Ex Parte 7,096,242 et al

Patent Trial and Appeal Board

Sep 27, 2012

95000351 (P.T.A.B. Sep. 27, 2012)

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.
95/000,351 02/21/2008 7,096,242 72037.00200.RE02 7268
43023 7590 09/27/2012
Thomas Swenson
1118 13th Street
A-5
BOULDER, CO 80302
EXAMINER
DEB, ANJAN K
ART UNIT PAPER NUMBER
3992
MAIL DATE DELIVERY MODE
09/27/2012 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
____________

WINBOND ELECTRONICS CORPORATION OF TAIWAN and
WINBOND ELECTRONICS CORPORATION AMERICA OF
SAN JOSE, CA
Requesters

v.

THE QUANTUM WORLD CORPORATION
Patent Owner and Appellant
____________

Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,2421
Technology Center 3900
____________

Before KARL D. EASTHOM, STEPHEN C. SIU, and JOSIAH C. COCKS,
Administrative Patent Judges.

COCKS, Administrative Patent Judge.

DECISION ON APPEAL

1 The patent involved in this reexamination proceeding appeal (the “’242
patent”) issued to Scott Wilbur on August 22, 2006.
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

2
A. STATEMENT OF THE CASE
Summary
Patent Owner The Quantum World Corporation (“Quantum”)2 appeals
under 35 U.S.C. §§ 134(b) and 315(a) the Examiner’s decision to reject
claims 14, 20, and 22-25.3 The Third-Party Requesters sought
reexamination of the ‘242 Patent in a paper filed February 21, 2008 but have
filed no briefs in this appeal. The Examiner filed an Examiner’s Answer in
support of his decision.4 We have jurisdiction under 35 U.S.C. §§ 134(b)
and 315(a).
We reverse the Examiner’s decision to reject claims 22-25.
We affirm the Examiner’s decision to reject claims 14 and 20.
The Invention
The ‘242 Patent characterizes its disclosed invention as relating to
“random number generators, and more particularly to a random number
generator that generates true binary random sequences.” (‘242 patent, 1:15-
17.) A distinction is drawn between “true random number generators” and
“pseudorandom number generators.” (Id. at 1:24-28.) That distinction is
expressed as (id.):

2 See Patent Assignment Abstract of Title, Reel 018787 Frame 0706 which
was entered into the record of this proceeding as “Title Report” on March 7,
2008.

3 See Quantum’s “CORRECTED APPEAL BRIEF” filed April 28, 2010
(“App. Br.”).

4 See Examiner’s Answer mailed July 1, 2010.
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

3
Random number generators may be divided into two types: true
random number generators, which generate numbers from a non-
deterministic source, and pseudorandom number generators, which
generate numbers from a deterministic algorithm.

By way of example uses, the ‘242 Patent states “[a] high quality
random number generator is necessary for functions from Monte Carlo-
based studies used in computer modeling of the global economy, to
cryptography, to parapsychological studies, to marketing research, to lottery
games.” (Id. at 1:28-32.)
Claims 22 and 25 are independent claims. Claims 23 and 24 depend
from claim 22. Claims 14 and 20 depend, respectively, from claims 8 and
16, which have been canceled and are not on appeal.
Claim 22, which is illustrative of the appealed subject matter, reads as
follows (App. Br., CA-ii, Claims App’x.):
22. A true random number generator system comprising:

a hardware device for producing a binary true random
sequence of signals; and

a computer for utilizing said binary true random
sequence of signals;

wherein |B2| ≤ 0.002 and |SD(t)| ≤ 0.0004, where |B2| is
the fractional bias in the 1, 0 probability of said binary true
random sequence of signals and SD(t) is the serial dependence
as a function time of said binary true random sequence of
signals.

Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

4
The Prior Art
U.S. Patents
Cohn (or “‘941 patent”) 3,706,941 Dec, 19, 1972

Porter (or “‘366 patent”) 4,355,366 Oct. 19, 1982

Ogrodski (or “‘259 patent”) 4,799,259 Jan. 17, 1989

Brown et al. 4,853,884 Aug. 1, 1989
(“Brown” or “‘884 patent”)

Schulz (or “‘176 patent”) 4,905,176 Feb. 27, 1990

Reinhardt et al. 5,224,165 Jun. 29, 1993
(“Reinhardt” or “‘4165 patent”)

James, III 5,251,165 Oct. 5, 1993
(“James” or “‘1165 patent”)

Davis (or “‘828 patent”) 5,539,828 Jul. 23, 1996

Non Patent Literature
Richter, Manfred. “A noise generator for obtaining quasi-ideal random
numbers for stochastic simulation.” A dissertation submitted to the
Faculty of Electrical Engineering of the Rhineland-Westphalia
Technical Institute at Aachen in order to obtain the academic degree
of Doctor of Engineering. Dated 1992. (“Richter”)5

Murry, Hershell F. “A General Approach for Generating Natural Random
Variables.” IEE Transactions on Computers. December 1970.
(“Murry”)6

5 Exhibit B in the Evidence Appendix of Quantum’s Appeal Brief. (“Ex. B”)
6 Ex. F
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

5
Gude, Michael, “Concept for a High Performance Random Number
Generator Based on Physical Random Phenoms.” FREQUENZ,
Journal of Telecommunications, 1985, pp. 187-190. (“Gude”)7

Radin, Dean I., “Testing the Plausibility of PSI-Mediated Computer System
Failures.” Journal of Parapsychology, Vol. 54, March 1990, pp. 1-19.
(“Radin”)8

IBM Technical Disclosure Bulletin, Volume 30, Number 11, April 1988, pp.
333-335. (“IBM TDB”)9

CALMOS™ NM 810 RNG (Random Number Generator) data sheets.
November 1989. (“NM 810 RNG”)

Admitted Prior Art (“APA”) appearing at column 24, line 53 through
column 25, line 13 (e.g., see Ans., p. 27) and at column 2, lines 46-60 (e.g.,
see Ans., p. 31) of the ‘242 Patent.

The Involved Rejections
Anticipation
The Examiner entered the following rejections under 35 U.S.C.
§ 102(b):
(1) The anticipation of claims 22-25 by Richter;
(2) The anticipation of claims 14 and 20 by Radin;
(3) The anticipation of claim 14 by NM 810 RNG; and
(4) The anticipation of claims 22-25 by Gude.

7 Ex. J
8 Ex. K
9 Ex. L
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

6
Obviousness
The Examiner entered the following rejections under 35 U.S.C.
§ 103(a):
(5) The unpatentability of claim 20 over IBM TBD and Radin;
(6) The unpatentability of claim 20 over IBM TBD, Davis, and
APA;
(7) The unpatentability of claim 20 over Schulz, APA, and Radin;
(8) The unpatentability of claim 20 over NM 810 RNG and APA;
(9) The unpatentability of claims 14 and 20 over Cohn and APA;
(10) The unpatentability of claim 14 over Murry and APA;
(11) The unpatentability of claims 14 and 20 over Porter and Radin;
(12) The unpatentability of claims 14 and 20 over Porter, James, and
APA;
(13) The unpatentability of claims 14 and 20 over Porter, Davis, and
APA;
(14) The unpatentability of claims 14 and 20 over Gude and APA;
(15) The unpatentability of claim 20 over Ogrodski and Radin;
(16) The unpatentability of claim 20 over Ogrodski, James, and
APA;
(17) The unpatentability of claim 20 over Ogrodski, Davis, and
APA;
(18) The unpatentability of claims 14 and 20 over Brown and Radin;
(19) The unpatentability of claims 14 and 20 over Brown, James,
and APA;
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

7
(20) The unpatentability of claims 14 and 20 over Brown, Davis,
and APA;
(21) The unpatentability of claims 14 and 20 over Reinhardt, James,
and Radin;
(22) The unpatentability of claims 14 and 20 over Reinhardt, James,
and APA; and
(23) The unpatentability of claims 14 and 20 over Reinhardt, Davis,
and APA.
B. ANALYSIS
By our count, the Examiner has entered twenty-three separate
rejections in connection with Quantum’s claims. In challenging the
Examiner’s rejections, Quantum argues the claims in two groupings: (1)
claims 22-25; and (2) claims 14, 20, and 25.
Claims 22-25
Issues- Claims 22-25
With respect to claims 22-25 the following issues arise:
(1) Did the Examiner correctly find that Richter discloses a true
random number generator incorporating a “serial dependence” value of less-
than or equal to 0.0004 as required by claims 22 and 25?
(2) Did the Examiner correctly find that Gude discloses a true random
number generator incorporating a “serial dependence” value of less-than or
equal to 0.0004 as required by claims 22 and 25?
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

8
Discussion – Claims 22-25
Common features appearing in claims 22 and 25 are at issue in
connection with each of the separate anticipation rejections based on Richter
and Gude. In particular, claims 22 and 25 each requires the production of “a
binary true random sequence of signals.” (App. Br., CA-ii, Claims App’x.)
In connection with those signals, the claims require specific values for the
“serial dependence” SD(t) and the “fractional bias” B2. (Id. at CA-ii and CA-
ii.)
As explained in the ‘242 Patent, “serial dependence” SD(t) and
“fractional bias” B2 constitute parameters used to evaluate the “randomness
quality” of a given random sequence. (‘242 patent, 25:63-26:25.) In a
theoretical context, a true or optimally random sequence incorporates values
of “0” for each of SD(t) and B2. (Id.) However, real-world random
sequence generators have physical constraints which impose imperfection in
attaining ideal randomness such that those parameters are not zero. (Id.)
The invention of the ‘242 Patent provides a random number generator
configured so as to have values for SD(t) and B2 which are sufficiently small
in magnitude such that it would take an acceptably long period of years to
demonstrate defects in randomness, i.e., indication of non-randomness, with
a 95% confidence level. (Id. at 2:10-22; 27:1-16.)
Claims 22 and 25 each includes a clause conveying the same values
required for SD(t) and B2. In claim 22, the clause reads (App. Br., CA-ii,
Claims App’x.):
wherein |B2| ≤ 0.002 and |SD(t)| ≤ 0.0004, where |B2| is the
fractional bias in the 1, 0 probability of said binary true random
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

9
sequence of signals and SD(t) is the serial dependence as a function
time of said binary true random sequence of signals.

Richter
In accounting for the above-noted values of B2 and SD(t) in Richter,
the Examiner stated (Ans. 12):
Richter dissertation teaches that the fractional bias in the 1, 0
probability is less than 0.002 since the distribution P(0)=P(1) = ½ for
a symmetrical distribution and the serial dependency (autocorrelation)
is less than 0.0004 (ACF is of the order of magnitude of 10-100)
(Section 4.1 page 39 and Section 4.3.2 page, 58). Richter disclosed
that the exact uniform distribution P(0) = P(1) = ½ can be produced
with the help of a symmetrification circuit (see Richter dissertation:
Section 4.3.2).

Thus, the Examiner, pointing to pages 39 and 58 of sections 4.1 and
4.3.2, respectively, in Richter, found that “distribution P(0) = P(1) = ½”
constitutes a fractional bias of less than 0.0002 and an “autocorrelation”
function “ACF” with an order of magnitude of 10-100 constitutes a serial
dependence of less than 0.0004. That is the underlying factual basis for the
Examiner’s conclusion that Quantum’s claims 22 and 25 were anticipated.
Quantum contests the Examiner’s conclusion with respect to each of
the fractional bias values and serial dependence values. With respect to the
serial dependence requirement, Quantum submits the following (App. Br., p.
5) (emphasis in original).
On page 39, Richter states that the calculated autocorrelation
factor for the output of the noise generator is below 10-100. The
Examiner incorrectly relies on this passage as disclosing the serial
dependence of the RNG [Random Number Generator], which is what
the claims address. The correlation factor for the output of the noise
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

10
generator is not the same as the correlation factor at the output of the
RNG.
Section 4.1, in which the cited passage appears, is entitled “4.1
The Noise Generator.” The Richter reference explicitly states that the
correlation factor for the output of the noise generator is “so slight,
relative to the correlations that are brought about in the scanning
circuit by non-ideal components, that it can be disregarded.” Richter,
page 39. In other words, this autocorrelation factor will be dwarfed
by the autocorrelations introduced when the noise signal is digitized.
The Examiner’s identification of Richter’s confirmation of the
estimated 10-100 correlation coefficient (see RAN at 60) does not
change this fact – the measured values are still the measured values of
the noise generator, not the RNG, and the numbers will still be
dwarfed by the correlations introduced by the scanning circuit. The
Examiner’s identification of the correlations coefficient of the noise
generator as disclosing the limitations of claims 22-25 – which specify
the quality of the outputs of the RNG – is incorrect. As a result, the
Examiner has not identified any disclosure by Richter of the serial
dependence levels, SD(t), required by claims 22-25.

Thus, Quantum submits that the “autocorrelation” factor having a
value of 10-100 does not constitute a serial dependence value of a random
number generator (“RNG”) as required by claims 22 and 25. Instead,
Quantum is of the view that the referenced autocorrelation coefficient
simply constitutes a correlation factor associated with a noise generator
which is not in and of itself a true random number generator. In support of
its position, Quantum points to description in Richter that the autocorrelation
factor may be disregarded as inconsequential as compared with another
component – a scanning circuit.
In response to Quantum’s position, the Examiner seemingly modifies
his stance stating (Ans., p. 61):
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

11
Appellant’s arguments (Appeal Brief: pages 4-7) with respect to the
rejection of claims 22-25 under 35 U.S.C. 102(b) as being anticipated
by the Richter dissertation are not convincing because Richter
disclosed that the correlations introduced by non-ideal components in
the scanning circuit are slight and can be disregarded therefore
examiner believes that the correlation at the output of the RNG will be
close to the calculated correlation factor for the output of the noise
generator which is below 10-100.

Thus, the Examiner appears to urge that the disclosed correlation
value of 10-100 for the output noise generator does not itself constitute am
output value for a RNG. Rather, now according to the Examiner, the
correlation value of the output of the RNG will be “close to” the correlation
value of the noise generator and thus the pertinent claims are satisfied.
We are not persuaded by the Examiner. As set forth in Richter, the
random number generator that is disclosed is composed of “five primary
assemblies.” (Richter, p. 4210.) The noise generator and the scanning circuit
each form one of those assemblies. (Id.) In light of Richter’s disclosure at
page 3911, it is evident that the noise generator will have one correlation
coefficient (i.e., 10-100) and the scanning circuit will have another correlation
coefficient. It is not apparent what the actual magnitude of the latter
correlation coefficient is, however, the 10-100 value associated with the noise
generator is disclosed as being “so slight” that it may be “disregarded” as
compared with the correlation coefficient of the scanning circuit. (Richter,

10 We observe that Richter includes several duplicate page numbers,
including two pages “42.” The page referenced here is the first occurrence
of page 42 associated with chapter 4.

11 Richter also includes two pages “39.” The referenced page is the second
occurrence of page 39 associated with section 4.1.
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

12
p. 39.) Even assuming that the respective correlation values disclosed are
each properly viewed as forming “serial dependencies,” we do not discern
how the Examiner arrived at the above-noted conclusion as to the magnitude
of serial dependency of a RNG as a whole. Each of the noise generator and
scanning circuit are simply components of the RNG and, evidently, have
correlation values that differ wildly in terms of magnitude. The Examiner
does not meaningfully articulate why he believes that the correlation value
of the RNG will be “close to” that of any one of its subsidiary components,
e.g., a noise generator.
We have considered the Examiner’s view but, on this record, we do
not discern that the Examiner has suitably established that Richter itself
discloses a true random number generator system with a serial dependence
of less than or equal to 0.0004 as is required by each of claims 22 and 25.
Because anticipation requires that a single prior art reference disclose all
elements of the claimed invention, In re Spada, 911 F.2d 705, 708 (Fed. Cir.
1990), we do not sustain the rejection of claims 22-25 as anticipated by
Richter.
Gude
The Examiner also rejected claims 22-25 as anticipated by Gude. The
issue raised by Quantum in connection with that rejection is essentially the
same as that discussed above regarding the Richter rejection. In rejecting
claims 22-25, the Examiner stated the following (Ans., p. 22):
With respect to claims 22-25, the Gude publication discloses all of the
claimed limitations including true random number generator system
(see Fig. 1, 2 shown below) and producing an utilizing true random bit
stream (Fig. 6, 7) and that the fractional bias in the 1,0 probability of
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

13
the bit stream is less than 0.002 since the Gude publication disclosed
the distribution P0 = P1 = 0.5 (page 188 col. 2, see below) and the
serial dependency (autocorrelation) is less than 0.0004 (ACF < 10-50).

We again turn our focus to the “serial dependence” requirement of
claims 22 and 25, now in the context of Gude. The Examiner found that the
requirements was met in Gude by its disclosure of an “autocorrelation”
function (ACF) less than 10-50. In contesting the Examiner’s position,
Quantum argues the following (App. Br., p. 8):
The Examiner cites column 1 of page 188 as disclosing an ACF less
than 10-50. As discussed above with respect to Richter’s ACF of
10-100, Gude’s disclosure is the ACF of the analog noise signal, not
the output of the RNG.

Thus, as it did in connection with Richter, Quantum contends that the
ACF value relied upon by the Examiner as the required serial dependence is
a value associated with a noise signal as juxtaposed with the output of an
RNG.
In response to Quantum’s contention, the Examiner urges the
following (Ans., pp. 62-63):
Appellant’s arguments that the value of ACF less that [sic] 10-50
disclosed by Gude is for the analog noise signal and not the output of
the RNG is not convincing because Gude disclosed that when the
generator is stationary (as with radioactive generator) ideal uniform
distribution of the generated bits can be achieved. The “generated
bits” disclose by Gude is clearly not “analog noise signal” as stated by
the Appellant.

Thus, the Examiner equates Gude disclosure of an ACF of 10-50 with
something termed “generated bits” rather than a noise signal. Evidently, on
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

14
that basis the Examiner concludes that the claim requirement that a true
RNG has a serial dependence of less than or equal to 0.0004 is met in Gude.
We are not persuaded that the Examiner has directed us to content of
Gude expressing a value of the serial dependence of a true RNG. Page 188
of Gude includes the following statement: “[t]o achieve a low
autocorrelation function (ACF) of the generated bitstream the noise signal
must have a low ACF, too.” (Gude, p. 188, section 3.1.) The ensuing
discussion describes determination of an ACF in connection with an “RC-
filtered noise signal” culminating in the calculation of an “extremely low
value” ACF of less than 10-50. The Examiner has apparently equated Gude’s
disclosure of a “generated bitstream” with the output of a true RNG. Even
assuming that is correct, we share Quantum’s view that the discussion of
obtaining an “ACF” value of less than 10-50 pertains to the noise signal and
not to the generated bitstream. As with Ricther, in Gude, a noise signal is a
core component of its disclosed RNG which is made up of additional
components such as a sampling circuit. (Gude, p. 188, col. 2, section 4.) We
do not discern that the discussion in Gude, and on which the Examiner
relies, accounts for the serial dependence values of a true RNG that are
required by Quantum’s claims 22-25.
For the foregoing reasons, we do not conclude that the Examiner has
appropriately justified the proposed anticipation rejection involving Gude
with adequate underlying support. Accordingly, we do not sustain the
rejection.
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

15
Conclusion – Claims 22-25
(1) The Examiner did not correctly find that Richter discloses a true
random number generator incorporating a “serial dependence” value of less-
than or equal to 0.0004 as required by claims 22 and 25.
(2) The Examiner did not correctly find that Gude discloses a true
random number generator incorporating a “serial dependence” value of less-
than or equal to 0.0004 as required by claims 22 and 25.

Claims 14 and 2012
Issues – Claims 14 and 20
With respect to claims 14 and 20, the following issues arise:
(3) Did the Examiner correctly determine that Radin discloses
“utilizing” random numbers in association with a general-purpose computer?
(4) Did the Examiner correctly determine that Gude discloses an
“interface” between a true RNG and a computer?
(5) Did the Examiner correctly determine one with ordinary skill in
the art would have appreciated from other involved prior art that an
“interface” may be established between a true RNG and a computer?

12 We observe that Quantum also argued claim 25 along with claims 14 and
20. However, the only rejections advanced in connection with claim 25 are
the above-discussed anticipation rejections over each of Richter and Gude.
We have already concluded that those involving claim 25 cannot be
sustained. Therefore, we need not address further any additional arguments
advanced in conjunction with claim 25.

Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

16
(6) Did the Examiner correctly determine one with ordinary skill in
the art would have appreciated from the involved prior art that a “power
connection” may be established between a true RNG and a computer?
(7) Did the Examiner correctly determine that a “source of a clock
signal,” as required by claims 14 and 20, is inherent to the disclosures of
Murry and NM 810 RNG?
Discussion – Claims 14 and 20
Claim 14 is dependent on canceled independent claim 8 and claim 20
is dependent on cancelled independent claim 16. The Examiner rejected
claim 14 as anticipated by each of Radin and NM 810 RNG.13 The
Examiner also rejected claim 20 as anticipated by Radin.14 Finally, the
Examiner rejected claims 14 and 20 as obvious over a variety of
combinations of prior art.15
Claims 14 and 20 recite the production of a binary true random
sequence of signals. Each claim also requires (Claims App’x, App. Br., CA-
i, CA-ii):
a general-purpose computer for utilizing said binary true
random sequence of signals; and

13 These rejections are numbered (2) and (3) in “The Involved Rejections”
section of this opinion.

14 Rejection numbered (2).

15 Rejections numbered (5) through (23).

Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

17
an interface for applying said binary true random sequence of
signals to said general-purpose computer.16

Claim 14 and 20 further recite the presence of a random number
generator circuit with (id.):
a power connection operable to provide power from said
general-purpose computer to said random number generator circuit.

Lastly, the claims set forth “a source of a clock signal” as a part of the
random number generator circuit. (id.)
Each of the above-quoted features is challenged by Quantum in some
fashion in connection with one or more of the prior art rejections which have
been advanced by the Examiner. In shorthand, we refer to those features as:
(1) the “utilizing” requirement; (2) the “interface” requirement; (3) the
“power connection” requirement; and (4) the “clock signal” requirement.
(1) The “utilizing” requirement
We have reviewed Quantum’s Appeal Brief and the Examiner’s
Answer. In light of that review, we discern that the Examiner has relied on
several of the prior art references of record as accounting for the “utilizing”
feature of claims 14 and 20. However, Quantum’s challenge to the
Examiner’s rejections with respect to that feature focuses simply on Radin.
Radin is an article published in the Journal of Parapsychology which
is titled “Testing the Plausibility of PSI-Mediated Computer System
Failures.” Radin describes the use of a “truly random number generator”

16 These limitations involving “utilizing” the binary true random sequence of
signals and an “interface” to the general-purpose computer appear in each of
claims 8 and 16, on which claims 14 and 20 depend, respectively.
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

18
(Radin, p. 3) and includes a circuit including “a random number generator
on a chip” (id. at p. 4) Radin also expresses that a “Commodore Amiga
500TM computer was used to interface to this circuit.” (Id.) Throughout the
Examiner’s Answer, the Examiner has expressed that Radin discloses
utilization of a true random sequence of signals. (E.g., Ans., pp. 26, 29.)
Quantum makes the following argument in connection with alleged
deficiencies in Radin (App. Br., p. 13; see also p. 21) (emphasis in original):
Examiner states that “The Radin publication also disclosed
general-purpose personal computer (Commodore Amiga 500TM) (page
4, 2d paragraph) interfaced with random number generator and
utilizing said binary true random sequence of signals for performing
RNG psi experiments.” RAN at 62. The Examiner is correct that the
closest disclosure by Radin of “utilization” is the use of the numbers
to perform psi experiments. However, the psi experiments
described in the Radin publication consist only of the Author’s
attempt to mentally influence the randomness of the numbers
generated while eating girl scout cookies or viewing erotic pictures.
Radin at 10, 12-13. This is not utilization by a computer as is required
by the claims. Rather, to the extent that it can be called utilization at
all, it is utilization by Radin himself.

At the outset, although Quantum characterizes its claims as requiring
“utilization by a computer,” claims 14 and 20 (via their dependency on
claims 8 and 16, respectively) recite only that the computer is “for utilizing”
the binary true random sequence of signals. Thus, the claims do not require
any active step or actual procedure of utilizing the sequence of signals.
Instead, the computer must simply be capable of using, or be operable to
use, the signals once received. Quantum does not offer any explanation as to
why the computer disclosed in Radin, i.e., the Commodore Amiga 500TM
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

19
computer, is not a suitable component “for utilizing” data constituting a
sequence of signals regardless of whether the data is actually utilized.
In any event, even if claims 14 and 20 were construed as requiring
actually utilization of a random sequence of signals, we are unpersuaded by
Quantum that Radin does not reasonably disclose such utilization. Radin
clearly provides, with respect to the experiments disclosed therein, that “[a]ll
phases of the experiments were computer-controlled, including data
collection, storage, and analyses.” (Radin, Abstract.) In connection with the
data generated and stored on the computer, Radin also conveys that
“randomness checks were performed by a program running on the Amiga
500.” (Id. at p. 6, second paragraph) Furthermore, the computer is
programmed to provide “feedback” with respect to the collected data. In
particular, Radin states (id. at p. 9, second full paragraph):
For a novel form of feedback, I programmed the computer to
use synthesized speech phrases to announce when Sample 4 was about
to be collected (the computer spoke, “Data collection how”), when it
had ended (End data collection”), and the number of bits occurring in
that sample (“Bit sum equals [a number])”.

Lastly, it is apparent from the various tables and graphs in Radin’s
disclosure reflecting the pertinent data content that the computer operates to
ultimately output the data.
In light of Radin’s disclosure, we are not persuaded that a computer
which receives, stores, analyses, provides feedback in association with, and
outputs data content is not reasonably and readily understood as utilizing
that data. We are cognizant of Quantum’s assertion that (App. Br., pp. 10-
11):
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

20
The ‘242 specification provides several examples of utilization by a
computer: “A high quality random number generator is necessary for
functions from Monte Carlo-based studies used in computer modeling
of the global economy, to cryptography, to parapsychological studies,
to marketing research, to lottery Games.” Col. 1, ll. 28-32.

We observe that a utilization characterized as “parapsychological
studies” is seemingly in accord with the psi premised experiments set forth
in Radin. However, even if that comparison is not apt, the claims do not
specify or require any particular type of use of utilization. Example uses set
forth in the specification do not operate to import limitations into the claims.
See Prima Tek II, L.L.C. v. Polypap, S.A.R.L., 318 F.3d 1143, 1148 Fed. Cir.
2003) (“Notwithstanding the fact that the claim language must be examined
in light of the written description, limitations may not be read into the claims
from the written description.”) We have considered Quantum’s argument
with respect to the “utilizing” feature in conjunction with Radin, but, for the
above-discussed reasons, we are not persuaded that it is correct.
Accordingly, we conclude that Radin discloses a computer “for
utilizing” a binary true random sequence of signals as required by claims 14
and 20.
(2) The “interface” requirement
On review of the record, we discern that the Examiner appears to have
relied, in various instances, on Radin, NM 810 RNG, Cohn, Gude, James,
and Davis in accounting for the “interface” requirement of claims 14 and 20.
Quantum does not challenge the Examiner’s position with respect to Radin,
NM 810 RNG, or Cohn, but does dispute that the other above-noted
references are correctly viewed as accounting for that feature.
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

21
With respect to Gude, as is noted by Quantum (App. Br., pp. 14-15),
the reference sets forth that randomly generated data formed as a “bitstream”
are “stored on a data recording medium for later use on a computer.” (Gude,
p. 188, ll. 4-6.) Quantum submits that the term “interface” as it appears in
claims 14 and 20 indicates a direct connection and that the conveyance of
data from an RNG first to a data recording medium then subsequently
delivered to a computer does not establish such a connection between the
RNG and computer. (App. Br., p. 15.) The Examiner does not appear to
disagree with Quantum as to the meaning of “interface” as it is used in the
claims, but instead contends that the term “fast access” appearing in Gude’s
Figure 1 constitutes disclosure of an interface between an RNG and a
computer. (Ans., p. 69.)
On this record, we are unable to discern what “fast access” means or
what it, if anything, it conveys with respect to a connection or interface
between an RNG and a computer. The Examiner does not offer any
meaningful explanation justifying why he believes that the term “fast
access” establishes the required interface. Accordingly, we are unable to
agree with the Examiner in conjunction with Gude’s alleged disclosure of
the required interface of claims 14 and 20.
With respect to James, and Davis, Quantum does not contend that
those references do not disclose a direct connection between a type of RNG
and a computer. Rather, the basis of Quantum’s challenge is that the types
of RNGs disclosed in those references constitute pseudo RNGs rather than
true RNGs. According to Quantum, “an interface between a pseudo-RNG
and a computer does not suggest that a true RNG can be similarly interfaced
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

22
with a computer.” (App. Br., p. 22.) Evidently, in conjunction with that
argument, Quantum urges that the prior art “teaches away” from such an
interface with respect to a true RNG. (Id. at p. 23.)
Quantum attempts to substantiate its “teaching away” argument by
reference to portions of only a select number of references involved in this
appeal, namely; Richter, IBM TDB, Murry, and Gude. (Id. at pp. 23-24.)
Quantum takes the position that the portions of those references provide that
it was an established practice to isolate true RNGs from other circuitry.
Even if that is true, however, Quantum does not meaningfully articulate that
such a practice operates to itself “teach away” from other potentially less
established practices. A reference does not teach away if it merely expresses
a general preference for an alternative invention but does not “criticize,
discredit, or otherwise discourage” investigation into an applicant’s claimed
invention. DePuy Spine Inc. v. Medtronic Sofamor Danek, 567 F.3d 1314,
1327 (Fed. Cir. 2009) (quoting In re Fulton, 391 F.3d 1195, 1201 (Fed Cir.
2004)).
Here, even if the cited references convey that a particular association
of a computer and an RNG, which is presumably not tantamount to an
“interface” therebetween, is a practice that is established or even preferred,
we do not discern that the cited content criticizes, discredits or otherwise
discourages inquiry into other practices, such as those including creating an
“interface” between the components. Indeed, Quantum itself characterizes
the prior art as conveying that establishing electrical connection between a
true RNG and a computer “was disfavored and avoided where possible”
(App. Br., p. 25.) Thus, Quantum itself is evidently of the view that where
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

23
such connection or interface cannot be avoided, those of ordinary skill in the
art would have had the knowledge and competence to establish such an
interface when necessary. We also observe in passing that other of the prior
art of record in this appeal, such as Radin, clearly establish that skilled
artisans knew that “truly random” number generators could be connected to
or interfaced with general purpose computers. (Radin, p. 4.)
For the foregoing reasons, although we agree that the record does not
reasonably convey that Gude discloses the “interface” feature of claims 14
and 20, we do not agree that the other references applied by the Examiner in
accounting for that feature are not properly so applied.
(3) The “power connection” requirement
Based on our review of the record, we discern that the Examiner has
relied on each of Radin, NM 810 RNG, and APA as disclosing the “power
connection” requirement of claims 14 and 20. Quantum does not appear to
dispute the Examiner’s determinations with respect to that requirement in
connection with Radin and NM 810 RNG. Quantum, however, does
challenge the Examiner’s reliance on APA as accounting for a power
connection between a general-purpose and a true RNG.
The relevant portion of the ‘242 Patent which is said to constitute
admitted prior art, i.e., APA, appears at column 24, line 53 through column
25, line 13. The characterization of that which is admitted in the section of
the ‘242 Patent noted above is summarized by the Examiner as follows (Ans,
p. 34):
APA disclosed that general-purpose computer comprising a personal
computer such as IBM PC compatible computer and Apple computer
may be used as is it well known in the PC art that the general-
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

24
purpose computer comprising personal computers are capable of
interfacing circuits and supplying power to circuits (col. 24 line 53 to
col. 25 line 13).

Quantum does not dispute that the above-noted portion of the ‘242
Patent constitutes admitted prior art or contend that the Examiner’s
assessment of the content of the APA is incorrect. Indeed, Quantum states
the following (App. Br., p. 25):
The APA cited by the Examiner say only that interfacing and
powering generic add-on boards and integrated circuits with a
computer were known in the art.

Thus, each of the Examiner and Quantum characterize the APA as
providing that integrated circuits may be interfaced with and powered by a
computer. Rather, the basis of Quantum’s dispute is essentially the same as
that which was proffered in connection with the “interface” feature of the
claims, i.e., that certain references of record allegedly teach away from
providing a power connection between a true RNG and a computer.
For essentially the same reasons discussed above with respect to the
“interface” requirement, we reject Quantum’s teaching away argument.
We are not persuaded that the prior art relied upon by the Examiner in
rejecting claims 14 and 20 would not have reasonably conveyed to one with
ordinary skill in the art that a true RNG may be powered by a computer via a
power connection established between those two components.
(4) The “clock signal” requirement
Claims 14 and 20 reach require the feature of a “source of a clock
signal” as a part of the true random number generator systems set forth in
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

25
those claims. Although the Examiner determined that many of the prior art
references relied upon in rejecting claims 14 and 20 expressly set forth the
above-quoted clock feature, with respect to two references –NM 810 RNG
and Murry – the Examiner takes the position that the feature is inherent.
(E.g., Ans. pp. 19, 34.)
In particular, with respect to NM 810 RNG, the Examiner states that a
source of clock signal is inherent to a “strobe” disclosed in that reference.
(Ans., pp. 19, 75.) With respect to Murry, the Examiner maintains that the
source of a clock signal is “inherent to the circuit shown in Fig. 1,5 [sic].”
(Id. at pp. 34, 76.)
As is well settled, inherency may not be established by probabilities or
possibilities. MEHL/Biophile International Corp. v. Milgraum, 192 F.3d
1362, 1365 (Fed. Cir. 1999). The prior art must necessarily function in
accordance with, or include, the claimed limitations, in order to anticipate.
Id. We have considered the Examiner’s inherency position, but do not
conclude that the Examiner has sufficiently explained that a source of clock
signal is necessarily present. In that regard, we agree with Quantum. (App.
Br., pp. 17-20.) Even if a source of clock signal may possibly, or even
probably, be a part of the RNG circuits of NM 810 RNG and Murray, that is
insufficient to establish inherency.
Conclusion – Claims 14 and 20
(3) The Examiner correctly determined that Radin discloses
“utilizing” random numbers in association with a general-purpose computer.
(4) The Examiner did not correctly determine that Gude discloses an
“interface” between a true RNG and a computer.
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

26
(5) The Examiner correctly determined that one with ordinary skill in
the art would have appreciated from other involved prior art that a true RNG
may be “interfaced” with a computer.
(6) The Examiner correctly determined that one with ordinary skill in
the art would have appreciated from the involved prior art that a “power
connection” may be established between a true RNG and a computer.
(7) The Examiner did not correctly determine that a “source of clock
signal,” as required by claims 14 and 20, is inherent to the disclosures of
Murry and NM 810 RNG.
Summary
In light of our analysis set forth above, we summarize our decision as
follows:
Of the twenty-three rejections involved in this appeal we sustain all
but the following:
The anticipation of claims 22-25 by Richter;
The anticipation of claim 14 by NM 810 RNG;
The anticipation of claims 22-25 by Gude;
The unpatentability of claim 20 over NM 810 RNG and APA;
The unpatentability of claim 14 over Murry and APA;
The unpatentability of claims 14 and 20 over Gude and APA.
C. ORDER
Anticipation – 35 U.S.C. § 102(b)
The Examiner’s decision to reject:
(1) claims 22-25 as anticipated by Richter is reversed;
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

27
(2) claims 14 and 20 as anticipated by Radin is affirmed;
(3) claim 14 as anticipated by NM 810 RNG is reversed; and
(4) claims 22-25 as anticipated by Gude is reversed;
Obviousness – 35 U.S.C. § 103(a)
The Examiner’s decision to reject:
(5) claim 20 over IBM TBD and Radin is affirmed;
(6) claim 20 over IBM TBD, Davis, and APA is affirmed;
(7) claim 20 over Schulz, APA, and Radin is affirmed;
(8) claim 20 over NM 810 RNG and APA is reversed;
(9) claims 14 and 20 over Cohn and APA is affirmed;
(10) claim 14 over Murry and APA is reversed;
(11) claims 14 and 20 over Porter and Radin is affirmed;
(12) claims 14 and 20 over Porter, James, and APA is affirmed;
(13) claims 14 and 20 over Porter, Davis, and APA is affirmed;
(14) claims 14 and 20 over Gude and APA is reversed;
(15) claim 20 over Ogrodski and Radin is affirmed;
(16) claim 20 over Ogrodski, James, and APA is affirmed;
(17) claim 20 over Ogrodski, Davis, and APA is affirmed;
(18) claims 14 and 20 over Brown and Radin is affirmed;
(19) claims 14 and 20 over Brown, James, and APA is affirmed;
(20) claims 14 and 20 over Brown, Davis, and APA is affirmed;
(21) claims 14 and 20 over Reinhardt, James, and Radin is
affirmed;
(22) claims 14 and 20 over Reinhardt, James, and APA is affirmed;
and
Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

28
(23) claims 14 and 20 over Reinhardt, Davis, and APA is affirmed.

In accordance with 37 C.F.R. § 41.79(a)(1), the “[p]arties to the
appeal may file a request for rehearing of the decision within one month of
the date of: . . . [t]he original decision of the Board under § 41.77(a).” A
request for rehearing must be in compliance with 37 C.F.R. § 41.79(b).
Comments in opposition to the request and additional requests for rehearing
must be in accordance with 37 C.F.R. § 41.79(c) & (d), respectively. Under
37 C.F.R. § 41.79(e), the times for requesting rehearing under paragraph (a)
of this section, for requesting further rehearing under paragraph (d) of this
section, and for submitting comments under paragraph (c) of this section
may not be extended.
An appeal to the United States Court of Appeals for the Federal
Circuit under 35 U.S.C. §§ 141-144 and 315 and 37 C.F.R. § 1.983 for an
inter partes reexamination proceeding “commenced” on or after November
2, 2002 may not be taken “until all parties' rights to request rehearing have
been exhausted, at which time the decision of the Board is final and
appealable by any party to the appeal to the Board.” 37 C.F.R. § 41.81. See
also MPEP § 2682 (8th ed., Rev. 8, July 2010).

AFFRIMED-IN-PART; REVERSED-IN-PART

Appeal 2012-006289
Reexamination Control 95/000,351
Patent 7,096,242

29
PATENT OWNER:
HEIM, PAYNE & CHORUSH LLP
600 TRAVIS, SUITE 6710
HOUSTON, TX 77002

THIRD-PARTY REQUESTER:
SHERMAN & STERLING
801 PENNSYLVANIA AVE., NW
SUITE 900
WASHINGTON, DC 20004-2634