Apple Inc.

Patent Trials and Appeals Board

Nov 16, 2021

2020005414 (P.T.A.B. Nov. 16, 2021)

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.
14/721,630 05/26/2015 Su Khiong Yong 8888-40501 1007
81310 7590 11/16/2021
Kowert Hood Munyon Rankin & Goetzel (Apple)
1120 S. Capital of Texas Hwy
Building 2, Suite 300
Austin, TX 78746
EXAMINER
LEE, TSU-CHANG
ART UNIT PAPER NUMBER
2128
NOTIFICATION DATE DELIVERY MODE
11/16/2021 ELECTRONIC
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.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
patent_docketing@intprop.com
PTOL-90A (Rev. 04/07)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________

Ex parte SU KHIONG YONG, CHRISTIAAN A. HARTMAN,
KRISTEN A. MCINTYRE, YONG LIU, and
CHIU NGOK E. WONG

Appeal 2020-005414
Application 14/721,630
Technology Center 2100
____________

Before RICHARD M. LEBOVITZ, JOHN A. EVANS, and
JAMES W. DEJMEK, Administrative Patent Judges.

LEBOVITZ, Administrative Patent Judge.

DECISION ON APPEAL
The Examiner rejected claims 1–3 and 5–21 under 35 U.S.C. § 103 as
obvious. Pursuant to 35 U.S.C. § 134(a), Appellant1 appeals from the
Examiner’s decision to reject the claims. We have jurisdiction under
35 U.S.C. § 6(b).
We REVERSE and set forth new ground rejection pursuant to
37 C.F.R. § 41.50(b).

1 We use the word “Appellant” to refer to “applicant” as defined in
37 C.F.R. § 1.42 (2019). Appellant identifies the real party in interest as
Apple Inc. Appeal Br. 2.
Appeal 2020-005414
Application 14/721,630

2
STATEMENT OF THE CASE
The Examiner rejected claims 1–3 and 5–21 in the Final Office Action
as follows:
Claims 1–3 and 6 under 35 U.S.C. § 103 as obvious in view of
Goering et al. (“Service Discovery Using Bloom Filters,” Proceedings of the
twelfth annual conference of the Advanced School for Computing and
Imaging, 219–227, June 2006) (“Goering”), Haddad et al. (US 9,179,305
B2, issued Nov. 3, 2015) (“Haddad”), and Yang et al. (US 7,333,464 B2,
issued Feb. 19, 2008) (“Yang”). Final Act. 3.
Claim 5 under 35 U.S.C. § 103 as obvious in view of Goering,
Haddad, Yang, and Santoro et al. (US 8,010,407 B1, issued Aug. 30, 2011)
(“Santoro”). Final Act. 8.
Claims 7 and 10–20 under 35 U.S.C. § 103 as obvious in view of
Yang and Haddad. Final Act. 9.
Claims 8, 9, and 21 under 35 U.S.C. § 103 as obvious in view of
Yang, Haddad, and Stephenson et al. (US 7,535,884 B2, issued May 19,
2009) (“Stephenson”). Final Act. 16.
There are three independent claims, claims 1, 7, and 15. Claim 1 is
illustrative and reproduced below (bracketed numbers have been added to
reference certain limitations in the claim):
Appeal 2020-005414
Application 14/721,630

3
1. A wireless device, comprising:
a radio; and
a processing element operatively coupled to the radio;
wherein the radio and the processing element are
configured to:
[1] select a plurality of parameter values for use in
generating a probabilistic data structure;
[2] generate, using the selected plurality of parameter
values, the probabilistic data structure, wherein the probabilistic
data structure provides a service hint map for at least one
service available via the wireless device;
[3] wirelessly transmit the probabilistic data structure;
and
[4] wirelessly transmit an indication of one or more of
the plurality of parameter values used to generate the
probabilistic data structure, wherein a false-positive probability
associated with the probabilistic data structure can be
determined using the indication of one or more of the plurality
of parameter values.

CLAIM INTERPRETATION
We begin with the interpretation of claim 1. During prosecution,
claims are given their broadest reasonable interpretation in light of the
Specification. In re Morris, 127 F.3d 1048, 1054 (Fed. Cir. 1997).
As an overview, the claim relates to using a Bloom filter
(“probabilistic data structure”) to discover services advertised/available to a
wireless device. Specifically, claim 1 is directed to wireless device
comprising a radio and a processing element operatively coupled to the
radio. The radio/processing element are “configured to” perform certain
functions recited in the claim that we have numbered [1] to [4].
The phrase “configured to” requires that “the apparatus as provided
must be ‘capable’ of performing the recited function, not that it might later
be modified to perform that function.” Typhoon Touch Techs., Inc. v. Dell,
Appeal 2020-005414
Application 14/721,630

4
Inc., 659 F.3d 1376, 1380 (Fed. Cir. 2011) (analyzing Microprocessor
Enhancement Corp. v. Texas Instruments, Inc., 520 F.3d 1367 (Fed. Cir.
2008)). Therefore, claim 1 does not require that the recited functions actually
are performed, but instead, only that the radio/processing element is able to
perform the function without any modification, such as additional program
instructions, etc.
All the limitations in the claim refer to a “probabilistic data structure.”
The only example of a probabilistic data structure described in the
Specification is of a “Bloom filter.” Spec. ¶ 32. The Specification discloses
that a Bloom filter “may be used to test whether or not an element is a
member of a set.” Id. The Specification describes the Bloom filter as being
used to discover the set of services available via a wireless device. Spec.
Title (“Dynamic Bloom Filter Operation for Service Discovery”), ¶¶ 5–7, 37
(“an AP providing a wireless network in a mall might provide such services
as a mall directory and/or map service (e.g., with location service), on-sale
and coupon service(s), printing services, etc.”).
The Specification discloses that a Bloom filter “may be configured
such that it is possible to obtain a false positive result (with a configurable
false positive probability depending on the configuration of the filter), but
that a false negative result is not possible.” Spec. ¶ 32.
The first two numbered limitations [1] and [2] of the claim require
that the radio and processing element be “configured to” select “parameter
values” for the data structure and to generate the data structure using the
parameter values. The Specification describes parameters as values used to
generate the Bloom filter. Spec. ¶ 7. The Specification identifies examples of
parameter values that are used in generating the Bloom filter, such as “the
Appeal 2020-005414
Application 14/721,630

5
number of elements to be inserted in the Bloom filter, i.e., the number of
services which are to be advertised using the Bloom filter” (Spec. ¶ 58),
“false positive indication rate” (Spec. ¶ 59), “number of hash functions (and
possibly more particularly a specific set of hash functions)” (Spec. ¶ 60), and
“size (e.g., a number of bits) of the Bloom filter” (Spec. ¶ 60) (the latter two
are described as calculated based service number and false indication rate)
(Spec. ¶ 60).
Limitation [2] does not require that the data structure can be generated
using only the “plurality of parameter values” recited in limitation [1]
because there is no term in limitation [2] that would restrict the capability to
generate the data structure to solely the selected parameter values.
The radio and the processing element must be capable of selecting
parameter values and generating a probabilistic data structure, such as a
Bloom filter. Although the disclosure identified by Appellant as supporting
the various limitations of claim 1 describe specific wireless configurations
and structures (Appeal Br. 4), there does not appear to be a specific
identified structure that enables the radio and processing element to perform
the parameter value selection and data structure generation. Thus, for
example, a “processing element” with “program instructions” to accomplish
[1] and [2] would meet the claim limitations. See Spec. ¶¶ 29, 124, 129.
Limitations [3] and [4] of the claim recite the functional limitations of
to “wirelessly transmit” the data structure and parameter values,
respectively. A radio and processing element capable of wireless
transmission would meet these limitations in the claim because the claim is
not a method claim, and therefore only requires that these elements are
enabled to perform the recited function of wireless transmission.
Appeal 2020-005414
Application 14/721,630

6
Because the claim is directed to a “wireless device” having the
capability of performing the functions in the claim, we do not interpret [3]
and [4] as separate steps that would require [3] the wireless transmission of
the data structure and [4] “indication of one or more of the plurality of
parameter values” to be transmitted independently of each other or in
separate messages, fields, or transmissions. In fact, the Specification
discloses that “the number of services advertised in the Bloom filter and the
number of hash functions used to generate the Bloom filter,” namely the
parameter values, “may be indicated as part of ‘Bloom filter information’
provided in conjunction with the Bloom filter itself.” Spec. ¶ 63. In other
words, the indication of the parameter values may be provided as part of the
transmission of the data structure, itself. The Specification also describes
another embodiment in which the parameter values may be part of an
information field and therefore not a part of the Bloom filter. Spec. ¶ 83; see
also claim 20, which distinguishes [3] and [4] from being transmitted “in
conjunction” with each other from being transmitted “distinct” from each
other. Claim 1 recites that an “indication” of the parameter values is
transmitted, which is interpreted to mean that the value or something that
points to the value is transmitted.
The second part of limitation [4] recites “wherein a false-positive
probability associated with the probabilistic data structure can be determined
using the indication of one or more of the plurality of parameter values.”
This limitation does not require that the “false-positive probability” is
determined because it is not a step in a method. The claim recites “wherein”
the probability can be determined by the parameters. The “wherein” clause
is separate from the “configured to” clause. Consequently, the claim does
Appeal 2020-005414
Application 14/721,630

7
not require the radio/processing element to be capable of calculating the
false-positive probability using one of the parameters of the claim.

REJECTIONS
Appellant argued 14 separate groups of claims on a claim by claim
basis. Each of these groups is addressed below.

1. CLAIMS 1, 3, and 6
The Examiner found that Goering describes selecting parameter
values for a Bloom filter, generating the Bloom filter, and then wirelessly
transmitting it, which the Examiner found meets limitations [1]–[3] of claim
1. Final Act. 4. The Examiner found that Goering does not describe
transmitting the parameter values for the Bloom filter as in limitation [4] of
the claim, but found that Haddad does so. Id. at 5. The Examiner explained it
would have been obvious to transmit the parameter values to enable local
service discovery. Id. For determining [4] “a false-positive probability
associated with the probabilistic data structure . . . using the indication of
one or more of the plurality of parameter values,” the Examiner cited to
Yang, which the Examiner found discloses this limitation of the claim. Id. at
6.
Appellant does not dispute in the Appeal Brief that Goering describes
a wireless device capable of selecting parameter values that can be used in
generating a Bloom filter and a device that is also capable of generating it
and transmitting it wirelessly. Appellant, however, disputes the Examiner’s
finding that Haddad describes a wireless device that is capable of [4]
wirelessly transmitting “an indication of one or more of the plurality of
Appeal 2020-005414
Application 14/721,630

8
parameter values used to generate the probabilistic data structure.” Appeal
Br. 7–9. Appellant also argues that Yang does not remedy the deficiency in
Haddad and Goering, and does not describe the last part of limitation [4] of
“a false-positive probability associated with the probabilistic data structure
can be determined using the indication of one or more of the plurality of
parameter values.” Appeal Br. 8–9.

Limitation [4]
Limitation [4] requires a wireless device that is capable of wirelessly
transmitting “an indication of one or more of the plurality of parameter
values used to generate the probabilistic data structure” and “wherein” “a
false-positive probability associated with the probabilistic data structure”
can be determined “using the indication of one or more of the plurality of
parameter values.”

Haddad
We first address Haddad. The Examiner cited to Haddad at Figure 1,
2:3–6; 8:61–63; 10:45–47 (Final Act. 5); 14:35–37 (Ans. 4) as describing
and making the first part of limitation [4] obvious to “wirelessly transmit an
indication of one or more of the plurality of parameter values used to
generate the probabilistic data structure.” After reviewing these disclosures,
we conclude that Appellant is correct that the Examiner did not meet the
burden of establishing that Haddad describes transmitting the parameters
used to generate the Bloom filter.
The disclosures in Haddad cited by the Examiner are directed to
transmitting “identifiers,” which are elements which are used as inputs to the
Appeal 2020-005414
Application 14/721,630

9
Bloom filter, not parameters used to generate the filter. As indicated supra
at 5, the number of identifiers used would be considered a parameter.
Haddad discloses “[t]he pairing identifiers are inputs to the filter and the
output of the filter is a bit vector.” Haddad 7:28–29; see also 2:3–6. The “bit
vector” is the Bloom filter. Spec. ¶ 55 (“At least in some instances, the
Bloom filter may take the form of a bit vector (e.g., a bit array).”).
The Examiner states in the Answer that Haddad’s disclosure at 14:35–
37 of “transmitting to [a] peer node a second message that includes the
matched pairing identifier and the identifier indicated in the bit vector”
meets the claimed limitation because “each bit in the Bloom filter can be
considered a parameter for the data structure.” Ans. 4. The Examiner’s
finding is not supported by a preponderance of the evidence because the
identifier is an input used to construct the bits in the bit filter. Haddad 7:28–
29. Haddad explains that to make the bit vector “[a]ccording to some
aspects, a hash operation or multiple hash operations are performed on the
inputs (e.g., pairing identifiers) to derive the output bit vector.” Haddad
7:30–32. The hash values are the parameters, not the pairing identifiers. See
Spec. ¶ 60 (supra at 5). Therefore, the Examiner’s reliance on the wireless
transmission of pairing identifiers as a description of limitation [4] is
deficient.

Yang
Yang was cited by the Examiner for the second part of limitation [4]
of determining the false-positive probability. Yang discloses an equation to
compute a false positive rate for the Bloom filter, using at least some of the
parameter values to generate it, such as the hash function and the number of
Appeal 2020-005414
Application 14/721,630

10
keys used as input. Yang 7:53–61. (Appellant did not dispute that the false
positive rate is the same as the recited “false-positive probability”.) Yang
refers to calculating the false positive rate at 8:31–32 cited by the Examiner.
Final Act. 6. The Examiner further found that Yang describes “[u]sing the 6
bits of the index field 324 to calculate the false positive rate” at 8:31–32 as
meeting the claim limitation. Final Act. 6 (emphasis omitted). The index
field 324 is described as being a part of a “special information 300”
broadcast by a wireless device. Yang 5:58–64 (“the special information may
be encoded in a customized Information Element (IE) in the broadcasting
beacon. The special information 300 includes . . . an index field 324”).
Appellant states that the index field 324 indicates which of the Bloom
filters is used in an invitation list. Appeal Br. 9. Appellant argues that the
value of the index field 324 “does not allow for a determination of the false
probability ratio of the probabilistic data structure transmitted as in claim 1,”
but instead “the length of the index field 324” is used “to illustrate the
reduction in false positives associated with selecting a Bloom Filter from
multiple possible Bloom Filters as illustrated in Figure 7.” Id.
We agree with Appellant that what the Examiner identified in Yang as
a teaching of determining a false-positive probability associated with the
probabilistic data structure using the indication of one or more of the
plurality of parameter values” is deficient. Specifically, as indicated by
Appellant, the index field 324 identified by the Examiner is used to calculate
the false positive rate Pl of the outcome of a plurality of the Bloom filters
and does not use the index field to determine false positive rate Pf for an
individual filter. Yang 8:30–36.
Appeal 2020-005414
Application 14/721,630

11

New Ground of Rejection
Although we identified errors in the rejection of claim 1 as set forth
by the Examiner, the rejection, itself, is affirmable, but for different reasons
than those given by the Examiner. This is explained in more detail below.
Because our reasoning differs from the Examiners, we designate the
rejection as a new ground of rejection pursuant to 37 C.F.R. § 41.50(b). The
new ground of rejection is as follows:
Claim 1 is rejected under 35 U.S.C. § 103 as obvious in view of
Goering, Haddad, and Yang.
We leave it to the Examiner to determine whether dependent claims 3
and 6 are obvious based on the cited references.

Discussion
Before a claim can be compared to the prior art, it must be properly
interpreted to determine whether it is described or suggested by the prior art.
In this case, there was no explicit interpretation of claim 1. The Examiner
“read” claim limitations on the cited prior art references without interpreting
their full scope. As a result, the broadest reasonable interpretation of claim 1
was not taken into account when claim 1 was compared to the disclosures in
Goering, Haddad, and Yang. We do so now.
We focus attention on limitations [3] and [4]. As explained in the
Claim Interpretation section, we interpreted the term “configured for”
limitation of [3] and [4] to require that the radio and processing element of
the claimed wireless device are capable of transmitting the probabilistic data
structure and parameter values, respectively. Further, as explained supra at
Appeal 2020-005414
Application 14/721,630

12
6, the limitations do not require that the wireless transmission of the data
structure and “one or more of the plurality of parameter values” are
transmitted independently of each other or in separate messages, fields, or
transmissions. Consistently, the Specification discloses an embodiment in
which parameters are “provided in conjunction with the Bloom filter itself.”
Spec. ¶ 63. In other words, the parameter values may be provided as part of
the transmission of the data structure, itself. In the Reply Brief, Appellant
argued that “claim 1 explicitly requires that the wireless device transmits
both ‘the probabilistic data structure’ (e.g., a Bloom filter) and ‘an indication
of one or more of the plurality of parameter values used to generate the
probabilistic data structure’” which is true, but the two may be transmitted in
conjunction with each other, rather than in “distinct” transmissions (see
claim 20 reciting that the two are “distinct,” invoking the doctrine of claim
differentiation). Reply Br. 3.
We now turn to Goering. As the Examiner found, Goering discloses
that “[a] broadcast message will contain the Bloom filter that represents
services offered by the node itself.” Goering, 4.1. Overview. The filter
contains the number of bits as shown in Table 1 of Goering. “3.1 Overview
Bloom Filters.” Goering discloses that a Bloom filter “consists of an array of
w bits.” The bit number, w, is a parameter. See supra at 5. Therefore, when
the Bloom filter is broadcast, in conjunction with it, the number of bits in the
array is also broadcast.2 In view of the breadth of limitation [4] of claim 1,
we find that Goering’s disclosure that the wireless device is capable of

2 Although the number of bits in the array is not provided as a single
number, the number of bits could be counted and therefore this number
points to the actual value of the parameter and is an “indication” of it. See
supra at 6.
Appeal 2020-005414
Application 14/721,630

13
transmitting the filter which contains the bit number meets the first part of
limitation [4]. As explained above, the claim does not exclude the filter and
the parameter value from being transmitted in the same broadcast, namely in
conjunction with each other. There is no requirement in the claim that the
transmitting is done by separate steps.
The second part of limitation [4] of “a false-positive probability
associated with the probabilistic data structure can be determined using the
indication of one or more of the plurality of parameter values” is met by
Yang. There is no requirement in the claim that the probability is
determined. There is also no requirement that the device is actually capable
of determining the probability because the claim uses the terminology
“wherein” the probability “can be determined” with the parameter values,
which is separate from the “configured to” clause. See supra at 6. For these
reasons, Yang’s disclosure at 7:55 of the formula (P l=(1-e-nk/m)k) for
calculating the false positive rate that uses bits (m bit length) as one of the
values to plug into the equation meets the claim limitation. Indeed, even if
we interpret the claim as requiring that the wireless device be capable of
calculating the false-positive probability, the limitation is still met by Yang
because Yang’s device makes the calculation. Yang 7:53–61.
Appellant argued that Yang performs the calculation in the
transmitting device and not the receiving device (Appeal Br. 9–10, 13–14),
but the claim is not a method claim and does not distinguish between a
transmitting and receiving device.
Appeal 2020-005414
Application 14/721,630

14
2. CLAIM 2
Claim 2 recites “wherein the probabilistic data structure comprises a
Bloom filter, wherein the selected plurality of parameter values describe a
design and/or implementation of the Bloom filter.”
The Examiner referenced the use of the attenuated Bloom filters in
Goering as meeting the claim limitation. Final Act 6. The Examiner further
explained in the Answer that Goering discloses the bit array which is also a
parameter of the Bloom filter design. Ans. 5. In response, Appellant
contends that “such parameters do not teach the features of the claims at
least because these parameters are not transmitted in addition to the Bloom
filter or other probabilistic data structure.” Reply Br. 5. The deficiency in
Appellant’s argument is that the parameter values, as explained above, do
not have to be transmitted “in addition” to the Bloom filter. The rejection is
affirmed, but designated as a new ground of rejection because of its
dependency on claim 1.
3. CLAIM 5
Claim 5 depends from claim 1 and further recites: “wherein the radio
and the processing element are further configured to: wirelessly transmit
information directly advertising one or more services available via the
wireless device in addition to advertising services available via the wireless
device using the probabilistic data structure.”
The Examiner cites Santoro at 9:39–41 and 8:67–9:1 to meet the
claimed limitation. Final Act. 9. The Examiner found it would have been
obvious to one of ordinary skill in the art to include “direct advertising” as
described by Santoro to “efficiently identify target businesses.” Id.
(emphasis omitted).
Appeal 2020-005414
Application 14/721,630

15
Appellant argues that Santoro “concerns marketing techniques ‘to
efficiently identify target businesses to solicit for purchase of advertising
products’. In other words, Santoro relates to selling advertisements, a very
different subject than claim 5.” Appeal Br. 12.
The Examiner responded by newly citing Santoro at 5:44–46. Ans. 6.
Santoro describes targeting businesses “to solicit for purchase of
advertising products.” Santoro, Abstract. Santoro further explains:
One embodiment of the present invention provides a computer
implemented methodology for identifying target businesses to
solicit for purchase of advertising products.
Santoro 1:45–47.
A Bloom filter is used to store IDs of listings. Santoro 8:67–9:1.
Santoro at 5:43–46 (cited by Examiner) discloses: “In general, the client
103, network 101, and/or map server system 107 may be in communication
via any type of wired or wireless connection, using a wide variety of
communication protocols.”
The term “advertising” as used in the claim relates to transmitting or
broadcasting the service available on the wireless device and is not the same
as “advertising products” described by Santoro, which is advertising for an
actual product. See Santoro 3: 9–11.
Claim 5 requires the claim to be configured to transmit (1) services
available via the wireless device in addition to (2) services available via the
wireless device using the. In other words, not all the services advertised by
the device are transmitted in the Bloom filter.
Although Santoro may be interpreted to advertise via probabilistic
data structure, the Examiner also did not explain how the advertising is “in
Appeal 2020-005414
Application 14/721,630

16
addition to advertising services available via the wireless device using the
probabilistic data structure.” (emphasis added).
The rejection of claim 5 is reversed.

4. CLAIMS 7 and 14
Independent claim 7 is reproduced below (bracketed numbers added
to reference the two recited steps in the claim):
7. A method for discovering services advertised by a first
wireless device, comprising:
by a second wireless device:
[1] receiving information regarding services available via
the first wireless device,
wherein the information comprises:
a Bloom filter; and
an indication of one or more parameters of the
Bloom filter, wherein a false-positive probability of the Bloom
filter can be determined using the indication of one or more
parameters of the Bloom filter; and
[2] determining whether the Bloom filter indicates that at
least a first service is or is not available via the first wireless
device.

We interpret this claim similarly to claim 1. The claim recites
“receiving information regarding services available via the first wireless
device.” The information comprises the Bloom filter and the indication of
the parameters. There is no requirement in the claim that the Bloom filter
and the parameters are received in separate transmission, messages, or fields.
Consequently, as discussed above, we interpret the transmission of a Bloom
filter comprising bits to meet the claim limitation of receiving an indication
of parameters (the bit number is a parameter). The Examiner did not reject
Appeal 2020-005414
Application 14/721,630

17
the claim over Goering, but instead relied solely upon Haddad and Yang.
Final Act. 9–10.
The Examiner found that Yang receiving information services
available via the first wireless device, where the information comprises a
Bloom filter and the parameters of the Bloom filter that can be used to
determine false positive probability. Final Act. 9–10 (citing Yang at 8:14–
16; 8:31–32; Figs. 3, 7, 8; 3:58–60). The Examiner found that Yang does not
“detail” the recited parameter indication, but found that Haddad provides
such an indication, citing 10:45–47. Final Act. 10. The Examiner determined
it would have been obvious to one of ordinary skill in the art to combine
Yang and Haddad for automating service discovery and setting up a wireless
network. Id.
Appellant argues that Yang does not describe “receiving a Bloom
filter or ‘an indication of one or more parameters of the Bloom filter’” as
recited in claim 7. Appeal Br. 13. Appellant states that the disclosure in
Yang identified by the Examiner as describing the receipt of a Bloom filter
by a wireless device is a description of creating Bloom filters, not
transmitting them. Id.
We agree with Appellant that the Examiner did not identify disclosure
in Yang where a Bloom filter is received from by a second wireless device
from a first wireless device. The disclosure identified by the Examiner,
particularly at columns 7–8 and Figures 3 and 7 of Yang, relate to creating
Bloom filters; the Examiner did not guide us to a clear statement in Yang,
nor provide an adequate explanation to establish that Yang meets the
limitation of claim 7 of “receiving information regarding services available
via the first wireless device, wherein the information comprises: a Bloom
Appeal 2020-005414
Application 14/721,630

18
filter.” Although we find the Examiner’s findings and explanation deficient,
we are not making any determination as to whether Yang discloses the
disputed limitation in other parts of its disclosure.
Nonetheless, Goering cited in the rejection of claim 1, describes
receipt of a Bloom filter by a second wireless device from a first wireless
device. Goering 4.1 Overview (“A broadcast message will contain the
Bloom filter that represents services offered by the node itself and the d-1
attenuated Bloom filters that represent services from other nodes.”); 3.1
Overview Bloom Filters (“Although Bloom filters are already quite small, it
is possible to compress them before transmission.”). The transmission of a
Bloom filter, as discussed above, is reasonably interpreted to meet the claim
limitation of receiving an indication of at least one parameter.3
As for the requirement in claim 7 of “determining whether the Bloom
filter indicates that at least a first service is or is not available via the first
wireless device,” Goering also discloses this step:
Goering Abstract (“Attenuated Bloom filters are used to distribute
services to nodes in the neighborhood and thus enable local service
discovery.”);
Goering 4.1 Overview (“Nodes will announce the services they have
to offer through local broadcast messages, see section 4.2. All nodes will
keep information about what services can be reached through their direct
neighbors in attenuated Bloom filters. A node will store d attenuated Bloom

3 See also Goering for another transmitted parameter (4.6 Hash function:
“Nodes that want to exchange service-broadcasts should use the same set of
hash functions that have to be defined beforehand.”). Defining a function
beforehand reasonably suggests that the hash function is transmitted to the
wireless device.
Appeal 2020-005414
Application 14/721,630

19
filters per node it receives a broadcast from, which represent the number of
hops away the service might be located.”).
Goering therefore describes all the limitations of claim 7, anticipating
it. Claim 7 is rejected under 35 U.S.C. § 102(a)(1) as anticipated by Goering.
We designate this as a new ground of rejection pursuant to 37 C.F.R.
§ 41.50(b).
We leave it to the Examiner to determine whether claim 14 is
anticipated by Goering, or obvious based on Goering or any additional cited
references.
5. CLAIMS 8, 9, and 21
Claim 8 depends from claim 7 and recites:
determining the false-positive probability of the Bloom
filter; and
if the Bloom filter indicates that the first service is
available and the false-positive probability of the Bloom filter is
below a false-positive threshold, sending a generic
advertisement service (GAS) query to the first wireless device
regarding the first service.

The Examiner cited Yang as describing “the false-positive probability
of the Bloom filter is below a false-positive threshold,” citing to Figure 7,
item 704 of “USE SAME BLOOM FILTER TO CHECK FOR FALSE
POSITIVES.” Final Act. 16. The Examiner further describes Stephenson as
“sending a generic advertisement service (GAS) query to the first wireless
device regarding the first service.” Id. at 17 (emphasis omitted).
The Examiner determines that the limitation beginning with “if” is
conditional and therefore optional and not given patentable weight. Ans. 8.
Appellant argues, in response, that the limitation is not conditional.
Reply Br. 6. Appellant also contends that Yang step 704 “describes actions
Appeal 2020-005414
Application 14/721,630

20
of the transmitting device, and is thus not applicable to the second, e.g.,
receiving, device as in claims 8, 9, and 21.” Id. Appellant further argues
“there is no hint in 704, Col. 8, lines 14-36, or elsewhere in Yang of
comparing the false positive probability to a ‘threshold’ at all.” Id.
The claim requires an indication that “the first service is available”
and that “the false-positive probability of the Bloom filter is below a false-
positive threshold.” When both are satisfied, then the claim requires
“sending a generic advertisement service (GAS) query to the first wireless
device regarding the first service.” The claim therefore requires that the
threshold is made. The sending of the GAS query, however, is conditional
because it doesn’t take place unless “the first service is available” and when
“the false-positive probability of the Bloom filter is below a false-positive
threshold” are satisfied. The latter two conditions therefore must be
performed to meet the claim limitation.
As argued by Appellant, the Examiner did not identify a description in
Yang of determining “if . . . the false-positive probability of the Bloom filter
is below a false-positive threshold.” Yang does not describe, at the
disclosure cited by the Examiner, that checking for false-positives is done by
determining that “the false-positive probability of the Bloom filter is below a
false-positive threshold.” Consequently, we reverse the obviousness
rejection of claim 8, and of claim 9, which depends from claim 8.
Claim 21 depends from claim 17, and has similar language about
using a threshold. We also reverse the rejection for similar reasons as for
claim 8 because the Examiner did not show that Yang is capable of making
the recited threshold determination.
The rejection of claims 8, 9, and 21 is reversed.
Appeal 2020-005414
Application 14/721,630

21
6. CLAIMS 10 and 19
Claim 10 depends from claim 7, and recites “wherein the information
regarding services available via the first wireless device further comprises
information directly indicating availability of one or more services via the
first wireless device.” Claim 19 depends from claim 15 and has the same
limitation.
The purpose of Goering is to determine service availability between
interconnected nodes of wireless networks by using Bloom filters. The
following citations provide evidence of this fact:
Goering Title (“Service Discovery Using Bloom Filters”).
Goering Abstract (“A protocol to perform service discovery in
ad-hoc networks is introduced in this paper. Attenuated Bloom filters
are used to distribute services to nodes in the neighborhood and thus
enable local service discovery. . . . Methods to optimize the used
bandwidth, which is a scarce resource in wireless networks, are
investigated.”).
Goering 1 Introduction (“In the near future many people will
carry multiple devices like a phone or a PDA with connected
peripherals. People want to be able to find and connect these devices
together as they may need information or services from one of their
devices. Further people want to be able to find devices of other users
or organizations in the vicinity that offer services to the public that
they can make use of.”).

The Bloom filter is used to determine what services are available:
Appeal 2020-005414
Application 14/721,630

22
Goering 3.1 Overview Bloom Filters (“In the example Bloom
filter given in Table 1 several services are represented. When a user
wants to test whether a color printer is one of these services, a hash
function will be used on the string ‘Color Printer’. Suppose this hash
function returns (0, 3, 6). This means the color printer is probably
represented in this filter as the bits 0, 3 and 6 are all enabled.”).
Goering therefore describes all the limitations of claims 10 and 19, as
well as the limitations of their respective base claims (i.e., claims 7 and 15),
discussed herein. Accordingly, we find Goering anticipates claims 10 and
19.
Claims 10 and 19 are rejected under 35 U.S.C. § 102 as anticipated by
Goering. We designate this as a new ground of rejection pursuant to
37 C.F.R. § 41.50(b).
7. CLAIM 11
Claim 11 depends from claim 7, and further recites “wherein the
indication of one or more parameters includes an indication of a number of
services advertised using the Bloom filter and a set of hash functions used
with the Bloom filter.”
Claim 7 is a method claim. In combination with the limitations of
claim 7, it requires a second wireless device receiving from a first wireless
device information comprising: (1) a Bloom filter, (2) an indication of one or
more parameters of the Bloom filter, and (3) “an indication of a number of
services advertised using the Bloom filter,” and (4) “a set of hash functions
used with the Bloom filter.” All four must be received by the second
wireless device.
Appeal 2020-005414
Application 14/721,630

23
The Examiner cites disclosure from Yang of listing services provided
by a computing device (at 10:3–4), of a Bloom filter with different hash
functions (at 8:14–15), and of the transmission of service bits (10:47–49).
Final Act. 11; Ans. 12.
We agree with Appellant’s argument that the Examiner erred. Appeal
Br. 16–17. There is no discussion in Yang of transmitting “a set of hash
functions used with the Bloom filter.” The Examiner identifies disclosure in
Yang where a Bloom filter “with different k hash functions . . . is used to
generate the invitation list in the invitation list field.” Yang 8:14–17. The
invitation list field is broadcast. Yang 11: 26–32. The Examiner did not
identify a teaching in Yang where the hash functions are broadcast.
However Goering teaches that a hash function must be used on the
string represented by the Bloom filter:
Goering 3.1 Overview Bloom Filters (“When a user wants to test
whether a color printer is one of these services, a hash function will be
used on the string ‘Color Printer’.”).
Goering also discloses an embodiment where the same hash function
is used for the sending and receiving node and the functions are defined
beforehand:
Goering 4.6 Hash function (“Nodes that want to exchange service-
broadcasts should use the same set of hash functions that have to be
defined beforehand.”).
Because both wireless device nodes use the same hash function in the
above-mentioned embodiment, it would have been obvious to one of
ordinary skill in the art to provide the hash function with the Bloom filter for
efficiency and because the time at which the hash functions are transmitted
Appeal 2020-005414
Application 14/721,630

24
is not critical as long as they are received when the second wireless device
needs to perform service discovery.
With regard to the number of services advertised using the Bloom
filter, we find that the number of services is indicated by, or represented by
the Bloom filter, and therefore is inherent. The claim does not require
recognition of the number of services or its use in performing additional
operations.
We have already addressed the transmission of the Bloom filter and
the “bits” as parameters. Thus, all four elements are described by Goering as
being received by a second wireless device.
Goering therefore describes all the limitations of claim 11, as well as
the limitations of base claim 7,discussed above, anticipating it. Claim 11 is
rejected under 35 U.S.C. § 102(a)(1) as anticipated by Goering. We
designate this as a new ground of rejection pursuant to 37 C.F.R. § 41.50(b).

8. CLAIM 12
Claim 12 depends from claim 11, and further comprises the second
device “calculating the false-positive probability of the Bloom filter based
on a number of hash functions used with the Bloom filter.”
The Examiner cited disclosure in Yang of calculating the false
positive rate. Final Act. 12. However, as we discussed earlier, Yang does not
calculate the false positive rate after transmitting the Bloom filter, but does it
for the purpose of determining the error rate in preparing an invitation list
(Yang 8:30–36) that is subsequently broadcast (Yang 11:27–32).
Appeal 2020-005414
Application 14/721,630

25
Goering does not disclose “calculating the false-positive probability of
the Bloom filter based on a number of hash functions used with the Bloom
filter” after its transmission to a second wireless device.
Having determined that the Examiner erred in finding the limitation
described by Yang and finding Goering deficient, as well, we reverse the
rejection of claim 12.
9. CLAIM 13
Claim 13 depends from claim 12 and further recites “calculating a
length of the Bloom filter based on the false-positive probability of the
Bloom filter and the number of services advertised using the Bloom filter.”
The Examiner cited Yang’s determination of the probability of false-
positive rate of a Bloom filter at 7–8. Ans. 13–14. Yang teaches that a false
positive rate can be calculated, but it uses the knowledge of the Bloom filter,
such as the number of elements. Yang 7:53–58 (“The false positive rate P1
of a Bloom Filter 1 can be computed using [formula]. Therefore, given that
the invitation list is nine bytes (72 bits), a Bloom Filter with 4 hash functions
would have a false positive rate around 20% if the number of elements n
equals to 20.”). The Examiner did not explain how the length of the filter is
calculated “based on the false-positive probability of the Bloom filter and
the number of services advertised using the Bloom filter,” as recited in the
claim.
The rejection of claim 13 is reversed for the reasons set forth for claim
12 and because the Examiner did not explain how Yang meets the additional
limitation of claim 13.
Appeal 2020-005414
Application 14/721,630

26
10. CLAIM 15
Independent claim 15 is reproduced below (annotations added as
bracketed numbers):
15. A second wireless device, comprising:
a radio; and
a processor operatively coupled to the radio;
wherein the radio and the processor are configured to:
[1] receive information regarding services available via a
first wireless device, wherein the information includes:
a probabilistic data structure; and
an indication of one or more parameters used to
generate the probabilistic data structure; and
[2] determine whether the probabilistic data structure
provides a positive or negative indication of the availability via
the first wireless device of each of one or more services.

Claim 15 is directed to a wireless device comprising a radio and a
processor operatively coupled to the radio.
The radio and processor are configured to [1] receive information via
a first wireless device. The information includes (1) a probabilistic data
structure; and (2) an indication of one or more parameters used to generate
the probabilistic data structure.
The radio and processor are also configured to [2] “determine whether
the probabilistic data structure provides a positive or negative indication of
the availability via the first wireless device of each of one or more services.”
We interpret this claim as we did claim 1. The radio and processor of
the wireless device must be capable of receiving the probability data
structure and one of the parameters used to generate it. The latter two
elements are not required by the claim to be in separate messages,
broadcasts, fields, etc., because the claim has no restriction on the form of
Appeal 2020-005414
Application 14/721,630

27
the information that is received. Therefore, as for claim 1, the parameters
can be provided as part of the transmission of the data structure, itself.
We interpret the “probabilistic data structure” to include a Bloom
filter. Thus, a configured radio/processor that can receive a Bloom filter
meets the limitation of the claim. We have the same interpretation of
“parameters” as “parameter values” as in claim 1.
The radio and processor of the wireless device must also be capable of
[2] determining “whether the probabilistic data structure provides a positive
or negative indication of the availability via the first wireless device of each
of one or more services.” We interpret this to mean that data structure is
configured to determine whether a service is available via the device.
The claim is not a method claim. To meet the claim limitation, it must
show that the device is capable of performing the functions recited in the
claim, not they necessarily have been performed.
The Examiner rejects the claim based on Yang and Haddad. Final
Act. 9. The Examiner only cites Yang as meeting the limitations of the
claim. We reverse the rejection because Goering is closer prior art than Yang
as it describes the basic functions of a Bloom filter in broadcasting services
to a wireless node. We set forth a new ground of rejection pursuant
37 C.F.R. § 41.50(b) as follows.
Goering discloses that the Bloom filter is transmitted to a node and the
node therefore receives it, providing evidence that the radio/processor is
configured to/capable of “receive information regarding services available.”
Goering 4.1 Overview (“A broadcast message will contain the Bloom filter
that represents services offered by the node itself and the d-1 attenuated
Bloom filters that represent services from other nodes.”); 3.1 Overview
Appeal 2020-005414
Application 14/721,630

28
Bloom Filters (“Although Bloom filters are already quite small, it is possible
to compress them before transmission.”).
The Bloom filter is a “probabilistic data structure” and the data
structure/Bloom filter comprises bits, which as explained for claim 1, supra
at 26, is a parameter. We refer to the discussion in claim 1 for this limitation
of the claim, rather than repeating it here. Thus, the first part [1] of claim 15
is met by Goering.
The claim also requires that the radio/processor is capable to [2]
“determine whether the probabilistic data structure provides a positive or
negative indication of the availability via the first wireless device of each of
one or more services.” We interpreted this to mean that data structure/Bloom
filter is configured to determine whether a service is offered by the device.
As explained below, this limitation is met by Goering.
The purpose of Goering is to determine service availability between
interconnected nodes of wireless networks by using Bloom filters. See
Goering Title, Abstract, and 1 Introduction.
The Bloom filter is used to determine what services are available.
Goering 3.1 Overview Bloom Filters (“In the example Bloom filter given in
Table 1 several services are represented. When a user wants to test whether a
color printer is one of these services, a hash function will be used on the
string ‘Color Printer’. Suppose this hash function returns (0, 3, 6). This
means the color printer is probably represented in this filter as the bits 0, 3
and 6 are all enabled.”).
Determining what services are available using the Bloom filter
accomplishes the last part of claim 15 of being capable to “determine
whether the probabilistic data structure provides a positive or negative
Appeal 2020-005414
Application 14/721,630

29
indication of the availability via the first wireless device of each of one or
more services.” We do not discern any difference between discovering
services available via a Bloom filter received from a broadcasting node and
this last limitation of the claim.
Goering therefore describes all the limitations of claim 15,
anticipating it. Claim 15 is rejected under 35 U.S.C. § 102(a)(1) as
anticipated by Goering. We designate this as a new ground of rejection
pursuant to 37 C.F.R. § 41.50(b).

11. CLAIM 16
Claim 16 depends from claim 15 and further comprises “determine
that a first service is not available via the first wireless device if the
probabilistic data structure provides a negative indication of the availability
via the first wireless device of the first service.” We interpret this to mean
that when the Bloom filter is used to determine whether a service is offered,
and a negative is obtained, it is determined that a first service is not available
via the first wireless device.
We agree with Appellant that Yang is deficient. Appeal Br. 18–19.
However, determining whether a service is not available and therefore
returning a “negative” is a part of discovering whether or not services are
available via the first wireless node/device. (Indeed, it is known that false
negatives cannot occur with a Bloom filter. Haddad 8:5–6; Spec. ¶ 32.)
Accordingly, we find this limitation met by Goering.
In addition to disclosing all the limitations of base claim 15 (see
above), Goering also describes all the limitations of claim 16, anticipating it.
Claim 16 is rejected under 35 U.S.C. § 102(a)(1) as anticipated by Goering.
Appeal 2020-005414
Application 14/721,630

30
We designate this as a new ground of rejection pursuant to 37 C.F.R. §
41.50(b).

12. CLAIM 17
Claim 17 depends from claim 15 and further recites “wherein the
radio and the processor are further configured to: determine a false-positive
probability of the probabilistic data structure based at least in part on the
indication of one or more parameters used to generate the probabilistic data
structure.”
There is no requirement in the claim that probability is determined.
However, because the radio/processor are “configured to” determine the
false-positive probability, we interpret the claim to require that the
radio/processor are capable of making the determination.
The Examiner found this limitation met by Yang. Ans. 18. We agree
with the Examiner that Yang’s disclosure at 7:55 of the formula (P l=(1-e-
nk/m)k) for calculating the false positive rate at 7:55 which uses bits (m bit
length) as one of the values to plug into the equation establishes that the
device is capable of making the recited determination. One of ordinary skill
in the art would have reason to perform the calculation in Yang because of
the importance of false-positives in service discovery. Goering 3 Service
Discovery Using Attenuated Filter.
Appellant’s argument focuses on the field 324 and the device which
performed the false-probability determination. Appeal Br. 19. However, the
claim is not directed to a method and which device performs the steps is
irrelevant.
Appeal 2020-005414
Application 14/721,630

31
Pursuant to 37 C.F.R. § 41.50(b), a new ground of rejection of claim
17 under 35 U.S.C. § 103 as obvious in view of Goering and Yang.

13. CLAIM 18
Claim 18 depends from claim 15 and further recites “wherein the
radio and the processor are further configured to: calculate one or more
additional parameters of the probabilistic data structure based at least in part
on the indication of one or more parameters used to generate the
probabilistic data structure.”
Claim 18 requires the device to be capable of calculating additional
parameters based on at least one of the parameters used to generate the
probabilistic data structure.
The Examiner cites to Yang 7:36–60 as meeting the limitation and
Haddad 10:45–47. Ans. 19.
The disclosure cited by the Examiner in Yang describes the creating
the Bloom filter and then using the same hash function to determine whether
a key is contained in it:
The Bloom Filter represents n elements (also called keys) in a
bit vector to support membership check-up. The basic operation
of the Bloom Filter is to compute k hash values given a key X
(e.g., the Host Name). The k hash values range from 1 to m.
The filter then sets k bits in a bit-vector with m bit length at the
bit positions corresponding to the k hash values. This repeats
until all the remaining n-1 keys are set into the bit vector. The
Bloom filter verifies whether a key X' has been input into the
bit vector by using the same k hash functions to generate k hash
values. If all the bits at the k locations corresponding to the k
hash values in the bit-vector are set, X' is believed to have been
input into the Bloom Filter.
Yang 7:40–53.
Appeal 2020-005414
Application 14/721,630

32
The hash function, which is a parameter of the Bloom filter, is used to
generate hash values and to verify whether X’ is contained within the filter
at k locations. The hash values are a parameter of the Bloom filter and thus a
parameter (hash function) is used to calculate additional parameters (hash
value, k location). See Spec. ¶ 7. Appellant’s arguments did not address this
specific description in Yang. Appeal Br. 20.
Pursuant to 37 C.F.R. § 41.50(b), a new ground of rejection is set
forth of claim 18 under 35 U.S.C. § 103 as obvious in view of Goering and
Yang.
14. CLAIM 20
Claim 20 depends from claim 15 and recites “wherein the
probabilistic data structure comprises a Bloom filter, wherein the indication
of one or more parameters used to generate the probabilistic data structure is
distinct from the Bloom filter.”
We interpret “distinct” to indicate that the parameters are not a part of
the Bloom filter, as we found in claims 1 and 15, but instead are separate, for
example in separate field or transmission. This embodiment is described in
the Specification paragraph 83 (“The IE [information element] may also
include a Bloom filter information field 606, which as previously noted may
include certain information relating to the Bloom filter. For example, as
shown, the Bloom filter information field may include a sub-field 608 for
indicating the number of services offered by the wireless device.”).
As discussed above, Goering discloses that “[n]odes that want to
exchange service-broadcasts should use the same set of hash functions that
have to be defined beforehand.” Goering 4.6 Hash function. This disclosure
establishes that in one embodiment a parameter used to generate the
Appeal 2020-005414
Application 14/721,630

33
probabilistic data structure is provided beforehand to the device and
therefore the device is configured to receive it as a separate indication.
Goering therefore describes all the limitations of claim 20 (as well as
the limitations of base claim 15, as discussed above), anticipating it. Claim
20 is rejected under 35 U.S.C. § 102(a)(1) as anticipated by Goering. We
designate this as a new ground of rejection pursuant to 37 C.F.R. § 41.50(b).

CONCLUSION
In summary:
Claim(s)
Rejected
35
U.S.C.
§
Reference(s)/Basis Affirmed Reversed New
Ground
1–3, 6 103 Goering, Haddad,
Yang
1–3, 6 1, 2
5 103 Goering, Haddad,
Yang, Santoro
5
7, 10–20 103 Yang, Haddad 7, 10–20
8, 9, 21 103 Yang, Haddad,
Stephenson
8, 9, 21
7, 10, 11,
15, 16,
19, 20
102 Goering 7, 10,
11, 15,
16, 19,
20
17, 18 103 Goering, Yang 17, 18
Overall
Outcome
1–3, 5–21 1, 2, 7,
10, 11,
15–20

TIME PERIOD FOR RESPONSE
This decision contains a new ground of rejection pursuant to
37 C.F.R. § 41.50(b). 37 C.F.R. § 41.50(b) provides “[a] new ground of
Appeal 2020-005414
Application 14/721,630

34
rejection pursuant to this paragraph shall not be considered final for judicial
review.”
37 C.F.R. § 41.50(b) also provides that the Appellant, WITHIN TWO
MONTHS FROM THE DATE OF THE DECISION, must exercise one of
the following two options with respect to the new ground of rejection to
avoid termination of the appeal as to the rejected claims:
(1) Reopen prosecution. Submit an appropriate
amendment of the claims so rejected or new Evidence relating
to the claims so rejected, or both, and have the matter
reconsidered by the examiner, in which event the proceeding
will be remanded to the examiner. . . .
(2) Request rehearing. Request that the proceeding be
reheard under § 41.52 by the Board upon the same Record. . . .

Further guidance on responding to a new ground of rejection can be
found in the Manual of Patent Examining Procedure § 1214.01.
REVERSED; 37 C.F.R. § 41.50(b)