency. Spec. ¶ 0014.

Appeal 2010-002365
Application 10/989,067

7
Su
4. Su teaches mobile nodes (“nodes exhibit some degree of
regularity in mobility patterns,” p.491, Section I, Introduction, left column,
2nd para.) with wireless links between nodes (“mobile ad hoc networks do
not have any fixed wired communication infrastructure,” p.491, Section I,
Introduction, left column, 1st para.) (endnote omitted). See also Fig. 1.
5. Su “can predict the future state of the network topology
and provide a transparent network access during the period of topology
changes.” Page 491, Introduction, left column, 2nd para.).
6. Su uses “mobility prediction to enhance unicast and
multicast routing protocols.” Page 491, Introduction, right column, first full
paragraph.
7. “GPS position information is piggybacked on data
packets during a live connection and is used to estimate the expiration time
of the link between two adjacent nodes. Based on this prediction, routes are
reconfigured before they disconnect.” Page 491, Introduction, right column,
1st full para.
8. Section II of Su presents “the method used to predict the
link expiration time (LET) and various ways to enhance unicast and multi-
cast routing protocols by using mobility prediction.” Page 491; right
column, 2nd full para.
9. “The destination predicts the change in topology ahead of
time and determines when a flow needs to be rerouted or ‘handoffed.’” Page
492, left column, Section B.1, 1st para.

Appeal 2010-002365
Application 10/989,067

8
Chung-Ju
10. Chung-Ju teaches a fuzzy logic controller and a fuzzy
rule base that is a control knowledge-base characterized by a set of linguistic
statements in the form of “if-then” rules. Col. 5, ll. 35-41.
11. Chung-Ju’s inference engine has decision-making logic
that acquires input linguistic terms and uses an inference method to obtain
output linguistic terms. Col. 5, ll. 42-46.
12. A Neural Fuzzy Connection Admission Controller
(NFCAC) has a bi-directional communication with a Network Resource
Estimator that receives data from a Bandwidth Estimator which estimates an
equivalent capacity required for new connection from traffic parameters
before a new connection is accepted. That is, the Network Resource
Estimator is updated when a new connection is accepted or when an existing
connection is disconnected by the NFCAC as indicated by the bi-directional
arrow in Fig. 2 between the Network Resource Estimator and the NFCAC.
Col. 6, ll. 39-50. Accord Reply Br. 4(quoting the aforesaid citation).

THE REJECTION OF CLAIM 1 AS
OBVIOUS OVER SU IN VIEW OF CHUNG-JU
Relying upon Appellant’s definitions in the application, the Examiner
finds that “network services” can be either routing or mobility or both (FF 2)
and that Su teaches mobility prediction (FF 6, 8) and rerouting or handoff
(FF 9). Accordingly, the Examiner finds that Su teaches predicting network
Appeal 2010-002365
Application 10/989,067

9
services. Ans. 24, 2nd full paragraph.2 Appellant agrees that Su is concerned
with node mobility prediction. App. Br. 6-7.
Finding that Appellant defines “network resources” as frequency (FF
3; Ans. 24), the Examiner also finds that Su estimates expiration time of a
link between two adjacent nodes (FF 7) because a destination in Su predicts
a change in topology ahead of time and determines when a flow needs to
rerouted or handed off (FF 9). The Examiner further finds that a handoff
involves a data flow to start occupying a link that it did not previously
occupy, and that this implies a data flow is starting to occupy a frequency
bandwidth of the link it did not previously occupy. Therefore, the Examiner
finds that the estimation of the expiration time of a link between two
adjacent nodes for a rerouting or a handoff in Su is substantively the same as
the recited predicting future-needed network resources. Ans. 24-25.
Appellant does not persuasively counter these findings. Reply B. 2-3.
Regarding predicting the claimed future needed network
configurations, the Examiner finds that Su teaches predicting a future state
of a network topology and providing transparent network access during a
period of topology changes (FF 5). Accordingly, the Examiner concludes
that predicting a future statement of network topology is substantively the
same as predicting a future needed network configuration since a network
topology is a type of network configuration of network nodes. Ans. 25.

2 The second full paragraph on page 24 of the Answer is concerned with the
issue of “network services.” We therefore presume that the last two lines of
the second full paragraph contain a typographical error when it refers to
“network resources” rather than “network services.” Our presumption is
buttressed by the Examiner’s discussion of “network resources” in the third
paragraph of page 24 of the Answer.
Appeal 2010-002365
Application 10/989,067

10
Appellant agrees that Su predicts the future state of network topology. App.
Br. 7, ll. 1-2.
Although Appellant generally discusses Su’s teachings (Reply Br. 2,
3rd full para.), he does not persuasively contest the Examiner’s correlation of
various parts of Su to the three claimed prediction requirements. Instead, he
contends that the Examiner is impermissibly using the same portion of Su
(page 492, section B.1) to show that Su teaches both of the separately
claimed prediction of network services and the prediction of network
resources. Reply Br. 3, comparing Ans. 24, ll. 4-12 with Ans. 24, l. 14-Ans.
25 l. 3. The cited passage in Su is: “The destination predicts the change in
topology ahead of time and determines that a flow needs to be rerouted or
‘handoffed.’” Reply Br. 3. We agree that the Examiner refers to the cited
passage twice. Nevertheless, for the following reasons, the Examiner’s dual
references to the cited passage of Su is not dispositive.
The Examiner refers to Su’s p.491 to show mobility prediction and,
therefore, Appellant’s prediction of network services. (FF 2; Ans. 24).
The Examiner then refers to the cited passage on p.492 to show
frequency and, therefore, Appellant’s prediction of network resources. (FF
3; Ans. 25).
The Examiner’s second reference at Ans. 24 to the cited passage to
show prediction of network services is unnecessary and therefore immaterial
because, as shown above, the Examiner has additionally relied upon a
different part of Su (page 491) to show Su’s prediction of network services.
We therefore conclude that the Examiner has not erred because the
Examiner relied on two different portions of Su to map prediction of
network services and prediction of network resources.
Appeal 2010-002365
Application 10/989,067

11
Regarding Chung-Ju, the Examiner finds that its estimating
capabilities (FF 12) suggest that a NFCAC sends information to a Network
Resource Estimator that is used for estimating network resource usage
involving a new connection. Consequently, the Examiner concludes, the
reference teaches predicting (estimating) network resources using fuzzy
logic and a knowledge base of fuzzy if-then rules. Ans. 5, 26. See also, FF
10-11.
Appellant responds that Chung-Ju describes an admission controller
and does not teach or suggest the prediction of the three claimed prediction
requirements using a fuzzy logic and a knowledge base of fuzzy if-then
rules. Reply Br. 4-5. We divide Appellant’s contention into two parts: (1)
Chung-Ju does not teach or suggest any prediction; (2) Chung-Ju does not
teach or suggest the prediction of the three claimed prediction requirement
using a fuzzy logic and a knowledge base of if-then rules.
First, we find that “predict” means “to declare in advance; esp. foretell
on the basis of observation, experience, or scientific reason.” Webster’s
NEW COLLEGIATE DICTIONARY 906 (1973). In Chung-Ju, a Network
Resource Estimator estimates the capacity required for new connection from
traffic parameters before a new connection is accepted. That is, the Network
Resource Estimator is updated when a new connection is accepted or when
an existing connection is disconnected by the NFCAC. (FF 15). As the
Examiner found, these features result in a prediction (i.e., foretelling) of
network resources using fuzzy logic and a knowledge base of fuzzy if-then
rules. Ans. 26. Appellant agrees that Chung-Ju’s Network Resource
Estimator needs to be updated at various times (Reply Br. 5) and
consequently does not persuasively refute this finding.
Appeal 2010-002365
Application 10/989,067

12
Secondly, Appellant relies upon the fact that Chung-Ju does not
expressly teach or suggest all of the three claimed prediction requirements
using fuzzy logic and a knowledge base of fuzzy if-then rules. Reply Br. 5.
Appellant also argues that a combination of Su with Chung-Ju cannot result
in the claimed invention because such a combination would result in a fuzzy
logic admission control scheme being used in Su’s network. This argument
misapprehends the Examiner’s position.
The Examiner does not suggest that it would have been obvious to
incorporate the entire Chung-Ju system into the Su device. Instead, the
Examiner concludes that, because Chung-Ju teaches predicting network
resources using fuzzy logic and a knowledge base of fuzzy if-then rules for
decision making (Ans. 27), it would have been obvious to apply Chung-Ju’s
decision making to all of the claimed prediction requirements shown in Su in
order to provide more robust and accurate network predictions and network
adjustment as Su’s mobile nodes move into, away, and within an ad hoc
network. Ans. 5, 26-27. We agree.
The elements from each reference could have been combined by the
known method of applying Chung-Ju’s fuzzy teachings to the Su device in
the same way that they are applied in the Chung-Ju system. With such a
combination, the Su device and the Chung-Ju fuzzy teachings would
predictably perform the same functions as they would perform separately.
We therefore affirm the rejection of representative claim 1 and claims
2, 5, 6, 9, and 10 not separately argued with particularity.

Appeal 2010-002365
Application 10/989,067

13
THE REJECTION OF THE REMAINING CLAIMS OVER
SU IN VIEW OF VARIOUS OTHER REFERENCES
We also give separate consideration of claims 3, 7, 8, 11-14, 16-24,
and 26-31 which were rejected over various combinations of Su and Chung-
Ju with Tonjes, Bellur, McKinnin, and Haas. See In re McDaniel, 293 F.3d
1379, 1384 (Fed. Cir. 2002). We affirm the rejections of those claims
because the rejections are not separately argued and Appellant expressly
relies solely upon the recitations of the three claimed prediction
requirements for patentability: “It is these combinations of features which
are not fairly taught or suggested in the cited references and which
patentably define over the cited references.” (emphasis added). App. Br. 6.

DECISION
The Examiner’s decision rejecting claims 1-3, 5-14, 16-24, and 26-31
is affirmed.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1). See 37 C.F.R.
§ 1.136(a)(1)(iv) (2010).

AFFIRMED

rwk