Ex parte Smith

Board of Patent Appeals and Interferences

Jun 20, 2001

08614324 (B.P.A.I. Jun. 20, 2001)

The opinion in support of the decision being entered today was not
written for publication and is not binding
precedent of the Board.
Paper No. 35
UNITED STATES PATENT AND TRADEMARK OFFICE
____________
BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
____________
Ex parte GORDON J. SMITH
____________
Appeal No. 1998-2472
Application No. 08/614,324
____________
ON BRIEF
____________
Before JERRY SMITH, GROSS, and BLANKENSHIP, Administrative
Patent Judges.
GROSS, Administrative Patent Judge.
DECISION ON APPEAL
This is a decision on appeal from the examiner's final
rejection of claims 4 through 8 and 12 through 24, which are
all of the claims pending in this application.
Appellant's invention relates to a method for detecting
defects in a disk drive storage device. The method includes
the steps of deriving a test signal indicative of a parameter
to be measured, detecting an abnormal value by identifying a
sample of the test signal that exceeds a threshold value,
Appeal No. 1998-2472
Application No. 08/614,324
2
identifying a window of samples relative to the abnormal
value, and analyzing the window of samples with a neural
network. Claim 4 is illustrative of the claimed invention,
and it reads as follows:
4. A method for detecting defects in a disk drive
storage device, said disk drive storage device comprising at
least one disk, said method comprising the steps of:
deriving a test signal from said disk drive, said test
signal being indicative of a disk surface parameter to be
measured, said test signal varying as a function of at least
one input parameter over a range of values of said at least
one input parameter;
identifying a sample of said test signal exceeding a
predetermined threshold value to detect an abnormal value of
said test signal indicative of a parameter to be measured;
identifying a window of a plurality of samples of said
signal relative to said detected abnormal value in response to
detecting said abnormal value;
providing a neural network for receiving said identified
window of said plurality of samples including at least one
neural network data structure defining data path adaptive
weights for said disk drive storage device; and
analyzing said identified window of said plurality of
samples with said neural network to determine whether a
predefined defect exists in said disk drive storage device.
The prior art references of record relied upon by the
examiner in rejecting the appealed claims are:
Meyer 4,942,609 Jul. 17,
1990
Appeal No. 1998-2472
Application No. 08/614,324
Since the examiner did not permit entry of the Reply Brief (Paper No.1
30, filed June 20, 1997), we will not consider the arguments made therein.
3
Sheppard et al. (Sheppard) 5,130,936 Jul. 14,
1992
Coker et al. (Coker) 5,168,413 Dec. 01,
1992
Claims 4, 5, 7, 8, 12, 13, 15 through 18, and 20 through
24 stand rejected under 35 U.S.C. § 103 as being unpatentable
over Meyer in view of Sheppard.
Claims 6, 14, and 19 stand rejected under 35 U.S.C. § 103
as being unpatentable over Meyer in view of Sheppard and
Coker.
Reference is made to the Examiner's Answer (Paper No. 29,
mailed April 15, 1997) for the examiner's complete reasoning
in support of the rejections, and to appellant's Brief (Paper
No. 28, filed February 21, 1997) for appellant's arguments
thereagainst.1
OPINION
As a preliminary matter, we note that appellant indicates
on pages 11-12 of the Brief that the claims do not stand or
fall together. Appellant argues the claims in the following
four groups: (1) claims 4, 7, 8, 17, and 20; (2) claims 5, 6,
13, 14, 18, 19, 23, and 24; (3) claims 21 and 22; and (4)
Appeal No. 1998-2472
Application No. 08/614,324
We note that the examiner added Coker to the primary combination of2
Meyer and Sheppard to reject claims 6, 14, and 19. However, since appellant
has relied solely on the arguments for claims 4, 12, and 17, respectively,
with no separate arguments regarding Coker, we will treat claims 6, 14, and 19
with the claims from which they depend, claims 5, 13, and 18.
4
claims 12, 15, and 16. However, the limitation argued for
group two does not appear in claims 18 and 19 and the
limitation argued for group 3 does not appear in claim 22.
Accordingly, we will treat the claims according to the
following six groups: (1) claims 4, 7, 8, 17, and 20; (2)
claims 5, 6, 13, 14, 23, and 24; (3) claims 18 and 19, (4)
claim 21; (5) claim 22, and (6) claims 12, 15, and 16, with
claims 4, 5, 18, 21, 22, and 12, respectively, as
representative.2
We have carefully considered the claims, the applied
prior art references, and the respective positions articulated
by appellant and the examiner. As a consequence of our
review, we will affirm the obviousness rejections of claims 4,
7, 8, 12, 15 through 20, and 22, but reverse the obviousness
rejection of claims 5, 6, 13, 14, 21, 23, and 24.
With regard to the first group of claims, according to
the examiner (Answer, page 4), Meyer discloses all of the
limitations of claim 4 except for analyzing the samples using
a neural network which includes a neural network data
Appeal No. 1998-2472
Application No. 08/614,324
5
structure defining data path adaptive weights. The examiner
turns to Sheppard for a suggestion to use such a neural
network in Meyer's detection of defects. Specifically, the
examiner asserts (Answer, page 4) that "Sheppard et al.
teaches neural network means 100 (Figure 12) for diagnosing a
system defect based on a number of test samples, ... so that
the type of defect of a system can be identified with a great
degree of certainty (col. 3)."
Appellant, in the Brief, does not contest the examiner's
analysis of Meyer. Therefore, we will focus our attention on
the combinability of Sheppard's neural network with Meyer's
method. Appellant argues (Brief, page 15) that neither Meyer
nor Sheppard suggests the last two steps of claim 4, which
involve a neural network. Further, appellant states (Brief,
page 16) that "[t]here is no suggestion found in the cited
Meyer et al.
reference or the Sheppard et al. reference, other than the
teachings of applicant's own application, to justify the
modifications of the Meyer et al. reference as proposed by the
Examiner."
Sheppard's diagnostic tester "has broad application in
determining the nature or condition of a variety of systems,
Appeal No. 1998-2472
Application No. 08/614,324
6
ranging, for example, from determining faults and malfunctions
in electronic or electromechanical systems" (see Sheppard,
column 6, lines 60-64). Thus, Sheppard's tester would clearly
apply to defect detection in a disk drive storage device such
as Meyer's. Further, Sheppard states (column 8, lines 10-21)
that for analyzing test data "[e]xamples of commonly used
diagnostic approaches include ... neural networks," among
others. Therefore, Sheppard would have suggested to the
skilled artisan that neural networks are commonly used for
analyzing test data. Accordingly, appellant's argument
notwithstanding, it would have been obvious to one of ordinary
skill in the art to employ such a commonly used neural network
in Meyer's system to analyze the test data "to determine
whether a predefined defect exists," as recited in claim 4.
Appellant also contends (Brief, pages 16-17) that only
appellant teaches diagnosing the type of defect existing in a
disk drive. However, claim 4 merely requires determining the
existence of a defect, which Meyer clearly does. Therefore,
we will affirm the rejection of claim 4 and the claims grouped
therewith, claims 7, 8, 17, and 20.
Regarding the second group of claims, appellant argues
(Brief, page 17) that the references do not teach or suggest a
Appeal No. 1998-2472
Application No. 08/614,324
7
square wave data pattern. We agree. There is no indication
in either reference that the test signal used is a square wave
data pattern. Consequently, we must reverse the rejection of
claims 5, 6, 13, 14, 23, and 24.
Claim 18 requires that the parameter signal be indicative
of the flyheight of the transducer head with respect to the
disk surface. Meyer states (column 3, lines 42-46) that
"[t]he amplitude of the test output signal is proportional to
the magnitude of the air bearing disturbance." Since the air
bearing disturbance is a disruption in the flying height of
the read/ write transducer, the signal of Meyer is indicative
of the flyheight. Therefore, we will affirm the rejection of
claims 18 and 19.
As to claim 21, appellant contends (Brief, page 17) that
neither reference teaches or suggests normalizing the samples,
as recited in claim 21. We agree. Although Sheppard shows
normalizing values in the flowcharts of Figures 10 and 11,
nowhere does Sheppard suggest a reason for normalizing the
values in the method of Meyer. Accordingly, we will reverse
the rejection of claim 21.
Regarding claims 22 and 12, appellant asserts (Brief,
page 18) that neither reference discloses or suggests using a
Appeal No. 1998-2472
Application No. 08/614,324
8
window of samples which includes a selected number both before
and after the detected abnormal value. Meyer teaches (column
3, lines 52-54) using a set of five adjacent tracks. Although
Meyer does not explicitly state that the window should include
a certain number of samples both before and after a detected
abnormal value, the indication to use five adjacent tracks
would have suggested to the skilled artisan to check the
tracks both before and after the abnormal value. Therefore,
we will reverse the rejection of claims 22, 12 and the claims
grouped with claim 12, claims 15 and 16.
Appeal No. 1998-2472
Application No. 08/614,324
9
CONCLUSION
The decision of the examiner rejecting claims 4 through 8
and 12 through 24 under 35 U.S.C. § 103 is affirmed as to
claims 4, 7, 8, 12, 15 through 20, and 22 and reversed as to
claims 5, 6, 13, 14, 21, 23, and 24. Accordingly, the
decision of the examiner is affirmed-in-part.
No time period for taking any subsequent action in
connection with this appeal may be extended under 37 CFR
§ 1.136(a).
AFFIRMED-IN-PART
JERRY SMITH )
Administrative Patent Judge )
)
)
)
) BOARD OF PATENT
ANITA PELLMAN GROSS ) APPEALS
Administrative Patent Judge ) AND
) INTERFERENCES
)
)
)
HOWARD B. BLANKENSHIP )
Administrative Patent Judge )
Appeal No. 1998-2472
Application No. 08/614,324
10
Appeal No. 1998-2472
Application No. 08/614,324
11
JOAN PENNINGTON
535 NORTH MICHIGAN AVENUE
UNIT 1804
CHICAGO, IL 60611