Ex Parte Strolle

Board of Patent Appeals and Interferences

Feb 24, 2006

08869589 (B.P.A.I. Feb. 24, 2006)

1 Application for patent filed June 5, 1997, entitled
"Method and Apparatus for Performing Bandedge Equalization,"
which is based on and claims priority under 35 U.S.C. § 119(e)(1)
from U.S. Provisional Application 60/019,308, filed June 7, 1996.
2 Claim 8 should be objected to since it depends from
rejected independent claim 1.
- 1 -
The opinion in support of the decision being
entered today was not written for publication
and is not binding precedent of the Board.
_______________
UNITED STATES PATENT AND TRADEMARK OFFICE

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES

Ex parte CHRISTOPHER HUGH STROLLE

Appeal No. 2006-0401
Application 08/869,5891

ON BRIEF

Before THOMAS, BARRETT, and DIXON, Administrative Patent Judges.
BARRETT, Administrative Patent Judge.
DECISION ON APPEAL
This is a decision on appeal under 35 U.S.C. § 134(a) from
the final rejection of claims 1, 9, 10, 12, 15, and 16.
Claims 2-8, 11, 13, and 14 are allowed.2
We affirm.
Appeal No. 2006-0401
Application 08/869,589
- 2 -
PRIOR DECISION
A prior decision (pages referred to as "D__") was entered on
February 11, 2004, in this application by Appeal No. 2002-1376.
The decision affirmed the anticipation rejection of claims 1, 9,
10, 12, 15, and 16. This decision is incorporated-by-reference.
BACKGROUND
The invention relates to a method and apparatus for
performing equalization of the amplitudes of the bandedges of a
broadband signal. The equalization of the amplitudes of the
bandedges allows a bandedge timing recovery circuit to produce
substantially jitter-free or stress-free timing signals.
Claims 1 and 12 are reproduced below, where the underlining
indicates limitations added after the prior decision.
1. Apparatus for equalizing the amplitudes of the bandedges
of a broadband signal comprising:
a pre-equalizer for adjusting the amplitudes of the
bandedges of said broadband signal in response to a control
signal such that the amplitudes of the bandedges are made
equal;
a bandedge filter, connected to said pre-equalizer, for
extracting a bandedge signal from said broadband signal; and
a bandedge signal processor, connected to said bandedge
filter, for generating said control signal in response to
said bandedge signal.
Appeal No. 2006-0401
Application 08/869,589
- 3 -
12. A method of equalizing the amplitudes of the bandedges
of a broadband signal comprising the steps of:
adjusting the amplitudes of the bandedges of said
broadband signal in response to a control signal such that
the amplitudes of the bandedges are made equal;
extracting a bandedge signal from said broadband
signal; and
generating said control signal in response to said
bandedge signal.
THE REFERENCE
The examiner relies on the following reference:
Norrell et al. (Norrell) 5,793,821 August 11, 1998
(filed June 7, 1995)
THE REJECTION
Claims 1, 9, 10, 12, 15, and 16 stand rejected under
35 U.S.C. § 102(e) as being anticipated by Norrell.
We refer to the final rejection (pages referred to as
"FR__") entered April 27, 2004, and the refiled examiner's answer
(pages referred to as "EA__") entered July 8, 2005, for a
statement of the examiner's rejection, and to the appeal brief
(pages referred to as "Br__") filed September 29, 2004, and the
reply brief (pages referred to as "RBr__") filed
February 22, 2005, (and refiled September 8, 2005) for a
statement of appellant's arguments thereagainst.
Appeal No. 2006-0401
Application 08/869,589
- 4 -
OPINION
Claims 1 and 12
In our previous decision, which we incorporate-by-reference,
we found that Norrell discloses "adjusting the amplitudes of the
bandedges of said broadband signal in response to a control
signal" (claims 1 and 12) because it discloses adaptive
equalization to compensate for amplitude and phase distortion at
the bandedges (col. 2, lines 10-16):
[T]he power envelope of a signal can be adversely affected
by channel impairments, particularly amplitude and phase
distortion at the band edges. To compensate for such
impairments, many receiver designs include an adaptive
equalizer which compensates for amplitude and phase
distortion on the transmission channel in the data recovery
path. (Emphasis added.)
See D6. That is, channel frequency response deviates from the
ideal of constant amplitude and linear phase (constant delay) and
"equalization" compensates for these nonideal characteristics by
filtering before further processing is done. An "adaptive
equalizer" is one that uses feedback to control the equalization.
Norrell teaches adaptive equalization to compensate for amplitude
and phase distortion at the bandedges and adaptive equalization
adjusts the amplitude and phase in response to a control signal.
Norrell is primarily concerned with compensating for delay
(phase) distortion, but states that the embodiment "also
advantageously compensates for the effects of amplitude
distortion on the communications channel" (col. 7, line 67, to
Appeal No. 2006-0401
Application 08/869,589
- 5 -
col. 8, line 2) and shows an adaptive equalizer as delay line 506
in Fig. 5 (col. 9, lines 34-36) which is "long enough to
compensate for amplitude and delay distortion in general"
(col. 9, lines 45-46). Norrell discloses that "[prior art]
equalization boosts the desired energy at the bandedges, but also
boosts the unwanted energy near the bandedges" (emphasis added)
(col. 9, lines 14-16), which teaches amplifying at the bandedges
for equalizaiton (D6-7). In our previous decision, we stated
that "[a]mplitude equalization means attenuating or amplifying to
make the amplitudes equal" (D8).
In response to our previous decision, appellant amended
independent claim 1 to recite "a pre-equalizer for adjusting the
amplitudes of the bandedges of said broadband signal in response
to a control signal such that the amplitudes of the bandedges are
made equal" (underlining indicates amendment) and amended
claim 12 to recited "adjusting the amplitudes of the bandedges of
said broadband signal in response to a control signal such that
the amplitudes of the bandedges are made equal" (underlining
indicates amendment). The examiner relies on our statement that
"[a]mplitude equalization means attenuating or amplifying to make
the amplitudes equal" (D8) to meet the new limitations (EA11).
Appellant argues that Norrell does not specifically teach
that the bandedges are to have equal amplitudes (Br13; RBr2). It
is argued that the equalizer delay line 506 at Fig. 5 implements
Appeal No. 2006-0401
Application 08/869,589
- 6 -
a time delay and the filter 504 and delay line 506 "merely
compensate for differential delay distortion between the upper
and lower bandedges" (emphasis omitted) (Br15). It is argued
that compensation of time delay differences between the upper and
lower bandedge frequencies is devoid of any teaching of adjusting
the amplitudes of bandedges to make the bandedges equal
(Br15-16). Appellant points out where Norrell discusses
compensating for delay distortion, and argues that this is not
amplitude distortion (Br16). It is argued that to the extent
Norrell compensates for amplitude distortion, Norrell merely
passes the energy in the regions centered at the lower and upper
bandedges and does not teach a control signal to actively adjust
the amplitudes of the bandedges (Br16). It is argued that once
the delay distortion is compensated for in the delay line, the
Lower Band Edge Filter (LBEF) and Upper Band Edge Filter (UBEF)
perform amplitude compensation by attenuating the region between
the lower and upper bandedges, which does not adjust the
amplitudes such that the amplitudes of the bandedges are made
equal (Br17). Appellant argues (Br17; RBr2-3):
The cited section (Norrell, column 9, lines 11-15) is
directed to channel equalization and not the specific
adjustment of bandedges of a broadband signal in response to
a control signal such that the amplitudes of the bandedges
are made equal. Channel equalization generally involves
equalization of the entire frequency response and, as such,
is not the same as adjusting bandedges such that the
amplitudes of the bandedges are made equal. The purpose of
the cited section was to clarify the advantage of sharply
attenuating the midband, i.e., the region between the LBEF
Appeal No. 2006-0401
Application 08/869,589
- 7 -
and the UBEF, in Norrell. Moreover, the cited section is
devoid of any teaching of an adjustment of bandedges to make
them have equal amplitude in response to a control signal,
as generated by the bandedge filter and bandedge signal
processor in Appellant's invention. Thus, the cited section
does not teach the adjusting of amplitudes of the bandedges
of a broadband signal such that the amplitudes of the
bandedges are made equal as in claim 1 of Appellant's
invention.
It is argued that "Norrell teaches adjusting of the amplitude of
the original signal, including the bandedges but nowhere does it
teach specifically making the bandedge amplitudes equal" (Br18).
It is true that Norrell does not expressly state that the
amplitudes of the bandedges are made equal. Nevertheless,
Norrell discloses adaptive equalization to compensate for
amplitude and phase distortion at the bandedges (col. 2,
lines 10-16), which implies that the amplitudes are made equal to
correct for amplitude distortion in the channel. Although
Norrell is concerned with delay distortion, it expressly states
that it also compensates for the effects of amplitude distortion
(col. 7, line 67, to col. 8, line 2; col. 9, lines 1-16 & 44-46).
Norrell uses an adaptive equalizer (filter 504 and delay
line 506) and the adjusting of amplitudes and delay (not just
delay as argued by appellant) is done by adjusting the
coefficients to the filter 504 which control the equalizer. We
disagree with appellant's statement that Norrell is limited to
channel equalization: since the coefficients are based on the
outputs from the LBEF and UBEF, the amplitude and delay
Appeal No. 2006-0401
Application 08/869,589
- 8 -
equalization are based on the bandedge signals. In our opinion,
Norrell's teachings of bandedge amplitude equalization implies
that making the amplitudes of the bandedges equal is inherent.
Appellant does not point out the error in our statement that
"[a]mplitude equalization means attenuating or amplifying to make
the amplitudes equal" (D8) by providing any arguments,
explanation, or evidence, such as dictionary definitions, book
descriptions of amplitude equalization, Rule 132 declarations of
what Norrell teaches one skilled in the art, etc., but merely
argues that Norrell does not expressly state that the amplitudes
of the bandedges are made equal. Therefore, appellant leaves it
us to make a technical determination that the amplitude
compensation in Norrell does not inherently adjust the amplitudes
of the bandedges to be equal without providing any evidence.
Appellant takes no risk of making an argument that might be
wrong. If appellant had argued that the Norrell's teaching of
amplitude equalization of bandedges does not make the amplitudes
equal, and provided some explanation and evidence, instead of
just arguing that Norrell does not expressly teach making the
amplitudes equal, this would be a different situation.
To the extent that appellant relies on the limitation of a
"broadband signal," this term is not defined and appellant has
not argued how the term distinguishes over the signal in Norrell.
Appeal No. 2006-0401
Application 08/869,589
- 9 -
Appellant argues that Norrell does not teach a pre-equalizer
for making the amplitudes of the bandedges equal as recited in
claim 1 (Br14).
Claim 1 does not recite that the "pre-equalizer" is an
equalizer functionally placed before ("pre-") an "equalizer" and
such a limitation will not be implied. The term "pre-equalizer"
is just a name unless it is defined by other structure or
functional relationships. The filter 504 and delay line 506
perform the recited function of the "pre-equalizer" and,
therefore, are considered to be a "pre-equalizer."
For the reasons stated above, we affirm the anticipation
rejection of claims 1 and 12.
Claims 9, 10, 15, and 16
The adaptive equalizer in Norrell compensates for amplitude
(and delay) distortion and we find that this implies inherently
making the amplitudes of the bandedges equal. Appellant has not
explained why this is not true or provided any evidence to the
contrary, but merely argues that it is not expressly disclosed,
and so has not shown error in our finding. If the amplitudes of
bandedges are made equal, this must be done by attenuating one
bandedge to bring it down to the level of the other bandedge as
recited in claims 9 and 15, or by amplifying one of the bandedges
to bring it up to the level of the other as recited in claims 10
and 16, or, more likely, by a combination of attenuation and
Appeal No. 2006-0401
Application 08/869,589
- 10 -
amplification. The anticipation rejection of claims 9, 10, 15,
and 16 is affirmed.
CONCLUSION
The rejection of claims 1, 9, 10, 12, 15, and 16 is
affirmed.
No time period for taking any subsequent action in
connection with this appeal may be extended under 37 CFR
§ 1.136(a)(1)(iv).
AFFIRMED
JAMES D. THOMAS )
Administrative Patent Judge )
)
)
)
) BOARD OF PATENT
LEE E. BARRETT ) APPEALS
Administrative Patent Judge ) AND
) INTERFERENCES
)
)
)
JOSEPH L. DIXON )
Administrative Patent Judge )
Appeal No. 2006-0401
Application 08/869,589
- 11 -
MOSER IP LAW GROUP / SARNOFF CORPORATION
1040 BROAD STREET
2ND FLOOR
SHREWSBURY, NJ 07702