Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This action is in response to the amendment filed 5/22/2026. Claims 1, 2, and 4-20 are pending. Claims 1 (a machine), 8 (a machine), and 14 (a machine) are independent.
Response to Arguments
Applicant's arguments filed 5/22/2026 have been fully considered but they are not persuasive.
On page 8 of the response filed 5/22/2026, Applicant argues that Chowdhury “EIFFeL: Ensuring Integrity for Federated Learning” does not disclose: “wherein the one or more parameters of the privacy-preserving configuration comprise at least one of a random number generator (RNG) or a secret-shared non-interactive proof (SNIP) configuration to use for splitting the UE data into the plurality of cryptographically secure input shares”.
This argument is not persuasive. Chowdhury explicitly discloses the use of SNIP (§ 1) and splitting UE data (§ 4.4).
Applicant’s assertion that the SNIP parameters of Chowdhury are distinct from those claimed is not persuasive. The claim sets forth generic parameters and generic use of the SNIP algorithm and Chowdhury discloses SNIP and parameters used therein to split UE data. Therefore, Chowdhury reasonably anticipates the limitations of claim 14.
The amendments to the other independent claims are similar to those of claim 14 and Applicant’s arguments are not persuasive for the reasons discussed above.
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 14 and 16-18 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Chowdhury et al., “EIFFeL: Ensuring Integrity for Federated Learning” (published 2022).
As to claim 14, Chowdhury discloses a machine comprising:
One or more processors configured to, when executing instructions stored in a memory, perform operations comprising: (Chowdhury § 7.1 Performance Evaluation)
receiving an indication of a privacy-preserving configuration for secure data aggregation, the privacy-preserving configuration comprising one or more parameters to use for converting a cryptographically secure input share containing user equipment (UE) data to an aggregated output share; (“Broadcast. The server broadcasts the current parameters of the model M to all the clients.” Chowdhury § 2.1. “Setup Phase. In the setup phase, all parties are initialized with the system-wide parameters, namely the security parameter 𝜅, the numuber of clients 𝑛 out of which only 𝑚 < ⌊ 𝑛−1 3 ⌋ can be malicious, public parameters for the key agreement protocol 𝑝𝑝 $←− KA.param(𝜅), and a field F where |F| ≥ 2 𝜅 .” Chowdhury § 4.4) wherein the one or more parameters of the privacy-preserving configuration include at least one of a random number generator (RNG) or a secret-shared non-interactive proof (SNIP) configuration to use for splitting the UE data into the plurality of cryptographically secure input shares; (“EIFFeL uses secret-shared non-interactive proofs (SNIP; [24]) which are a type of zero-knowledge proofs that are optimized for the client server setting.” Chowdhury § 1. See § 4.4 discussing the setups of the SNIP protocol as modified by EIFFeL. “Setup Phase. In the setup phase, all parties are initialized with the system-wide parameters, namely the security parameter 𝜅, the numuber of clients 𝑛 out of which only 𝑚 < ⌊ 𝑛−1 3 ⌋ can be malicious, public parameters for the key agreement protocol 𝑝𝑝 $←− KA.param(𝜅), and a field F where |F| ≥ 2 𝜅 .” Chowdhury § 4.4. see also the plurality of mentions of SNIP throughout Chowdhury.)
receiving the cryptographically secure input share containing the UE data; and (“Round 3 (Verify Proof). In this round, every client C𝑖 partakes in the verification of the proofs 𝜋𝑗 of all other clients C𝑗 ∈ C\𝑖 , under the supervision of the server S.” Chowdhury § 4.4)
converting the cryptographically secure input share to the aggregated output share in accordance with the one or more parameters of the privacy-preserving configuration for secure data aggregation. (“every client C𝑖 ∈ C \ C∗ generates its share of the aggregate, (𝑖, U𝑖) with U𝑖 = Í C𝑗 ∈C\C∗ 𝑢𝑗𝑖, and sends that share to the server S.” Chowdhury § 4.4)
As to claim 16 Chowdhury discloses the machine of claim 14 and further discloses:
receiving a secret-shared non-interactive proof (SNIP) (“EIFFeL ensures input privacy by using Shamir’s threshold secret sharing scheme [60] (Sec. 4.1). Input integrity is guaranteed via SNIP and verifiable secret shares (VSS) which validates the correctness of the secret shares (Sec. 4.1)” Chowdhury § 2.4) associated with the cryptographically secure input share; and (“Round 2 (Generate and Distribute Proofs). Every client generates shares of its private update 𝑢𝑖 and the proof 𝜋𝑖 , and distributes these shares to the other clients C” Chowdhury § 4.4)
prior to converting the cryptographically secure input share to the aggregated output share, (“Round 4 (Compute Aggregate). This is the final round of EIFFeL where the aggregate of the well-formed updates is comp” Chowdhury § 4.4) verifying the cryptographically secure input share based at least on the SNIP associated with the cryptographically secure input share. (“Round 3 (Verify Proof). In this round, every client C𝑖 partakes in the verification of the proofs” Chowdhury § 4.4).
As to claim 17 Chowdhury discloses the machine of claim 16 and further discloses:
wherein verifying the cryptographically secure input share comprises exchanging the SNIP with one or more application functions (AFs) that are configured to receive and process other cryptographically secure input shares containing the UE data. (“Round 3 (Verify Proof). In this round, every client C𝑖 partakes in the verification of the proofs” Chowdhury § 4.4. The clients being the application functions).
As to claim 18 Chowdhury discloses the machine of claim 14 and further discloses:
18. The one or more processors of claim 14, the operations further comprising registering (“To prevent the server from simulating an arbitrary number of clients, the clients register themselves with a specific user ID on the public bulletin board B and are authenticated with the help of standard public key infrastructure (PKI).” Chowdhury § 4.4) for one or more data collection, model training, (use of the bulletin board for federated learning throughout § 4.4)
.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1, 4-10, 12, 13, 15, 19, and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chowdhury et al., “EIFFeL: Ensuring Integrity for Federated Learning” (published 2022), in view of Mueck et al., US 2024/0298194 (filed 2022).
As to claim 1, Chowdhury discloses a machine comprising:
One or more processors configured to, when executing instructions stored in a memory, perform operations comprising: (Chowdhury § 7.1 Performance Evaluation)
receiving an indication of a privacy-preserving configuration for secure data aggregation, the privacy-preserving configuration comprising one or more parameters to use for splitting user equipment (UE) data into a plurality of cryptographically secure input shares; (“Broadcast. The server broadcasts the current parameters of the model M to all the clients.” Chowdhury § 2.1. “Setup Phase. In the setup phase, all parties are initialized with the system-wide parameters, namely the security parameter 𝜅, the number of clients 𝑛 out of which only 𝑚 < ⌊ 𝑛−1 3 ⌋ can be malicious, public parameters for the key agreement protocol 𝑝𝑝 $←− KA.param(𝜅), and a field F where |F| ≥ 2 𝜅 .” Chowdhury § 4.4)
wherein the one or more parameters of the privacy-preserving configuration comprise at least one of a random number generator (RNG) or a secret-shared non-interactive proof (SNIP) configuration to use for splitting the UE data into the plurality of cryptographically secure input shares (“EIFFeL uses secret-shared non-interactive proofs (SNIP; [24]) which are a type of zero-knowledge proofs that are optimized for the client server setting.” Chowdhury § 1. See § 4.4 discussing the setups of the SNIP protocol as modified by EIFFeL. “Setup Phase. In the setup phase, all parties are initialized with the system-wide parameters, namely the security parameter 𝜅, the numuber of clients 𝑛 out of which only 𝑚 < ⌊ 𝑛−1 3 ⌋ can be malicious, public parameters for the key agreement protocol 𝑝𝑝 $←− KA.param(𝜅), and a field F where |F| ≥ 2 𝜅 .” Chowdhury § 4.4. see also the plurality of mentions of SNIP throughout Chowdhury.)
generating the plurality of cryptographically secure input shares according to the one or more parameters of the privacy-preserving configuration; and (“Round 2 (Generate and Distribute Proofs). Every client generates shares of its private update 𝑢𝑖 and the proof 𝜋𝑖 , and distributes these shares to the other clients C” Chowdhury § 4.4)
Chowdhury does not disclose the term: “instructing radio frequency (RF) circuitry to”
Mueck discloses an architecture for analytics with wireless devices, see e.g. Fig. 11. And radio network elements noted in Mueck ¶¶ 197 and 211.
A person of ordinary skill in the art before the effective filing date of the claimed invention would have combined Chowdhury with Mueck by implementing the disclosure of Chowdhury in a mobile network including wireless devices as the clients of Chowdhury. It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine Chowdhury with Mueck in order to allow a federated learning model to be exploited by UEs that contribute updates to expand local models for high-accuracy inference, Mueck ¶ 22; thereby providing ML services in ubiquitous wireless telephony.
As to claim 4, Chowdhury in view of Mueck discloses the machine of claim 1 and further discloses:
the operations further comprising: receiving, during a downlink protocol data unit (PDU) session, (“transport level packet marking in the uplink and downlink,” Mueck ¶ 211. Radio interface to the UE)
data associated with a machine learning (ML) model; and (“Broadcast. The server broadcasts the current parameters of the model M to all the clients.” Chowdhury § 2.1.)
training the ML model with the UE data in accordance with the one or more parameters of the privacy-preserving configuration for secure data aggregation. (“Server. There is a single untrusted server, S, who coordinates
the updates from different clients to trainM.
A single training iteration in FL consists of the following steps:
• Broadcast. The server broadcasts the current parameters of the
modelM to all the clients.
• Local computation. Each client C𝑖 locally computes an update,
𝑢𝑖 , on its dataset 𝐷𝑖 .” Chowdhury § 2.1.)
As to claim 5, Chowdhury in view of Mueck discloses the machine of claim 4 and further discloses:
the operations further comprising instructing the RF circuitry to transmit, during an uplink PDU session, (“transport level packet marking in the uplink and downlink,” Mueck ¶ 211. Radio interface to the UE)
data associated with the ML model trained with the UE data in accordance with the one or more parameters of the privacy-preserving configuration for secure data aggregation. (“Aggregation. The server S collects the client updates and aggregates them.” Chowdhury § 2.1. see also Round 2 in § 4.4)
As to claim 6, Chowdhury in view of Mueck discloses the machine of claim 1 and further discloses:
wherein instructing the RF circuitry to transmit the plurality of cryptographically secure input shares (“see e.g. Fig. 11. And radio network elements noted in Mueck ¶¶ 197 and 211.”) comprises outputting the plurality of cryptographically secure input shares for transmission to a respective plurality of application functions (AFs) that are configured to convert the plurality of cryptographically secure input shares to a respective plurality of aggregated output shares. (“every client C𝑖 ∈ C \ C∗ generates its share of the aggregate, (𝑖, U𝑖) with U𝑖 = Í C𝑗 ∈C\C∗ 𝑢𝑗𝑖, and sends that share to the server S.” Chowdhury § 4.4. the clients being the application function.)
As to claim 7, Chowdhury in view of Mueck discloses the machine of claim 1 and further discloses:
… that is configured to generate the analytic data based at least on the plurality of cryptographically secure input shares containing the UE data.
(“The server S can choose a different Valid(·) for every iteration (the protocol described above corresponds to a single iteration of model training in FL).” Chowdhury § 4.4. “The server broadcasts the current parameters of the Model M to all the clients” Chowdhury § 2.1)
Chowdhury in view of Mueck as combined in claim 1 does not explicitly disclose:
outputting a request for analytic data; and
receiving the analytic data from a network data analytics function (NWDAF)
Mueck further discloses:
outputting a request for analytic data; and (“Data types to be provided by the AIS/selected AI agent to AIS consumer (e.g., UE)” Mueck ¶ 77. Also ¶¶ 121, 154. “A UE (AIS consumer) to communicate to the AIS information relating to a user/client application-specific task (e.g., intention to drive a vehicle from location A to location B” Mueck ¶ 44)
receiving the analytic data from a network data analytics function (NWDAF) (“Per 3GPP Technical Standard 23.501, the Network Exposure Function (NEF) supports a number of functionalities, among which are the following (NWDAF stands for Network Data Analytics Function): …. (b) Exposure of analytics: NWDAF analytics may be securely exposed by NEF for external party … (c) Retrieval of data from external party by NWDAF: Data provided by the external party may be collected by NWDAF” Mueck ¶¶ 33-36).
A person of ordinary skill in the art before the effective filing date of the claimed invention would have combined Chowdhury with Mueck by implementing the disclosure of Chowdhury in a mobile network including the elements of the AIS in the existing NWDAF network element. It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine Chowdhury with Mueck in order to allow a federated learning model to be exploited by UEs that contribute updates to expand local models for high-accuracy inference, Mueck ¶ 22; thereby providing ML services in ubiquitous wireless telephony.
As to claim 8, Chowdhury discloses a machine comprising:
when executing instructions stored in a memory, perform operations comprising: (Chowdhury § 7.1 Performance Evaluation)
𝜅, the numuber of clients 𝑛 out of which only 𝑚 < ⌊ 𝑛−1 3 ⌋ can be malicious, public parameters for the key agreement protocol 𝑝𝑝 $←− KA.param(𝜅), and a field F where |F| ≥ 2 𝜅 .” Chowdhury § 4.4)
receiving a plurality of aggregated output shares computed from the plurality of cryptographically secure input shares containing the UE data; and (“every client C𝑖 ∈ C \ C∗ generates its share of the aggregate, (𝑖, U𝑖) with U𝑖 = Í C𝑗 ∈C\C∗ 𝑢𝑗𝑖, and sends that share to the server S. Note that, herein, C𝑖 uses its own share of the update, (𝑖, 𝑢𝑖𝑖), as well. The server recovers the aggregate U = Í C𝑖 ∈C\C∗ U𝑗 using robust reconstruction: U ← SS.robustRecon({(𝑖, U𝑖)}C𝑖 ∈ C\C∗ ).” Chowdhury § 4.4)
generating analytic data based at least on the plurality of aggregated output shares computed from the plurality of cryptographically secure input shares containing the UE data. (“the server chooses random values (𝑙1, · · · ,𝑙𝑘 ) ∈ F 𝑘 and recovers the sum Í𝑘 𝑖=1 𝑙𝑖 · 𝑤 𝑜𝑢𝑡 𝑖 in Round 3…. Current deployments of FL involves a single server who wants to train the global model.” Chowdhury § 4.4).
Chowdhury does not disclose the term: “instructing radio frequency (RF) circuitry to”
Mueck discloses an architecture for analytics with wireless devices, see e.g. Fig. 11.
A person of ordinary skill in the art before the effective filing date of the claimed invention would have combined Chowdhury with Mueck by implementing the disclosure of Chowdhury in a mobile network including wireless devices as the clients of Chowdhury. It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine Chowdhury with Mueck in order to allow a federated learning model to be exploited by UEs that contribute updates to expand local models for high-accuracy inference, Mueck ¶ 22; thereby providing ML services in ubiquitous wireless telephony.
As to claim 9, Chowdhury in view of Mueck discloses the machine of claim 8 but does not further disclose:
receiving an indication of a privacy or sensitivity level associated with the UE data; and
selecting a cooperative data collection scheme based at least on the privacy or sensitivity level associated with the UE data, wherein the analytic data is generated according to the selected cooperative data collection scheme.
Mueck further discloses:
receiving an indication of a privacy or sensitivity level associated with the UE data; and (“A device user (or any other data contribution entity in the network) can indicate data attributes of a specific client application instantiated to the device/User Equipment (UE), or machine as private/confidential or publicly shareable.” Mueck ¶ 139)
selecting a cooperative data collection scheme based at least on the privacy or sensitivity level associated with the UE data, wherein the analytic data is generated according to the selected cooperative data collection scheme. (“Trust Level 0 provides the highest level of trust in which no output data filtering is performed. Accordingly, higher Trust Level numbers provide lower levels of trusts with more data filtering.” Mueck ¶ 174. See also ¶ 145)
A person of ordinary skill in the art before the effective filing date of the claimed invention would have combined Chowdhury with Mueck by implementing the disclosure of Chowdhury in a mobile network including wireless devices as the clients of Chowdhury. It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine Chowdhury with Mueck in order to allow a federated learning model to be exploited by UEs that contribute updates to expand local models for high-accuracy inference, Mueck ¶ 22; thereby providing ML services in ubiquitous wireless telephony.
As to claim 10, Chowdhury in view of Mueck discloses the machine of claim 8 and further discloses:
outputting the analytic data for transmission to the at least one UE or NF in accordance with the request. (“The server S can choose a different Valid(·) for every iteration (the protocol described above corresponds to a single iteration of model training in FL).” Chowdhury § 4.4. “The server broadcasts the current parameters of the Model M to all the clients” Chowdhury § 2.1)
Chowdhury in view of Mueck, as combined in claim 1, does not explicitly disclose:
receiving, from at least one UE or network function (NF), a request for the analytic data; and
Mueck further discloses:
receiving, from at least one UE or network function (NF), a request for the analytic data; and
(“Data types to be provided by the AIS/selected AI agent to AIS consumer (e.g., UE)” Mueck ¶ 77. Also ¶¶ 121, 154. “A UE (AIS consumer) to communicate to the AIS information relating to a user/client application-specific task (e.g., intention to drive a vehicle from location A to location B” Mueck ¶ 44)
outputting the analytic data for transmission to the at least one UE (“Per 3GPP Technical Standard 23.501, the Network Exposure Function (NEF) supports a number of functionalities, among which are the following (NWDAF stands for Network Data Analytics Function): …. (b) Exposure of analytics: NWDAF analytics may be securely exposed by NEF for external party … (c) Retrieval of data from external party by NWDAF: Data provided by the external party may be collected by NWDAF” Mueck ¶¶ 33-36).
A person of ordinary skill in the art before the effective filing date of the claimed invention would have combined Chowdhury with Mueck by implementing the disclosure of Chowdhury in a mobile network including the elements of the AIS in the existing NWDAF network element. It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine Chowdhury with Mueck in order to allow a federated learning model to be exploited by UEs that contribute updates to expand local models for high-accuracy inference, Mueck ¶ 22; thereby providing ML services in ubiquitous wireless telephony.
As to claim 12, Chowdhury in view of Mueck discloses the machine of claim 8 and further discloses:
the operations further comprising training a machine learning (ML) model based at least in part on the plurality of aggregated output shares computed from the plurality of cryptographically secure input shares containing the UE data. (“the server chooses random values (𝑙1, · · · ,𝑙𝑘 ) ∈ F 𝑘 and recovers the sum Í𝑘 𝑖=1 𝑙𝑖 · 𝑤 𝑜𝑢𝑡 𝑖 in Round 3…. Current deployments of FL involves a single server who wants to train the global model.” Chowdhury § 4.4).
As to claim 13 Chowdhury in view of Mueck discloses the machine of claim 12 and further discloses:
wherein the analytic data is generated using the ML model trained on the plurality of aggregated output shares computed from the plurality of cryptographically secure input shares containing the UE data. (“the server chooses random values (𝑙1, · · · ,𝑙𝑘 ) ∈ F 𝑘 and recovers the sum Í𝑘 𝑖=1 𝑙𝑖 · 𝑤 𝑜𝑢𝑡 𝑖 in Round 3…. Current deployments of FL involves a single server who wants to train the global model.” Chowdhury § 4.4).
As to claim 15 Chowdhury discloses the machine of claim 14 and further discloses:
the operations further comprising outputting the aggregated output share for transmission (“every client C𝑖 ∈ C \ C∗ generates its share of the aggregate, (𝑖,U𝑖 ) …, and sends that share to the server S.” Chowdhury § 4.4)
… that is configured to generate analytic data based at least on the aggregated output share.
(“The server S can choose a different Valid(·) for every iteration (the protocol described above corresponds to a single iteration of model training in FL).” Chowdhury § 4.4. “The server broadcasts the current parameters of the Model M to all the clients” Chowdhury § 2.1)
Chowdhury does not explicitly disclose:
to a network data analytics function (NWDAF)
Mueck discloses:
to a network data analytics function (NWDAF)
(“Per 3GPP Technical Standard 23.501, the Network Exposure Function (NEF) supports a number of functionalities, among which are the following (NWDAF stands for Network Data Analytics Function): …. (b) Exposure of analytics: NWDAF analytics may be securely exposed by NEF for external party … (c) Retrieval of data from external party by NWDAF: Data provided by the external party may be collected by NWDAF” Mueck ¶¶ 33-36).
A person of ordinary skill in the art before the effective filing date of the claimed invention would have combined Chowdhury with Mueck by implementing the disclosure of Chowdhury in a mobile network including the elements of the AIS in the existing NWDAF network element. It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine Chowdhury with Mueck in order to allow a federated learning model to be exploited by UEs that contribute updates to expand local models for high-accuracy inference, Mueck ¶ 22; thereby providing ML services in ubiquitous wireless telephony.
As to claim 19 Chowdhury discloses the machine of claim 14 and further discloses:
receiving, …
inferred data generated by a local privacy-preserving machine learning (ML) model. (“Round 2 (Generate and Distribute Proofs). Every client generates
shares of its private update 𝑢𝑖 and the proof 𝜋𝑖 , and distributes
these shares to the other clients C\𝑖 .” Chowdhury § 4.4, the receiving by the ‘other clients’)
Chowdhury does not explicitly disclose:
during an uplink protocol data unit (PDU) session,
Mueck discloses:
during an uplink protocol data unit (PDU) session,
(“transport level packet marking in the uplink and downlink,” Mueck ¶ 211. Radio interface to the UE)
A person of ordinary skill in the art before the effective filing date of the claimed invention would have combined Chowdhury with Mueck by implementing the disclosure of Chowdhury in a mobile network including wireless devices as the clients of Chowdhury. It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine Chowdhury with Mueck in order to allow a federated learning model to be exploited by UEs that contribute updates to expand local models for high-accuracy inference, Mueck ¶ 22; thereby providing ML services in ubiquitous wireless telephony.
As to claim 20 Chowdhury in view of Mueck discloses the machine of claim 19 and further discloses:
the operations further comprising outputting (“every client C𝑖 ∈ C \ C∗ generates its share of the aggregate, (𝑖,U𝑖 ) …, and sends that share to the server S.” Chowdhury § 4.4)
…
is configured to generate analytic data based at least on the inferred data. (“The server S can choose a different Valid(·) for every iteration (the protocol described above corresponds to a single iteration of model training in FL).” Chowdhury § 4.4. “The server broadcasts the current parameters of the Model M to all the clients” Chowdhury § 2.1)
Chowdhury in view of Mueck, as combined in claim 19 does not explicitly disclose:
the inferred data to a network data analytics function (NWDAF) that
Mueck further discloses:
the inferred data to a network data analytics function (NWDAF) that
(“Per 3GPP Technical Standard 23.501, the Network Exposure Function (NEF) supports a number of functionalities, among which are the following (NWDAF stands for Network Data Analytics Function): …. (b) Exposure of analytics: NWDAF analytics may be securely exposed by NEF for external party … (c) Retrieval of data from external party by NWDAF: Data provided by the external party may be collected by NWDAF” Mueck ¶¶ 33-36).
A person of ordinary skill in the art before the effective filing date of the claimed invention would have combined Chowdhury with Mueck by implementing the disclosure of Chowdhury in a mobile network including the elements of the AIS in the existing NWDAF network element. It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine Chowdhury with Mueck in order to allow a federated learning model to be exploited by UEs that contribute updates to expand local models for high-accuracy inference, Mueck ¶ 22; thereby providing ML services in ubiquitous wireless telephony.
Claim(s) 2 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chowdhury et al., “EIFFeL: Ensuring Integrity for Federated Learning” (published 2022), in view of Mueck et al., US 2024/0298194 (filed 2022), and Qian et al., US 2021/0374605 (filed 2020).
As to claim 2, Chowdhury in view of Mueck discloses the machine of claim 1 but does not further disclose:
instructing the RF circuitry to transmit an indication of a privacy or sensitivity level associated with the UE data, wherein a different cooperative data collection scheme for the UE data is selected based on the privacy or sensitivity level associated with the UE data.
Mueck further discloses:
instructing the RF circuitry to transmit an indication of a privacy or sensitivity level associated with the UE data. (“A device user (or any other data contribution entity in the network) can indicate data attributes of a specific client application instantiated to the device/User Equipment (UE), or machine as private/confidential or publicly shareable.” Mueck ¶ 139)
A person of ordinary skill in the art before the effective filing date of the claimed invention would have combined Chowdhury with Mueck by implementing the disclosure of Chowdhury in a mobile network including wireless devices as the clients of Chowdhury. It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine Chowdhury with Mueck in order to allow a federated learning model to be exploited by UEs that contribute updates to expand local models for high-accuracy inference, Mueck ¶ 22; thereby providing ML services in ubiquitous wireless telephony.
Chowdhury in view of Mueck does not disclose: wherein a different cooperative data collection scheme for the UE data is selected based on the privacy or sensitivity level associated with the UE data.
Quian discloses: wherein a different cooperative data collection scheme for the UE data is selected based on the privacy or sensitivity level associated with the UE data.
(see Quian Fig. 6. “a first electronic device may access, from a data store associated with the first electronic device, a plurality of initial gradients associated with a machine-learning model. The first electronic device may then select one or more of the plurality of initial gradients for perturbation. In particular embodiments, the first electronic device may generate, based on a gradient-perturbation model, one or more perturbed gradients for the one or more selected initial gradients, respectively. For each selected initial gradient: an input to the gradient-perturbation model may comprise the selected initial gradient having a value x, the gradient-perturbation model may change x into a first continuous value with a first probability or a second continuous value with a second probability, and the first and second probabilities may be determined based on x. In particular embodiments, the first electronic device may further send, from the first electronic device to a second electronic device, the one or more perturbed gradients.” Quian ¶ 37. “As local differential privacy only needs each user's data to be indistinguishable, the client system may partition its data into multiple sub-datasets, calculate gradients and add noises on each sub-dataset then send all noisy gradients to the remote server 265.” Quian ¶ 53)
A person of ordinary skill in the art before the effective filing date of the claimed invention would have combined Chowdhury in view of Mueck with Quian by implementing local differential privacy at the client terminals of Showdhury. It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine Chowdhury in view of Mueck with Quian in order to prevent disclosure of private user data by adding particular noise to the data to allow for privacy preserved federated learning, Quian ¶ 27.
Claim(s) 11 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chowdhury et al., “EIFFeL: Ensuring Integrity for Federated Learning” (published 2022), in view of Mueck et al., US 2024/0298194 (filed 2022), and Krishnan et al., US 2023/0064266 (filed 2021).
As to claim 11, Chowdhury in view of Mueck discloses the machine of claim 8 and further discloses:
…
data associated with a machine learning (ML) model; and (“Broadcast. The server broadcasts the current parameters of the model M to all the clients.” Chowdhury § 2.1.)
outputting the data associated with the ML model during a downlink protocol data unit (PDU) session. (“transport level packet marking in the uplink and downlink,” Mueck ¶ 211. Radio interface to the UE)
Chowdhury in view of Mueck, as combined in claim 1, does not explicitly disclose:
retrieving, from an analytics data repository function (ADRF), data associated with a machine learning (ML) model
Krishnan discloses:
retrieving, from an analytics data repository function (ADRF), data associated with a machine learning (ML) model
(“the network entity includes at least one of an ADRF,” ¶ 129. “At 530, the UE 115-c may alternatively input the requested information, from the network entity 205-a, to improve a machine learning model at the UE 115-c.” Krishnan ¶ 95)
A person of ordinary skill in the art before the effective filing date of the claimed invention would have combined Chowdhury in view of Mueck with Krishnan by obtaining ML model parameters, that the server of Chowdhury is required to distribute, from an ADRF function as detailed in Krishnan. It would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to combine Chowdhury in view of Mueck with Krishnan in order to allow incorporation of the elements of Chowdhury into 3GPP federated learning infrastructures; thereby increasing interoperability with existing deployed systems.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. See PTO-892, particularly:
Baroudi et al., US 10,492,064, discloses protecting privacy of a base station in a wireless sensor network.
Pandharipande et al., US 10,142,779, discloses privacy protected location sharing.
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MICHAEL W CHAO whose telephone number is (571)272-5165. The examiner can normally be reached M, W-F 8-5.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Rupal Dharia can be reached at (571) 272-3880. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/MICHAEL W CHAO/Primary Examiner, Art Unit 2492
Prosecution Insights