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 .
Priority
Acknowledgment is made of applicant’s claim for foreign priority under 35 U.S.C. 119 (a)-(d). The certified copy has been filed in parent Application No. IN 202341041894, filed on June 23, 2023.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on May 21, 2025 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statements are being considered by the examiner.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 18-22, 30, 31 is rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. Below is the claim-by-claim analysis.
Claim 18, 31
Step 1: Recites an apparatus (claim 18) and a method (claim 31). Therefore, they are directed to a statutory category of invention.
Step 2A Prong 1: The claim recites:
determine, based on the one or more rules, whether to switch from a current operational mode to another operational mode from the set of operational modes
Determining x based on y requires abstract reasoning and is thus a mental process.
Step 2A Prong 2: The judicial exception is not integrated into a practical application. The remaining limitations of the claim are directed either to insignificant extra-solution activity (“receive a configuration…”) or merely apply the judicial exception (“perform the current operational mode or the other operational mode based on the determination”).
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis mirrors the analysis of step 2A prong 2.
Claim 19
Step 1: Recites an apparatus. Therefore, it is directed to a statutory category of invention.
Step 2A Prong 1: The claim recites the abstract idea it inherits from the claim it depends on.
Step 2A Prong 2: The judicial exception is not integrated into a practical application. The remaining limitations of the claim are directed to insignificant extra-solution activity (“receive a request…”, “transmit the assistance information…”).
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis mirrors the analysis of step 2A prong 2.
Claim 20
Step 1: Recites an apparatus. Therefore, it is directed to a statutory category of invention.
Step 2A Prong 1: The claim recites the abstract idea it inherits from the claim it depends on.
Step 2A Prong 2: The judicial exception is not integrated into a practical application. The remaining limitations of the claim are directed to further information about the operational modes from the configuration (“wherein the assistance information further comprises one or more parameters related to the set of operational modes, wherein the one or more parameters indicate at least one of”), that isn’t enough to render it more than the judicial exception recited.
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis mirrors the analysis of step 2A prong 2.
Claim 21
Step 1: Recites an apparatus. Therefore, it is directed to a statutory category of invention.
Step 2A Prong 1: The claim recites the abstract idea it inherits from the claim it depends on.
Step 2A Prong 2: The judicial exception is not integrated into a practical application. The remaining limitation of the claim is directed to insignificant extra-solution activity (“transmit an indication…”).
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis mirrors the analysis of step 2A prong 2.
Claim 22
Step 1: Recites an apparatus. Therefore, it is directed to a statutory category of invention.
Step 2A Prong 1: The claim recites the abstract idea it inherits from the claim it depends on.
Step 2A Prong 2: The judicial exception is not integrated into a practical application. The remaining limitation of the claim are directed to further information about the operational modes from the configuration (“a first operational mode comprising…”, “a second operational mode comprising…”, “a third operational mode comprising…”, “a fourth operational mode comprising…” ), that isn’t enough to render it more than the judicial exception recited.
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis mirrors the analysis of step 2A prong 2.
Claim 30
Step 1: Recites an apparatus. Therefore, it is directed to a statutory category of invention.
Step 2A Prong 1: The claim recites:
determine the one or more rules based at least partly on the assistance information
Determining x based on y requires abstract reasoning and is thus a mental process.
Step 2A Prong 2: The judicial exception is not integrated into a practical application. The remaining limitations of the claim are directed to insignificant extra-solution activity (“transmit… a request for providing assistance…”, “receive… the assistance information…”).
Step 2B: The claim does not contain significantly more than the judicial exception. The analysis mirrors the analysis of step 2A prong 2.
Claim Rejections - 35 USC § 102
(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.
(a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention.
Claims 18, 21, 28-29, 31 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Soldati et al.
(“WO 2023017102”).
Regarding claim 18, Soldati teaches an apparatus comprising at least one processor, and at least one memory storing instructions that, when executed by the at least one processor (Page 10, Lines 26-27, “Embodiments of a method executed by a first network node to optimize the performance of an AI/ML model or algorithm in a radio communication network are disclosed”), cause the apparatus at least to:
receive a configuration indicating one or more rules for switching between a set of operational modes associated with at least one of: inference, data collection or training of a machine learning model, the machine learning model being associated with a network optimization function1 (Page 11, Lines 12-20, “The first network node 200 receives at least one SECOND MESSAGE from the second network node 202. The SECOND MESSAGE comprises one or more of… indications of at least a performance metric associated to the AI/ML model or algorithm for which optimization is desirable… instructions or recommendations… to re-train/update/optimize/ replace with a new model at least one AI/ML of model… to test/validate at least one AI/ML model”);
determine, based on the one or more rules, whether to switch from a current operational mode to another operational mode from the set of operational modes (Page 13, Lines 10-13, “In another exemplifying case, the first network node 200 may determine whether training, optimizing, updating, or replacing with a new model the AI/ML model or algorithm is necessary based on at least an indication of at least a performance metric associated to the AI/ML model or algorithm or based on instructions or recommendations to re-train/update/optimize the AI/ML model received with a SECOND MESSAGE”)
Regarding claim 21, Soldati teaches the apparatus according to claim 18, further being caused to: transmit an indication indicating the switch to the other operational mode, based on determining to switch from the current operational mode to the other operational mode (Page 23, Lines 30-33, “Transmit a THIRD MESSAGE to the third network node 500 (e.g., a Model Inference) (see step 502) or to the second network node 202, where the THIRD MESSAGE comprises… A trained, optimized, or updated AI/ML model (e.g., via a Model Deployment Update)”2)
Regarding claim 28, Soldati teaches the apparatus according to claim 18, wherein the network optimization function comprises at least one of: radio resource management, network energy saving, load balancing, mobility optimization, cell selection, carrier selection, scheduling, or beam management (Page 1, Lines 18-20, “identify the benefits of Al enabled NG-RAN through possible use cases e g. energy saving, load balancing, mobility management, coverage optimization, etc.”).
Regarding claim 29, Soldati teaches an apparatus comprising at least one processor, and at least one memory storing instructions that, when executed by the at least one processor (see claim 18 analysis), cause the apparatus at least to: transmit, to a radio access network node3, a configuration indicating one or more rules for switching between a set of operational modes associated with at least one of: inference, data collection or training of a machine learning model, the machine learning model being associated with a network optimization function (Page 24, Lines 20-30, The second network node 202 transmits at least one SECOND MESSAGE to the first network node 200, where the at least one SECOND MESSAGE comprises… instructions or recommendations… to re-train/update/optimize/ replace with a new model at least one AI/ML of model or algorithm”).
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.
Claims 19-20, 30 are rejected under 35 U.S.C. 103 as being unpatentable over Soldati et al. (“WO 2023017102”) in view of Zhu et al. (“US 12185423”, 2022).
Regarding claim 19, Soldati teaches a set of operational modes (see claim 18 analysis). Soldati fails to teach the receiving of a request for providing assistance information indicating the set of operational modes supported by the apparatus and transmitting the assistance information indicating the set of operational modes supported by the apparatus, wherein the assistance information is transmitted based on receiving the request. However, Zhu teaches receiving a request for providing assistance information4 indicating the set of [machine learning models] supported by the apparatus; and transmitting the assistance information indicating the set of [machine learning models] supported by the apparatus, wherein the assistance information is transmitted based on receiving the request (Figure 4).
Soldati and Zhu are considered analogous to the invention because all are directed towards network and wireless systems that incorporate machine learning for network optimization. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Soldati to incorporate the teachings of Zhu, and receive a request for providing assistance information prior to transmitting it. Doing so is a well-known procedure in the art and thus would have been an obvious method for initiating communication between two network nodes.
Regarding claim 20, Soldati teaches the use of operational modes (see claim 18 analysis) for the purpose of performing network optimization (Page 1, Lines 18-20, “identify the benefits of Al enabled NG-RAN through possible use cases e g. energy saving, load balancing, mobility management, coverage optimization, etc.”). Soldati fails to teach wherein the assistance information further comprises one or more parameters related to the set of operational modes, wherein the one or more parameters indicate at least one of: the machine learning model to use in at least one operational mode of the set of operational modes, or a non-machine-learning-based algorithm to use in at least one operational mode of the set of operational modes for performing the network optimization function.
However, Zhu teaches wherein the assistance information further comprises one or more parameters related to the set of [machine learning models], wherein the one or more parameters indicate at least one of: the machine learning model to use in5… the set of [machine learning models] (Column 2, Lines 7-11, “Some aspects described herein relate to a method of wireless communication performed by a user equipment (UE). The method may include transmitting capability information that indicates support for one or more model combinations of machine learning (ML) models”).
Soldati and Zhu are considered analogous to the invention because all are directed towards network and wireless systems that incorporate machine learning for network optimization. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Soldati to incorporate the teachings of Zhu, and request and indicate what is supported by the apparatus. Doing so reduces the risk of system errors by having both ends of the communication understand what rules and metrics apply to what is used.
Regarding claim 30, Soldati teaches the use of operational modes (see claim 18 analysis) and a set of rules for switching between operational modes (see claim 18 analysis) . Soldati fails to teach the transmitting, to the radio access network node, a request for providing assistance information indicating the set of operational modes supported by the radio access network node; receiving, from the radio access network node, the assistance information indicating the set of operational modes supported by the radio access network node; and determining the one or more rules based at least partly on the assistance information.
However, Zhu teaches:
transmitting, to the radio access network node, a request for providing assistance information indicating the set of [machine learning models] supported by the radio access network node;
receiving, from the radio access network node, the assistance information indicating the set of [machine learning models] supported by the radio access network node (Figure 4); and
determining the one or more [rules]6 based at least partly on the assistance information (Column 21, Lines 14-18, “ As shown by reference number 540, the UE and the network node may communicate based at least in part on the ML models of the model combination. In some aspects, the UE may use a selected model combination to configure wireless communication of the UE (e.g., a RAN configuration).”).
Soldati and Zhu are considered analogous to the invention because all are directed towards network and wireless systems that incorporate machine learning for network optimization. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Soldati to incorporate the teachings of Zhu, and request and indicate what is supported by the apparatus, and determine policy based on that. Doing so reduces the risk of system errors by having both ends of the communication understand what rules and metrics apply to what is used.
Claims 22-25, 27 are rejected under 35 U.S.C. 103 as being unpatentable over Soldati et al. (“WO 2023017102”) in view of Tang et al. (“WO 2022133884”) and in further view of Singh et al. (“US 20210342736”).
Regarding claim 22, Soldati teaches wherein the set of operational modes comprises:
a first operational mode comprising running the inference with the machine learning model for performing the network optimization function (Figure 1, Model Inference)
…
and a fourth operational mode comprising determining one or more performance metrics of the machine learning model (Page 3, Lines 4-7, “method performed by first network node comprises receiving, from at least one other network node, at least one message comprising… indications of at least a performance metric associated to the AI/ML model”).
Soldati fails to teach a second operational mode comprising deactivating the machine learning model, and using a non-machine-learning-based algorithm for performing the network optimization function, and a third operational mode comprising collecting labelled training data, while using the non-machine-learning-based algorithm for performing the network optimization function, and updating the machine learning model based on the labelled training data.
However:
Tang teaches teach a second operational mode comprising deactivating the machine learning model, and using a non-machine-learning-based algorithm for performing the network optimization function7 and a third operational mode comprising… while8 using the non-machine-learning-based algorithm for performing the network optimization function, and updating the machine learning model… (Page 39, Paragraph 115, “Advantageously, by allowing dynamic switching between AI mode and non-AI mode, if the base station notices that using AI mode is not efficient or effective, the base station can switch to the traditional method, and/or indicate a retraining procedure”); and
Singh teaches the collecting of labeled training data (Paragraph 16, “…noticing undesirable performance characteristics and initiating collection of labeled data for retraining”).
Soldati, Tang and Zhu are considered analogous to the invention because all are directed towards machine learning methods. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Soldati to incorporate the teachings of Tang and Zhu, and include non-ML algorithm(s) and collecting labeled training data as operational modes. Allowing the apparatus to switch between a machine learning model and a non-ML model allows for a maintenance of performance when the machine learning model degrades (see Paragraph 115 of Tang) and collecting labeled training data provides a means to update the degraded machine learning model, allowing for an improvement in its performance.
Regarding claim 23, Soldati teaches wherein the one or more rules indicate at least one of: switching from the first operational mode to the fourth operational mode at one or more pre-defined time intervals…9 (Page 16, Lines 24-33, “An indication of a time interval for monitoring the performance of the AI/ML model or algorithm, which may comprise one or more of… A starting time… An ending time… An indication of pause (applicable to a part of the monitoring or to the complete monitoring) and/or a pause duration/time to resume… An indication of resume”).
Regarding claim 24, Soldati teaches wherein the one or more performance metrics of the machine learning model are determined by comparing one or more outputs of the machine learning model with one or more corresponding outputs of the non-machine-learning-based algorithm, or with a reference threshold or range (Page 18, Paragraph 16-18, “One or more threshold for at least a performance metric associated to an AI/ML model or algorithm to be monitored in the list of… Minimum, maximum, average of the accuracy of the AI/ML model or algorithm”).
Regarding claim 25, Soldati teaches wherein the one or more performance metrics of the machine learning model are determined by comparing one or more outputs of the machine learning model with corresponding ground truth data (Page 16, Lines 13-14, “least one performance metric associated to an AI/ML model or algorithm to be monitored in the list of:… Accuracy”10).
Regarding claim 27, Soldati teaches wherein the one or more performance metrics comprise at least one of: accuracy, convergence time, reliability, or statistical significance (see claim 24 analysis).
Claim 26 is rejected under 35 U.S.C. 103 as being unpatentable over Soldati et al. (“WO 2023017102”) in view of Tang et al. (“WO 2022133884”) and in further view of Singh et al. (“US 20210342736”) and Batalov et al. (“US 11106994”, 2016).
Regarding claim 26, Soldati teaches a network optimization function (see claim 28 analysis). Soldati fails to teach determining one or more performance metrics of the non-machine-learning-based algorithm associated with the network optimization function by comparing one or more outputs of the non-machine-learning-based algorithm with the corresponding ground truth data; comparing the one or more performance metrics of the machine learning model and the one or more performance metrics of the non-machine-learning-based algorithm; and determining, based on the comparison, whether to use the machine learning model or the non-machine-learning-based algorithm for performing the network optimization function.
However, Batalov teaches:
determining one or more performance metrics of the non-machine-learning-based algorithm… by comparing one or more outputs of the non-machine-learning-based algorithm with the corresponding ground truth data (Column 5 Line 63 to Column 6 Line 4, “In another example, a baseline model may be represented in terms of a model that does not use machine learning, where weights may be assigned for certain simple or rule based outcomes. The non-machine learning baseline model may be represented using simple percentages, rules, a naive model with labels, thresholds to define outcomes, or as definitions what occurs if no action at all is taken. Further, the total value of the non-machine learning model may be calculated”, Column 4 , Equation 211);
comparing the one or more performance metrics of the machine learning model and the one or more performance metrics of the non-machine-learning-based algorithm (Column 5, Lines 4-7, “Then a machine learning model may be compared against the non-machine learning baseline model to see if classification using machine learning can improve the total value of the model to a customer”); and
determining, based on the comparison, whether to use the machine learning model or the non-machine-learning-based algorithm… (Column 5, Lines 7-9, “If the machine learning improves the total value of the model then the machine learning model may be selected over the non-machine learning model”).
Soldati and Batalov are considered analogous to the invention because all are directed towards machine learning methods. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the invention to have modified Soldati to incorporate the teachings of Batalov and included a non-ML method to compare to the ML algorithm, and selected the better performing one. As evidenced by the Tang et al., doing so is well-established in the field of network optimization, and explicitly comparing the outputs of both the ML and non-ML model allows for a more precise means of adjusting what method of prediction is used.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MATTHEWOS MESFIN whose telephone number is (571)270-0782. The examiner can normally be reached Monday-Friday 8am-5pm.
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, Cesar Paula can be reached at (571) 272-4128. 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.
/MATTHEWOS MESFIN/Examiner, Art Unit 2145
/CESAR B PAULA/Supervisory Patent Examiner, Art Unit 2145
1 Specific network optimization function listed in claim 28 analysis
2 Retraining/optimizing/updating are different “modes” from executing. Sending the updated model can be broadly interpreted as indicating a switch in operation.
3 ” A network node can be a Radio Access Network (RAN) node” Page 9, Line 27
4 Referred to in the reference as “capability”, can be broadly interpreted as the same thing
5 The use of an ML model in at least one operational mode is taught by Soldati, as evidenced by the claim 18 analysis.
6 The rules, as written in the claim, refer to a policy for network optimization, as what is taught in Soldati. Determining a generic set of rules (in this case wireless configuration for communicating with the RAN node) based on received assistance information from a RAN node is taught by Zhu and in combination with Soldati the whole limitation is covered.
7 Reference teaches a means of improving network scheduling with AI, ” …AI technologies, especially Ml (deep learning) technologies, had been introduced to telecommunication for improving the system performance and efficiency”, Page 23, Paragraph 69
8 Tang and Singh combined teach the collection of labeled training data while the non-ML algorithm runs.
9 Because the claim asks for at least one of the listed rules, the rest do not need to be listed.
10 Accuracy necessitates comparing a model output with a ground truth label. This is further backed up by the specification of the application, where it states that “In this example, the one or more performance metrics, such as accuracy, of the machine learning model are determined from time to time by comparing one or more outputs of the machine learning model with corresponding ground truth data, Paragraph 80
11 An example of a performance metric that the model can be optimized over
Prosecution Insights