Ex parte Tattersall

Board of Patent Appeals and Interferences

Dec 14, 2000

08353258 (B.P.A.I. Dec. 14, 2000)

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was not written for publication and is not binding precedent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. 41
UNITED STATES PATENT AND TRADEMARK OFFICE
____________
BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
____________
Ex parte GRAHAM D. TATTERSALL
____________
Appeal No. 1998-1903
Application No. 08/353,258
____________
HEARD: November 29, 2000
____________
Before HAIRSTON, GROSS, and LEVY, Administrative Patent Judges.
GROSS, Administrative Patent Judge.
DECISION ON APPEAL
This is a decision on appeal from the examiner's final
rejection of claims 1 through 8, 11 through 15, 17, and 21.
Claims 18 through 20 and 22 have been allowed, and claims 9,
10, and 16 have been objected to as depending from rejected
base claims.
Appellant's invention relates to a pattern recognition
device and method of training such a device in which the input
digital signal values are encoded. The code used either is
Appeal No. 1998-1903
Application No. 08/353,258
2
redundant with a normalized Hamming distance to normalized
input signal value distance relationship with a mean slope
greater than 1 for input signal value distances up to at least
10% of the maximum input signal value distance or has a higher
monotonicity than a binary code and has a number of bits
between that required for the binary code and that required
for a bar code. Claims 12 and 13 are illustrative of the
claimed invention, and they read as follows:
12. A method of training a pattern recognition device
which is arranged to receive a set of digital input signal
values representing a pattern of physical occurrences
distributed over time or space to be recognized comprising the
steps of:
encoding said digital input signal values into coded
input signal values to form a digital signal representation of
said pattern using a redundant code which has a normalized
Hamming distance to normalized input signal value distance
relationship with a mean slope greater than 1 for input signal
value distances up to a predetermined level of at least 10% of
the maximum input signal value distance; and
using the coded input signal values to recognize the
pattern of physical occurrences.
13. A method of training a pattern recognition device
which is arranged to receive a set of digital input signal
values representing a pattern of physical occurrences
distributed over time or space to be recognized comprising the
steps of:
encoding said digital input signal values using a code
employing a number of bits which is:
Appeal No. 1998-1903
Application No. 08/353,258
3
i) greater than that which a binary code would
require to represent an equal number of signal
levels; and
ii) less than that which a bar code would require to
represent an equal number of signal levels;
and being selected to have a higher monotonicity
than would such a binary code; and
using the coded input signal values to recognize the
pattern of physical occurrences.
The prior art references of record relied upon by the
examiner in rejecting the appealed claims are:
Johnston 4,782,459 Nov. 01,
1988
Aleksander and T.J. Stonham, "Guide to pattern recognition
using random-access memories," Computers and Digital
Techniques, vol. 2, No. 1 (February 1979), pp. 29-40.
(Aleksander)
Claims 1 through 8, 11 through 15, 17, and 21 stand
rejected under 35 U.S.C. § 103 as being unpatentable over
Johnston in view of Aleksander.
Reference is made to the Final Rejection (Paper No. 26,
mailed May 15, 1996) and the Examiner's Answer (Paper No. 37,
mailed January 21, 1998) for the examiner's complete reasoning
in support of the rejection, and to appellant's Brief (Paper
Appeal No. 1998-1903
Application No. 08/353,258
4
No. 28, filed November 14, 1996) and Supplemental Brief (Paper
No. 34, filed August 28, 1997) for appellant's arguments
thereagainst.
OPINION
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 reverse the obviousness rejection of claims 1
through 8, 11 through 15, 17, and 21.
Both the examiner and appellant apparently agree that
Johnston does not disclose a means for or step of encoding, as
recited in each of the independent claims. We too agree that
Johnston lacks any teaching of encoding. The examiner relies
on Aleksander to provide the particular encoding specified in
the claims. However, we find Aleksander's disclosure to be
insufficient to meet the limitations in the claims, as
discussed infra.
Aleksander discusses (second column, page 36) three types
of codes, a direct binary code, a reflexive Gray code (which
has the same number of bits as a binary code), and a 1-in-n
code (hereinafter referred to as a bar code). In particular,
Appeal No. 1998-1903
Application No. 08/353,258
5
Aleksander explains that both the binary code and the Gray
code have problems with the Hamming distance, and, whereas the
bar code overcomes Hamming distance problems, it is
inefficient and requires a large amount of input space.
Each of independent claims 1, 12, 14, and 21 requires the
code used for encoding to be redundant and the relationship of
the input signal value Hamming distance to the input signal
value distance to have a mean slope greater than 1. Neither
the binary code nor the Gray code is redundant, as recognized
by the examiner (Final Rejection, page 4). Nonetheless, the
examiner concludes (Final Rejection, pages 4-5), that a Gray
code satisfies the claimed code, since a Gray code has the
claimed "normalized Hamming distance to normalized signal
distance relationship" and "is more efficient than the
redundant code." The examiner further contends (Answer, page
4) that "the redundant code as claimed does not patentably
distinguish over the Gray code, and the invention would
perform equally well with the Gray code," since the redundant
code does not "provide any improvement over the Gray code in
the classification of input patterns."
Appeal No. 1998-1903
Application No. 08/353,258
6
The examiner, however, has applied an erroneous standard
for obviousness. The test is not whether the claimed code is
better than the prior art, but rather whether it would have
been obvious to modify the prior art to obtain the claimed
code. In other words, it is improper to ignore a limitation
whether or not the examiner believes it to provide an
improvement over the prior art. Thus, whether or not the "the
invention would perform equally well with the Gray code" is
irrelevant in determining the obviousness of the claimed
invention. Since the examiner has admitted that the Gray code
is not redundant, the Gray code alone fails to satisfy the
claimed code. The issue then becomes whether or not it would
have been obvious to modify the Gray code to have redundancy.
The examiner argues (Final Rejection and Answer, page 5)
that Aleksander suggests that there is a tradeoff between data
redundancy and Hamming-distance problems. As evidence, the
examiner refers to Aleksander's statement (page 36, column 2)
that the bar code "overcomes all the Hamming-distance problems
... [but] is, however, very inefficient and can require a
large amount of input space." However, although the bar code
is known to be redundant, Aleksander does not relate the
Appeal No. 1998-1903
Application No. 08/353,258
7
redundant characteristic of the bar code to the absence of
Hamming-distance problems. Therefore, Aleksander does not
suggest that redundancy is a result effective variable which
can be optimized. Thus, contrary to the examiner's assertion
(Final Rejection and Answer, page 5), Aleksander does not
suggest varying the redundancy of (or, rather, adding
redundancy to) the Gray code. In addition, there is no
teaching or even a hint in Aleksander as to how one would or
could add redundancy to the Gray code. The examiner's
motivation for doing so thus is based solely on hindsight.
Therefore, the examiner has failed to establish a prima facie
case of obviousness, and we cannot sustain the rejection of
claims 1, 12, 14, and 21 and their dependents, claims 2 and 7,
15, and 17.
Regarding claims 8 and 13, the examiner advances the same
line of reasoning considered supra for combining properties of
the bar code and the Gray code. Since we have already
determined that the references lack a teaching to modify the
Gray code, we will reverse the rejection of claims 8 and 13,
as well as claim 11, which depends from 8.
Appeal No. 1998-1903
Application No. 08/353,258
8
CONCLUSION
The decision of the examiner rejecting claims 1 through
8, 11 through 15, 17, and 21 under 35 U.S.C. § 103 is
reversed.
REVERSED
KENNETH W. HAIRSTON )
Administrative Patent Judge )
)
)
)
) BOARD OF PATENT
ANITA PELLMAN GROSS ) APPEALS
Administrative Patent Judge ) AND
) INTERFERENCES
)
)
)
STUART S. LEVY )
Administrative Patent Judge )
Appeal No. 1998-1903
Application No. 08/353,258
9
AG/RWK
Appeal No. 1998-1903
Application No. 08/353,258
10
NIXON & VANDERHYE
1100 NORTH GLEBE ROAD
8TH FLOOR
ARLINGTON, VA 22201