Ex Parte Jung et al

Patent Trial and Appeal Board

Jun 24, 2014

10909200 (P.T.A.B. Jun. 24, 2014)

UNITED STATES PATENT AND TRADEMARK OFFICE
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
10/909,200 07/30/2004 Edward K.Y. Jung SE1-0025-US 5385
80118 7590 06/24/2014
Constellation Law Group, PLLC
P.O. Box 580
Tracyton, WA 98393
EXAMINER
SYED, FARHAN M
ART UNIT PAPER NUMBER
2165
MAIL DATE DELIVERY MODE
06/24/2014 PAPER
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte EDWARD K. Y. JUNG and CLARENCE T. TEGREENE
____________

Appeal 2011-011634
Application 10/909,200
Technology Center 2100
____________

Before MAHSHID D. SAADAT, BRUCE R. WINSOR, and
CATHERINE SHIANG, Administrative Patent Judges.

WINSOR, Administrative Patent Judge.

DECISION ON APPEAL
Appellants appeal under 35 U.S.C. § 134(a) from the Final Rejection
of claims 1-27. Claims 28-108 are withdrawn from consideration. We have
jurisdiction under 35 U.S.C. § 6(b).
We affirm.

Appeal 2011-011634
Application 10/909,200

2
STATEMENT OF THE CASE
Introduction
Appellants’ disclosure relates to the retrieval of stored occurrence data
matching a selected representative feature using pattern recognition criteria.
(Spec. 5:18-29.) In a preferred embodiment, the occurrence data includes
sensor data generated by a plurality of networked remote sensor nodes or
motes. (Spec. 14:6-7.) A mote is typically composed of sensors, actuators,
computational entities, and/or communications entities, and is typically
understood to be a semi-autonomous computing, communication, and
sensing device. (Spec. 4:17-26; 5:9-14; 14:6-11.) In Appellants’ disclosed
invention, the motes use their sensors to acquire data related to one or more
parameters of an environment in which they are located, store the acquired
data in their own data storage device, and transmit the acquired data to a
central computing device where it is stored in a central digital storage
device. (Spec. 14:16-27; 15:10-26; 19:14-20:12; 21:4-10; 22:12-23:8.) The
sensor data is acquired and stored as a chronological sequence that may have
at least one representative signal feature defining an “occurrence”, such as a
car accident, a siren sound, a gun shot, etc. (Spec. 16:22-17:28.) A target
occurrence of interest is defined by an input selection performed by a user,
machine, or computer program, and the input selection may include the
selection of one or more representative features of the target occurrence.
(Spec. 17:24-28; 21:25-26; 23:14-27; 24:18-24.) Pattern recognition criteria
corresponding to the one or more representative features of the target
occurrence is selected, and the stored chronological sequence of sensor data
is automatically searched using the selected pattern recognition criteria to
recognize when one or more signal features contained within the stored data
Appeal 2011-011634
Application 10/909,200

3
sequence correlate to the signal features representative of the target
occurrence. (Spec. 17:1-28; 24:5-25:5.) An output indicative of a result of
the automatic search is generated. (Spec. 5:24-27; 21:26-27; 37:9-13; 40:27-
41:6; 41:15-42:15.) The functions of receiving an input selection, selecting
pattern recognition criteria, automatically searching the sensor data, and
providing an output are all performed by a computing device in accordance
with programmed instructions. (Spec. 5:18-29; 20:9-21; 22:1-11; 28:3-12.)
Exemplary independent claim 1 reads as follows:
1. An occurrence-data retrieval system, comprising:
(a) a data storage operable to store a plurality of instances of
occurrence-data provided by multiple sensors, each instance of the
occurrence-data having a representative feature;
(b) a central computing device operable to communicate with
the data storage; and
(c) instructions that cause a computing device to perform steps
including:
(i) receive from an input-selector an input selection
corresponding to a target-occurrence having a representative
feature;
(ii) select a pattern recognition criteria corresponding to
the representative feature of the target-occurrence, the pattern
recognition criteria including a chronological sequence;
(iii) automatically search the plurality of instances of
stored occurrence-data for data correlating to the target-
occurrence using the selected pattern recognition criteria; and
(iv) provide an output indicative of a result of the
automatic search.

Applied Prior Art
The Examiner relies on the following prior art in rejecting the claims:
Lamming US 5,539,665 July 23, 1996

Appeal 2011-011634
Application 10/909,200

4
Sam Madden et al. TinyDB: In-Network Query Processing in TinyOS,
September 2003 (hereinafter “Madden TinyDB”).
Samuel Madden et al. Supporting Aggregate Queries Over Ad-Hoc
Wireless Sensor Networks, June 21, 2002 (hereinafter “Madden 1”).
The Examiner’s Rejections
Claims 1-20 stand rejected under 35 U.S.C. § 103(a) as unpatentable
over Madden TinyDB, and Lamming. (See Ans. 5-11.)
Claims 21-27 stand rejected under 35 U.S.C. § 103(a) as unpatentable
over Madden TinyDB, Lamming, and Madden 1. (See Ans. 11-16.)
Appellants’ Contentions
1. With respect to the rejection of claim 1, Appellants contend the
combination of Madden TinyDB, and Lamming is improper because:
(a) there is no motivation to combine Madden TinyDB, and Lamming
and these references may not be properly combined in the manner suggested
by the Examiner because Madden TinyDB states that temporal queries are
not supported and, therefore, cannot operate in the manner taught by
Lamming (App. Br. 38);1
(b) Lamming stores a chronological sequence of data, but only
searches for discrete events and, therefore, does not teach searching the
stored data for data correlating to a target occurrence using a selected pattern
recognition criteria including a chronological sequence (App. Br. 38-39 );
and

1 We refer to the Appeal Brief filed March 31, 2011. The pages of the
Appeal Brief are not numbered. Consistently with the Table of Contents in
the Appeal Brief (App. Br. 2-3), we refer to the pages of the Appeal Brief
consecutively, with the page bearing the caption information referred to as
page 1 and the page bearing the signature of Appellants’ attorney referred to
as page 43.
Appeal 2011-011634
Application 10/909,200

5
(c) the Examiner has not met his burden to establish a prima facie case
of unpatentability by providing objectively verifiable evidence establishing
that Lamming teaches searching the stored data using a selected pattern
recognition criteria including a chronological sequence (App. Br. 39).
2. Appellants contend that dependent claims 2-4 are separately
patentable, but rely only on contentions 1a and 1b and offer no additional
arguments addressing the Examiner’s mapping of the teachings of Madden
TinyDB and Lamming to the subject matter of claims 2-42 (App. Br. 39-41).
Thus, claims 2-4 stand or fall with claim 1.
3. Appellants argue the patentability of dependent claims 5-
27based on the reasons asserted for base claim 1 (App. Br. 39). Thus,
claims 5-27 also stand or fall with claim 1.
The Examiner’s Findings – Claim 1
The Examiner finds (Ans. 5-6) Madden TinyDB teaches a query
processing system for extracting stored occurrence data from a network of
motes each containing one or more sensors (Madden TinyDB, § 2, 1st ¶).
Given a query specifying the user’s data interests, i.e., specifying a
representative feature of interest, the query processing system collects the
queried occurrence data from the network of motes in the environment,
filters it, and aggregates it together, i.e., the queried data matching a selected
representative feature is collected and placed in data storage (id.). The
aggregated data is routed out of the mote network to a central computing

2 To show error in the Examiner’s position, Appellants must explain why the
relied-upon disclosure does not teach or suggest the claimed feature under its
broadest reasonable interpretation in light of the Specification, rather than
merely reciting the claimed feature and alleging that it is different. See
discussion infra.
Appeal 2011-011634
Application 10/909,200

6
device in the form of a PC (id.). The query is input using a simple SQL-type
graphical user interface, i.e., an input selector, by which the user defines a
target occurrence of interest having at least one representative feature (id., §
2, 1st ¶; § 2.1, last ¶; §§ 3.3-4.1). The query processor automatically
searches the stored occurrence data using a selected pattern recognition
criteria to determine which occurrence data matches the at least one
representative feature of the target occurrence defined by the user’s query
(id., § 2, 1st ¶; §§ 2.1, 4, 4.1). An output indicative of a result of the
automatic search is generated (id., § 2, 2nd ¶; § 2.1, last ¶; §§ 3.3, 4.1). The
functions of receiving an inputted query, selecting pattern recognition
criteria, automatically searching the sensor data, and providing an output are
all performed by the query processing system, i.e., a computing device, in
accordance with programmed instructions (id., § 2, 1st ¶).
The Examiner finds (Ans. 6) Madden TinyDB does not teach that the
pattern recognition criteria used by the query processor include a
chronological sequence. The Examiner further finds (Ans. 6-7) Lamming
teaches an analogous system to that of Madden TinyDB for collecting,
storing, and retrieving occurrence data detected by a plurality of sensors. In
Lamming’s system, the stored occurrence data is organized as a set of
threads, with each of the threads comprising a chronological sequence of
time-stamped sensor data. (Lamming, col. 2, ll. 52-56; col. 2, l. 60-col. 3, l.
1; col. 5, ll. 38-59). The system is operable through a user interface to
enable a user to retrieve from the stored occurrence data a chronological
sequence of target occurrences defined by the user (id., col. 3, ll. 8-21, 43-
57). The user’s query may define the target occurrences of interest by
specifying a pattern within the stored chronological sensor data (id., col. 6,
Appeal 2011-011634
Application 10/909,200

7
ll. 52-58; col. 8, ll. 26-29; col. 9, ll. 65-67). The query processor
automatically searches the stored occurrence data using selected pattern
recognition criteria to determine which chronological sequences of
occurrence data match the pattern defined by the user’s query, and the
matching data sequences are displayed to the user (id., col. 7, ll. 1-19; col.
10, ll. 3-24; Fig. 5). Based on this teaching in Lamming, the Examiner
concludes (Ans. 7) that it would be obvious to modify Madden TinyDB to
include a chronological sequence in the pattern recognition criteria used by
the query processor in order to provide information relevant to activities in a
sensor network. That is, the stored sensor data can be used to recognize not
only point-in-time events occurring in the sensor network but also activities
or occurrences defined by a chronological sequence of sensor data.
Issues on Appeal
Has the Examiner erred in rejecting claims 1-27 as obvious over
various combinations of Madden TinyDB, Lamming, and Madden 1?

ANALYSIS
We have reviewed the Examiner’s rejections in light of Appellants’
contentions that the Examiner has erred. We agree with the Examiner’s
findings and conclusions with respect to the rejections under § 103.
Appellants’ Reply Brief (filed Oct. 6, 2011) brings to our attention the
adequate notice requirements discussed in In re Jung, 637 F3d 1356, 1362
(Fed. Cir. 2011). We find the rejections made by the Examiner give
adequate notice to Appellants to determine whether to continue prosecution
and to counter the rejections. See 35 U.S.C. § 132(a).

Appeal 2011-011634
Application 10/909,200

8
35 U.S.C. § 103 Rejections
As an initial matter, we note Appellants have styled their arguments as
a challenge to the existence of a prima facie case of unpatentability of the
claims at issue (App. Br. 39).
With respect to the rejections of record, and as set forth supra, the
Examiner specifically puts Appellants on notice that the Examiner
considered the query processing system of Madden TinyDB as
corresponding to the claimed occurrence-data retrieval system (see Ans. 5-
6).
The Examiner similarly makes explicit findings regarding the
elements disclosed by Lamming and the other prior art relied upon, and
explains how the elements are interpreted as corresponding to the respective
features of the appealed claims (see generally, Ans. 6-16). As explained in
In re Jung, 637 F3d 1356 (Fed. Cir. 2011), these findings and explanations
would have put reasonable Applicants on notice of the Examiner’s rejection,
and given them ample information with which to counter the grounds of
rejection, which is all that is required to establish a prima facie case of
unpatentability.
Moreover, in order to rebut a prima facie case of unpatentability,
Applicants are required to distinctly and specifically point out the supposed
errors in the Examiner’s action as well as the specific distinctions believed
to render the claims patentable over the applied references. See 37 C.F.R.
§ 41.37(c)(vii) (“A statement which merely points out what a claim recites
will not be considered an argument for separate patentability of the claim.”);
see also In re Lovin, 652 F.3d 1349, 1357 (Fed. Cir. 2011) (“[W]e hold that
the Board reasonably interpreted Rule 41.37 to require more substantive
Appeal 2011-011634
Application 10/909,200

9
arguments in an appeal brief than a mere recitation of the claim elements and
a naked assertion that the corresponding elements were not found in the
prior art.”); cf. In re Baxter Travenol Labs., 952 F.2d 388, 391 (Fed. Cir.
1991) (“It is not the function of this [board] to examine the claims in greater
detail than argued by an appellant, looking for [patentable] distinctions over
the prior art.”).
Appellants have failed to do this at least with respect to contention 1c,
which ignores the Examiner’s mapping of the teachings of Madden TinyDB
and Lamming to the subject matter of claim 1, and the articulated reasoning
with specific rational underpinning provided by the Examiner to support the
conclusion of obviousness (see Ans. 5-7). Therefore, with respect to
Appellants’ contention 1c, in each of the rejections of record, the Examiner
has cited specific portions of the prior art references, and provided an
explanation as to how the disclosed elements are interpreted as
corresponding to the features of the claims. The cited portions of the
references and the accompanying explanations collectively constitute the
“objectively verifiable evidence” which Appellants allege to be lacking.
Additionally, we are unpersuaded by Appellants’ above-noted
contention because the test of whether a reference teaches a claim limitation
is not whether the exact language is present but rather whether the
limitations are taught or suggested by the prior art when the claim is given
its broadest reasonable interpretation in light of the Specification. Cf. In re
Bond, 910 F.2d 831, 832-33 (Fed. Cir. 1990) (citing Akzo N.V. v. U.S. Int’l
Trade Comm’n, 808 F.2d 1471, 1479 & n.11 (Fed. Cir. 1986))
(interpretation of references “is not an ‘ipsissimis verbis’ test.”); Standard
Havens Prods., Inc. v. Gencor Indus., Inc., 953 F.2d 1360, 1369 (Fed. Cir.
Appeal 2011-011634
Application 10/909,200

10
1991). “[A] . . . reference . . . need not duplicate word for word what is in
the claims.”).
Regarding Appellants’ contention 1a, we agree with the Examiner’s
finding that Lamming teaches an analogous system to that of Madden
TinyDB for collecting, storing and retrieving occurrence data detected by a
plurality of sensors (Ans. 6-7). We also agree with the Examiner that
Lamming’s teaching of pattern recognition criteria capable of recognizing a
chronological sequence of stored occurrence data offers an improvement
over the query processing system for extracting stored occurrence data of
Madden TinyDB (id. 16-17). Specifically, Lamming’s system enables a user
to specify a pattern representing target occurrences of interest to the user
(Lamming, col. 3, ll. 8-21, 43-57; col. 6, ll. 52-58; col. 8, ll. 26-29; col. 9, ll.
65-67). The system automatically searches the stored occurrence data using
a selected pattern recognition criteria to determine which chronological
sequences of occurrence data match the pattern defined by the user’s query
and displays the matching data sequences to the user (id., col. 7, ll. 1-19; col.
10, ll. 3-24; Fig. 5). One of ordinary skill in the art would have recognized
from the teachings of Lamming that such an approach (using pattern
recognition criteria including a chronological sequence to identify stored
occurrence data matching a pattern defined by a query) could be applied to
Madden TinyDB by using the query processing system of Madden TinyDB
to apply a pattern recognition criteria including a chronological sequence to
the stored sensor data collected by the motes of Madden TinyDB. Further,
one of ordinary skill in the art would have appreciated that such a
modification to Madden TinyDB would improve the query processing
system by enabling it to identify occurrences defined by a chronological
Appeal 2011-011634
Application 10/909,200

11
sequence of sensor data. That is, using pattern recognition criteria including
a chronological sequence, as taught by Lamming, enables the system to
recognize not only point-in-time events occurring in the sensor network but
also activities or occurrences defined by a chronological sequence of sensor
data. Thus, Appellants’ contention 1a, and specifically the argument that
there is no motivation to combine Madden TinyDB and Lamming, does not
persuasively rebut the articulated reasoning with specific rational
underpinning provided by the Examiner to support the conclusion of
obviousness (see Ans. 5-7).
Further, the teaching-suggestion-motivation (TSM) test can be a
helpful insight, but cannot be used as a rigid and mandatory formula. See
KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 419 (2007). Rather, rejections
based on obviousness must be supported by “some articulated reasoning
with some rational underpinning” to combine known elements in the manner
required by the claim. Id. at 418 (citing In re Kahn, 441 F.3d 977, 977 (Fed.
Cir. 2006)). Other rationales for supporting the Examiner’s conclusion of
obviousness include, inter alia, applying known techniques to improve
similar devices, methods, or products in the same way. Id. at 417. Here, the
Examiner reasons that it would have been obvious to use the known
techniques of Lamming to improve the similar system of Madden TinyDB in
the same way as Lamming’s system is improved (see Ans. 7, 17).
Further, regarding Appellants’ contention 1a, and specifically the
argument that Madden TinyDB cannot operate in the manner taught by
Lamming because Madden TinyDB states that temporal queries are not
supported, one cannot show nonobviousness by attacking references
individually when the rejection is based on a combination of references. In
Appeal 2011-011634
Application 10/909,200

12
re Keller, 642 F.2d 413, 425 (CCPA 1981). Each reference cited by the
Examiner must be read, not in isolation, but for what it fairly teaches in
combination with the prior art as a whole. See In re Merck & Co., Inc., 800
F.2d 1091, 1097 (Fed. Cir. 1986). All of the features of the secondary
reference need not be bodily incorporated into the primary reference (see
Keller, 642 F.2d at 425) and the artisan is not compelled to blindly follow
the teaching of one prior art reference over the other without the exercise of
independent judgment (see Lear Siegler, Inc. v. Aeroquip Corp., 733 F.2d
881, 889 (Fed. Cir. 1984)).
Thus, contention 1a does not persuade us of error in the Examiner’s
position because Appellants have not persuaded us that the relied-upon
disclosure of the combined references does not teach or suggest the claimed
feature the Examiner mapped to that part of the reference. See KSR, 550
U.S. at 418.
Further, it is noted that Madden TinyDB describes storing the sensor
data in mote memory and retrieving it for a specified sample period
(Madden TinyDB, § 4.6). This is consistent with Lamming’s system of
storing and retrieving chronological sequences of sensor data. That is,
sensor data retrieved for a specified sample period is similar to sensor data
retrieved for a given chronological sequence. Still further, Madden TinyDB
also teaches using a relational database to store all of the sensor data for later
retrieval and analysis (id., § 6). Again, this is similar to Lamming’s system
of storing chronological sequences of sensor data for later retrieval and
analysis. This provides further evidence, in addition to that already
discussed above, that one of ordinary skill in the art would have readily
recognized the teachings of Lamming could be applied to Madden TinyDB.
Appeal 2011-011634
Application 10/909,200

13
Regarding Appellants’ contention 1b, we agree with the Examiner’s
findings that Lamming teaches occurrence data stored as a chronological
sequence, a user-defined target occurrence of interest specifying a pattern
within the stored chronological sensor data, and a query processor that
automatically searches the stored occurrence data using a selected pattern
recognition criteria including a chronological sequence to determine which
chronological sequences of occurrence data match the pattern defined by the
user’s query (Ans. 6-7). We further agree with the Examiner that this
interpretation is consistent with Appellants’ disclosure (see Spec. 18:29-
19:13), which defines pattern recognition criteria as including anything that
recognizes, identifies or establishes a correspondence with one or more
representative features of an occurrence (Ans. 18; see Spec. 19:4-6).
Further, Appellants’ argument that Lamming only searches for
discrete events is not persuasive because Lamming’s system stores
chronological sequences of occurrence data organized as threads (Lamming,
col. 2, ll. 52-56; col. 2, l. 60-col. 3, l. 1; col. 5, ll. 38-59), and enables a user
to retrieve from the stored occurrence data a chronological sequence of
target occurrences defined by the user (Lamming, col. 3, ll. 8-21, 43-57).
The matching chronological data sequence (not just discrete events) is
displayed to the user (Lamming, col. 7, ll. 1-19; col. 10, ll. 3-24; Fig. 5).
For these reasons, we are not persuaded that Lamming fails to teach
searching the stored data for data correlating to a target occurrence using a
selected pattern recognition criteria including a chronological sequence.
In view of our analysis above, Appellants’ contentions have not
persuaded us of error in the Examiner’s findings and conclusion that claims
Appeal 2011-011634
Application 10/909,200

14
1-27 are obvious over various combinations of Madden TinyDB, Lamming,
and Madden 1.

DECISION
The Examiner’s decision rejecting claims 1-27 under 35 U.S.C.
§ 103(a) 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)(iv).

AFFIRMED

kme