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 Office Action is in response to communications filed on 11/12/2024.
Claims 1-20 are pending and presented for examination.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 11/12/2024 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Objections
Claims 1 & 16 objected to because of the following informalities: these claims recite “the based station”, which includes a typographical error. For the purpose of this review, examiner is interpreting these limitations as “the base station”. Appropriate correction is required.
Claim 7 objected to because of the following informalities: this claim recites “wherein the plurality of properties of supported functionalities for each of the plurality of AI/ML use cases includes properties of at least one of an observation window or a prediction window different beam sets for an AI/ML channel state information (CSI) report use case.”, which is not proper English. For the purpose of this review, examiner is interpreting this claim as “wherein the plurality of properties of supported functionalities for each of the plurality of AI/ML use cases includes properties of at least one of an observation window or a prediction window of different beam sets for an AI/ML channel state information (CSI) report use case. Appropriate correction is required.
Claim 15 objected to because of the following informalities: this claim recites “therein the CPU limit is based on a UE capability or a predetermined value” which includes a typographical error. For the purpose of this review, examiner is interpreting this claim as “wherein the CPU limit is based on a UE capability or a predetermined value”.
Claim Interpretation
Several of the claims in the present application recite limitations that recite “artificial intelligence (AI)/Machine Learning (ML)” or “AI/ML”. For the purpose of this review, examiner is interpreting the “/” mark as an “OR” function. Specifically, examiner is interpreting “artificial intelligence (AI)/Machine Learning (ML)” as “artificial intelligence (AI) or Machine Learning (ML)” and interpreting “AI/ML” as “AI or ML”.
Several of the claims in the present application recite Markush groups in the format of “at least one of A, B or C” (see MPEP §2117). For the purpose of this review, the examiner is interpreting these Markush claims as a single element selection from a closed group of elements consisting of alternatives A, B, C, A&B, A&C, B&C, or A&B&C.
Claim 13 recites “wherein the CSI report further includes indices of a predetermined number of best models and their corresponding monitored metrics”. Page 52, last paragraph of the current application specification discloses that “With a different type of reporting, the UE may indicate the IDs of the best K models with the largest (i.e. best) metrics, as well as their corresponding metrics”. Page 45, last paragraph of the current application specification discloses that Model IDs are indexed from 0 to M-1 in ascending order of the data set IDs. Based on the above disclosures in the current application specification, Examiner is interpreting “wherein the CSI report further includes indices of a predetermined number of best models and their corresponding monitored metrics” as “wherein the CSI report further includes model IDs of a best K models and their corresponding monitored metrics”.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-6 & 16-20 are rejected under 35 U.S.C. 103 as being unpatentable over Cheng et al. (WO 2023/211343)(herein after “Cheng”) in view of Leng et al. (US 2024/0098533)(herein after “Leng”).
Regarding claim 1, Cheng discloses a method performed by a user equipment (UE), the method comprising: transmitting, to a base station, a plurality of properties of supported functionalities for each of a plurality of artificial intelligence (AI)/machine leaning (ML) use cases (Fig 3, [0064]-[0065] & [0067] disclose a method for a UE 200 to report (i.e. transmit) to a network node 300 ML feature set information for running different ML model feature sets (i.e. a plurality of ML use cases) in parallel including ML model-based functionalities or capabilities supported by the UE and subsets and parameter configurations for combinations of reported ML-based functionalities that can be configured simultaneously (i.e. a plurality of properties of supported functionalities for each ML use case). Fig 6 & [0167] disclose that network node 300 may be a base station.),
wherein each of the plurality of AI/ML use cases is configured with an individual use case identifier (ID) ([0080] & [0087] discloses that the UE may indicate subsets of models that can run concurrently (i.e. ML use cases) using subset identifications such as subset0={A,B], subset1={A,C], subset2={C,D,E] etc., where A, B, C, D, E represent ML model functionalities.);
transmitting, to the base station, a maximum number of AI/ML active functionalities that the UE supports across the plurality of AI/ML use cases ([0087]-[0096] disclose that the UE indicates to a gNB (i.e. transmits) a maximum number of concurrent models across model functionalities A, B, C, D & E.);
receiving, from the based station, a report configuration based on the plurality of properties of supported functionalities for each of the plurality of AI/ML use cases and the maximum number of AI/ML active functionalities that the UE supports across the plurality of AI/ML use cases ([0072] & [0096] disclose that the UE receives from the network node 300 (i.e. the base station) ML model configuration (i.e. a report configuration) for combinations of ML model functionalities aligned with the constraints provided by the UE ML model feature sets.);
Cheng fails to disclose but Leng teaches generating a report based on the report configuration (Fig 6 & [0089]-[0090] discloses the UE performs AI/ML model monitoring for configured AI/ML model use cases and performs related operations including reporting of assistance information for the AI/ML monitoring (i.e. generating a report based on the report configuration).); and
transmitting the report to the base station (Fig 6 & [0089]-[0090] discloses the NW receives from the UE (i.e. the UE transmits to the NW) the UE reported assistance information for AI/ML model monitoring. Fig 1 & [0049] disclose that the UE reported assistance information for AI/ML model monitoring must be transmitted to a base station in order to be transmitted to the NW.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have a method performed by a user equipment (UE), the method comprising: transmitting, to a base station, a plurality of properties of supported functionalities for each of a plurality of artificial intelligence (AI)/machine leaning (ML) use cases, wherein each of the plurality of AI/ML use cases is configured with an individual use case identifier (ID); transmitting, to the base station, a maximum number of AI/ML active functionalities that the UE supports across the plurality of AI/ML use cases; and receiving, from the based station, a report configuration based on the plurality of properties of supported functionalities for each of the plurality of AI/ML use cases and the maximum number of AI/ML active functionalities that the UE supports across the plurality of AI/ML use cases, as disclosed by Cheng, and generating a report based on the report configuration; and transmitting the report to the base station, as taught by Leng. The motivation to do so would have been to have a method for a UE to transmit to a base station ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE, receive a configuration for ML models and features for the UE to run, monitor and for which to report assistance information including ML model monitoring results, generate and transmit a report with the assistance information to the base station so that the base station can use the assistance information to evaluate the accuracy of the selected ML model and functionalities to determine if the base station should transmit a report reconfiguration to the UE to adapt the ML models and functionalities that are active at the UE in order to improve performance.
Regarding claim 2, Cheng in view of Leng disclose the method of claim 1.
Cheng discloses further comprising transmitting, to the base station, a maximum number of AI/ML active functionalities that the UE supports for each of the plurality of AI/ML use cases, wherein the report configuration is further based on the maximum number of AI/ML active functionalities that the UE supports for each of the plurality of AI/ML use cases ([0087]-[0096] disclose that the UE indicates to a gNB (i.e. transmits) a maximum number of concurrent models across model functionalities A, B, C, D & E. [0087] discloses that the maximum number of concurrent models can be indicated for each ML model feature set (i.e. each AI/ML use case), for example a first subset0={A,B] has a maximum number of concurrent models of 2, a second subset1={A,C] has a maximum number of concurrent models of 2, and a third subset2={C,D,E] has a maximum number of concurrent models of 3.).
Regarding claim 3, Cheng in view of Leng disclose the method of claim 1.
Cheng discloses further comprising transmitting, to the base station, an indication of how activated, configured functionalities are counted by the UE, wherein the report configuration is further based on how the activated, configured functionalities are counted by the UE ([0100] disclose that an active/operating ML model may be associated with a complexity indicator, and each ML model is counted by its complexity indicator towards a complexity cap. [0072] & [0096] disclose that the UE receives from the network node 300 (i.e. the base station) ML model configuration (i.e. a report configuration) for combinations of ML model functionalities aligned with the constraints provided by the UE ML model feature sets. Thus, network node 300 would base the ML model configuration on counting the complexity indicators and comparing against the complexity cap.).
Regarding claim 4, Cheng in view of Leng disclose the method of claim 3.
Cheng discloses wherein, for each use case ID, the indication includes a notification that the UE includes multiple models for one functionality ([0038], [0044] & [0069] discloses that each of the ML models may have a version number, each ML model (i.e. each ML model functionality) may be specified by a version number and there may be different versions (i.e. multiple models) of the same ML model functionality. Thus, as disclosed in [0087]-[0096], for each ML model set identification (e.g. subset 0, subset1 and subset2), the indications of the model functionalities A, B, C, D, E may each have a version number indicating there are multiple models for the same model functionality.).
Regarding claim 5, Cheng in view of Leng disclose the method of claim 3.
Cheng discloses wherein, for each use case ID, the indication includes a notification that the UE includes a single model for a group of configured functionalities ([0087] discloses that each ML model feature set identified as subset0, subset1 and subset2, etc. represents a single model for a group of ML model functionalities configured at the UE.).
Regarding claim 6, Cheng in view of Leng disclose the method of claim 1.
wherein the plurality of properties of supported functionalities for each of the plurality of AI/ML use cases includes properties of different beam sets for an AI/ML beam prediction use case ([0104]-[0108] disclose that ML model feature sets may include ML processing units (MLPUs) for beam management and beam selection (i.e. beam prediction) that represent computational complexity associated with the supported beam management functionality.).
Regarding claim 16, Cheng discloses a user equipment (UE), comprising: a transceiver ([0154] discloses that the UE may contain transceivers.); and
a processor ([0154] discloses that the UE may contain processors.) configured to:
transmit, to a base station, via the transceiver, a plurality of properties of supported functionalities for each of a plurality of artificial intelligence (AI)/machine leaning (ML) use cases (Fig 3, [0064]-[0065] & [0067] disclose a method for a UE 200 to report (i.e. transmit) to a network node 300 ML feature set information for running different ML model feature sets (i.e. a plurality of ML use cases) in parallel including ML model-based functionalities or capabilities supported by the UE and subsets and parameter configurations for combinations of reported ML-based functionalities that can be configured simultaneously (i.e. a plurality of properties of supported functionalities for each ML use case). Fig 6 & [0167] disclose that network node 300 may be a base station.),
wherein each of the plurality of AI/ML use cases is configured with an individual use case identifier (ID) ([0080] & [0087] discloses that the UE may indicate subsets of models that can run concurrently (i.e. ML use cases) using subset identifications such as subset0={A,B], subset1={A,C], subset2={C,D,E] etc., where A, B, C, D, E represent ML model functionalities.),
transmit, to the base station, via the transceiver, a maximum number of AI/ML active functionalities that the UE supports across the plurality of AI/ML use cases ([0087]-[0096] disclose that the UE indicates to a gNB (i.e. transmits) a maximum number of concurrent models across model functionalities A, B, C, D & E.),
receive, from the based station, via the transceiver, a report configuration based on the plurality of properties of supported functionalities for each of the plurality of AI/ML use cases and the maximum number of AI/ML active functionalities that the UE supports across the plurality of AI/ML use cases ([0072] & [0096] disclose that the UE receives from the network node 300 (i.e. the base station) ML model configuration (i.e. a report configuration) for combinations of ML model functionalities aligned with the constraints provided by the UE ML model feature sets.),
Cheng fails to disclose but Leng teaches generate a report based on the report configuration (Fig 6 & [0089]-[0090] discloses the UE performs AI/ML model monitoring for configured AI/ML model use cases and performs related operations including reporting of assistance information for the AI/ML monitoring (i.e. generating a report based on the report configuration).); and
transmit the report to the base station (Fig 6 & [0089]-[0090] discloses the NW receives from the UE (i.e. the UE transmits to the NW) the UE reported assistance information for AI/ML model monitoring. Fig 1 & [0049] disclose that the UE reported assistance information for AI/ML model monitoring must be transmitted to a base station in order to be transmitted to the NW.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have a user equipment (UE), comprising: a transceiver; and a processor configured to: transmit, to a base station, via the transceiver, a plurality of properties of supported functionalities for each of a plurality of artificial intelligence (AI)/machine leaning (ML) use cases, wherein each of the plurality of AI/ML use cases is configured with an individual use case identifier (ID), transmit, to the base station, via the transceiver, a maximum number of AI/ML active functionalities that the UE supports across the plurality of AI/ML use cases, and receive, from the based station, via the transceiver, a report configuration based on the plurality of properties of supported functionalities for each of the plurality of AI/ML use cases and the maximum number of AI/ML active functionalities that the UE supports across the plurality of AI/ML use cases, as disclosed by Cheng, and generate a report based on the report configuration; and transmit the report to the base station, as taught by Leng. The motivation to do so would have been to have a UE that can transmit to a base station ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE, receive a configuration for ML models and features for the UE to run, monitor and for which to report assistance information including ML model monitoring results, generate and transmit a report with the assistance information to the base station so that the base station can use the assistance information to evaluate the accuracy of the selected ML model and functionalities to determine if the base station should transmit a report reconfiguration to the UE to adapt the ML models and functionalities that are active at the UE in order to improve performance.
Regarding 17, Cheng in view of Leng disclose the UE of claim 16.
Cheng discloses wherein the processor is further configured to transmit, to the base station, via the transceiver, a maximum number of AI/ML active functionalities that the UE supports for each of the plurality of AI/ML use cases ([0087]-[0096] disclose that the UE indicates to a gNB (i.e. transmits) a maximum number of concurrent models across model functionalities A, B, C, D & E.),
and wherein the report configuration is further based on the maximum number of AI/ML active functionalities that the UE supports for each of the plurality of AI/ML use cases ([0087] discloses that the maximum number of concurrent models can be indicated for each ML model feature set (i.e. each AI/ML use case), for example a first subset0={A,B] has a maximum number of concurrent models of 2, a second subset1={A,C] has a maximum number of concurrent models of 2, and a third subset2={C,D,E] has a maximum number of concurrent models of 3.).
Regarding claim 18, Cheng in view of Leng disclose the UE of claim 16.
Cheng discloses wherein the processor is further configured to transmit, to the base station, via the transceiver, an indication of how activated, configured functionalities are counted by the UE ([0100] disclose that an active/operating ML model may be associated with a complexity indicator, and each ML model is counted by its complexity indicator towards a complexity cap.),
wherein the report configuration is further based on how the activated, configured functionalities are counted by the UE ([0072] & [0096] disclose that the UE receives from the network node 300 (i.e. the base station) ML model configuration (i.e. a report configuration) for combinations of ML model functionalities aligned with the constraints provided by the UE ML model feature sets. Thus, network node 300 would base the ML model configuration on counting the complexity indicators and comparing against the complexity cap.).
Regarding claim 19, Cheng in view of Leng disclose the UE of claim 16.
Cheng discloses wherein, for each use case ID, the indication includes: a notification that the UE includes multiple models for one functionality ([0038], [0044] & [0069] discloses that each of the ML models may have a version number, each ML model (i.e. each ML model functionality) may be specified by a version number and there may be different versions (i.e. multiple models) of the same ML model functionality. Thus, as disclosed in [0087]-[0096], for each ML model set identification (e.g. subset 0, subset1 and subset2), the indications of the model functionalities A, B, C, D, E may each have a version number indicating there are multiple models for the same model functionality.), or
a notification that the UE includes a single model for a group of configured functionalities ([0087] discloses that each ML model feature set identified as subset0, subset1 and subset2, etc. represents a single model for a group of ML model functionalities configured at the UE.).
Regarding claim 20, Cheng discloses a method performed by a base station, the method comprising: receiving, from a user equipment (UE), a plurality of properties of supported functionalities for each of a plurality of artificial intelligence (AI)/machine leaning (ML) use cases (Fig 3, [0064]-[0065] & [0067] disclose a method for a network node 300 ML to receive feature set information, reported from a UE 200, for running different ML model feature sets (i.e. a plurality of ML use cases) in parallel including ML model-based functionalities or capabilities supported by the UE and subsets and parameter configurations for combinations of reported ML-based functionalities that can be configured simultaneously (i.e. a plurality of properties of supported functionalities for each ML use case). Fig 6 & [0167] disclose that network node 300 may be a base station.),
wherein each of the plurality of AI/ML use cases is configured with an individual use case identifier (ID) ([0080] & [0087] discloses that the UE may indicate subsets of models that can run concurrently (i.e. ML use cases) using subset identifications such as subset0={A,B], subset1={A,C], subset2={C,D,E] etc., where A, B, C, D, E represent ML model functionalities.);
receiving, from the UE, a maximum number of AI/ML active functionalities that the UE supports across the plurality of AI/ML use cases ([0087]-[0096] disclose a gNB receiving, from the UE, an indication of a maximum number of concurrent models across model functionalities A, B, C, D & E.);
generating a report configuration based on the plurality of properties of supported functionalities for each of the plurality of AI/ML use cases and the maximum number of AI/ML active functionalities that the UE supports across the plurality of AI/ML use cases ([0072] & [0096] disclose that the UE receives from the network node 300 (i.e. the base station) ML model configuration (i.e. a report configuration generated by network node 300) for combinations of ML model functionalities aligned with the constraints provided by the UE ML model feature sets.); and
transmitting, to the UE, the report configuration ([0072] & [0096] disclose that the UE receives from the network node 300 (i.e. the base station) ML model configuration (i.e. a report configuration transmitted by network node 300) for combinations of ML model functionalities aligned with the constraints provided by the UE ML model feature sets.).
Cheng fails to disclose but Leng teaches receiving, from the UE, the report to the base station (Fig 6 & [0089]-[0090] discloses the NW receives from the UE the UE reported assistance information for AI/ML model monitoring. Fig 1 & [0049] disclose that the UE reported assistance information for AI/ML model monitoring must be received by a base station in order to be received by the NW.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have a method performed by a base station, the method comprising: receiving, from a user equipment (UE), a plurality of properties of supported functionalities for each of a plurality of artificial intelligence (AI)/machine leaning (ML) use cases, wherein each of the plurality of AI/ML use cases is configured with an individual use case identifier (ID); receiving, from the UE, a maximum number of AI/ML active functionalities that the UE supports across the plurality of AI/ML use cases; generating a report configuration based on the plurality of properties of supported functionalities for each of the plurality of AI/ML use cases and the maximum number of AI/ML active functionalities that the UE supports across the plurality of AI/ML use cases; and transmitting, to the UE, the report configuration, as disclosed by Cheng, and receiving, from the UE, the report to the base station, as taught by Leng. The motivation to do so would have been to have a method for a base station to receive from a UE ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE, transmit a configuration for ML models and features for the UE to run, monitor and for which to report assistance information including ML model monitoring results, receive a report with the assistance information from the UE so that the base station can use the assistance information to evaluate the accuracy of the selected ML model and functionalities to determine if the base station should transmit a report reconfiguration to the UE to adapt the ML models and functionalities that are active at the UE in order to improve performance.
Claim 7 is rejected under 35 U.S.C. 103 as being unpatentable over Cheng et al. (WO 2023/211343)(herein after “Cheng”) in view of Leng et al. (US 2024/0098533)(herein after “Leng”), as applied to claim 1, and further in view of Wang et al. (WO 2025/091459)(herein after “Wang”).
Regarding claim 7, Cheng in view of Leng disclose the method of claim 1.
Cheng fails to disclose but Wang further teaches wherein the plurality of properties of supported functionalities for each of the plurality of AI/ML use cases includes properties of at least one of an observation window or a prediction window different beam sets for an AI/ML channel state information (CSI) report use case ([0002] & [0087]-[0091]disclose a prediction window for reporting time beam prediction may be configured by a network device according to capabilities of a UE as part of an AI/ML configuration for a CSI prediction use case that uses the prediction window to perform beam management using multiple CIS-RS over the prediction window.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have the method of claim 1, as disclosed by Cheng in view of Leng, wherein the plurality of properties of supported functionalities for each of the plurality of AI/ML use cases includes properties of at least one of an observation window or a prediction window different beam sets for an AI/ML channel state information (CSI) report use case, as further taught by Wang. The motivation to do so would have been to have a method for a UE to transmit to a base station ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE for a CSI prediction use case for beam management, receive a configuration for ML models and features for the UE to run, monitor and for which to report assistance information including ML model monitoring results using CSI-RS transmissions over the prediction window, generate and transmit a report with the assistance information to the base station so that the base station can use the assistance information to evaluate the accuracy of the selected ML model and functionalities for beam management to determine if the base station should transmit a report reconfiguration to the UE to adapt the ML models and functionalities for beam management that are active at the UE in order to improve performance.
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Cheng et al. (WO 2023/211343)(herein after “Cheng”) in view of Leng et al. (US 2024/0098533)(herein after “Leng”), as applied to claim 1, and further in view of Kim et al. (US 2024/0214840)(herein after “Kim”).
Regarding claim 8, Cheng in view of Leng disclose the method of claim 1.
Cheng fails to disclose but Kim further teaches further comprising receiving, from the base station, an indication of inactive models ([0127] discloses a UE receiving, from a base station, model ID information about an AI/ML model being deactivated.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have the method of claim 1, as disclosed by Cheng in view of Leng, further comprising receiving, from the base station, an indication of inactive models, as further taught by Kim. The motivation to do so would have been to have a method for a UE to transmit to a base station ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE for a CSI prediction use case for beam management, receive a configuration for ML models and features for the UE to run, monitor and for which to report assistance information and indicating which ML models are inactive so that the UE can save battery life by not running ML inference routines for the indicated ML inactive models.
Claims 9-13 are rejected under 35 U.S.C. 103 as being unpatentable over Cheng et al. (WO 2023/211343)(herein after “Cheng”) in view of Leng et al. (US 2024/0098533)(herein after “Leng”) and Kim et al. (US 2024/0214840)(herein after “Kim”), as applied to claim 8, and further in view of Sun et al. (US 2025/0254549)(herein after “Sun”).
Regarding claim 9, Cheng in view of Leng and Kim disclose the method of claim 8.
Cheng fails to disclose but Sun further teaches further comprising receiving, from the base station, an indication of which of the inactive models for which the UE is to perform performance monitoring ([0062]-[0067]] discloses that multiple AI models may be configured at a terminal equipment by a network device that may be deactivated, and that the network device can configure transmit configuration parameters for monitoring the performance of one or multiple of the AI models that may be deactivated.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have the method of claim 8, as disclosed by Cheng in view of Leng and Kim, further comprising receiving, from the base station, an indication of which of the inactive models for which the UE is to perform performance monitoring, as further taught by Sun. The motivation to do so would have been to have a method for a UE to transmit to a base station ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE for a CSI prediction use case for beam management, receive a configuration for ML models and features for the UE to run, monitor and for which to report assistance information and indicating which ML models are inactive and for which of the inactive ML models the UE should perform performance monitoring, so that the UE can evaluate and report readiness of an inactive ML model for being activated to potentially improve performance.
Regarding claim 10, Cheng in view of Leng and Kim and Sun disclose the method of claim 9.
Cheng fails to disclose but Sun further teaches wherein the indication of which of the inactive models for which the UE is to perform the performance monitoring includes at least one model ID ([0069] discloses that when the configuration parameters for monitoring the performance of deactivated AI models are different for different deactivated AI models, then the network device may also transmit identifiers of the AI models corresponding to the different configuration parameters.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have the method of claim 9, as disclosed by Cheng in view of Leng and Kim and Sun, wherein the indication of which of the inactive models for which the UE is to perform the performance monitoring includes at least one model ID, as further taught by Sun. The motivation to do so would have been to have a method for a UE to transmit to a base station ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE for a CSI prediction use case for beam management, receive a configuration for ML models and features for the UE to run, monitor and for which to report assistance information and indicating which ML models are inactive and for which of the inactive ML models the UE should perform performance monitoring using identifiers for each inactive ML model to indicate specific performance monitoring parameters for each inactive ML model, so that the UE can separately evaluate and report readiness of different inactive ML models for being activated to potentially improve performance.
Regarding claim 11, Cheng in view of Leng and Kim and Sun disclose the method of claim 9.
Cheng fails to disclose but Sun further teaches further comprising receiving, from the base station, an indication of a performance metric type with which the UE is to perform the performance monitoring of the inactive models ([0072] discloses that the configuration parameters may include a threshold for performance metric, a filter coefficient for performance metric and/or a counter for counting a monitoring result of a performance metric.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have the method of claim 9, as disclosed by Cheng in view of Leng and Kim and Sun, further comprising receiving, from the base station, an indication of a performance metric type with which the UE is to perform the performance monitoring of the inactive models, as further taught by Sun. The motivation to do so would have been to have a method for a UE to transmit to a base station ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE for a CSI prediction use case for beam management, receive a configuration for ML models and features for the UE to run, monitor and for which to report assistance information and indicating which ML models are inactive and for which of the inactive ML models the UE should perform performance monitoring using identifiers for each inactive ML model to indicate specific performance monitoring parameters types for each inactive ML model, so that the UE can separately evaluate and report readiness based on the indicated performance monitoring parameter types of different inactive ML models for being activated to potentially improve performance.
Regarding claim 12, Cheng in view of Leng and Kim and Sun disclose the method of claim 11.
Cheng fails to disclose but Leng teaches wherein generating the report based on the report configuration comprises generating a channel state information (CSI) report including a performed metric corresponding to the performance metric type, based on the performance monitoring ([0093] discloses that the report of model monitoring assistance information may include AI/ML generated CSI feedback (i.e. a CSI report). [0081] disclose that, based on the performance monitoring, the UE reported assistance information may include CSI compression or CSI prediction metrics.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have the method of claim 11, as disclosed by Cheng in view of Leng and Kim and Sun, wherein generating the report based on the report configuration comprises generating a channel state information (CSI) report including a performed metric corresponding to the performance metric type, based on the performance monitoring, as further taught by Leng. The motivation to do so would have been to have a method for a UE to transmit to a base station ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE for a CSI prediction use case for beam management, receive a configuration for ML models and features for the UE to run, monitor and for which to report assistance information and indicating which ML models are inactive and for which of the inactive ML models the UE should perform performance monitoring using identifiers for each inactive ML model to indicate specific performance monitoring parameters metric types for CSI feedback in beamforming for each inactive ML model, so that the UE can separately evaluate and report readiness based on the indicated CSI metrics of different inactive ML models for being activated to potentially improve performance.
Cheng fails to disclose but Kim further teaches wherein the generating of the CSI report based on the performance monitoring is of the inactive models ([0127] discloses a UE receiving, from a base station, model ID information about an AI/ML model being deactivated.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have the method of claim 11, wherein generating the report based on the report configuration comprises generating a channel state information (CSI) report including a performed metric corresponding to the performance metric type, based on the performance monitoring, as disclosed by Cheng in view of Leng and Kim and Sun, wherein the generating of the CSI report based on the performance monitoring is of the inactive models, as further taught by Kim. The motivation to do so would have been to have a method for a UE to transmit to a base station ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE for a CSI prediction use case for beam management, receive a configuration for ML models and features for the UE to run, monitor and for which to report assistance information and indicating which ML models are inactive and for which of the inactive ML models the UE should perform performance monitoring using identifiers for each inactive ML model to indicate specific performance monitoring parameters metric types for CSI feedback in beamforming for each inactive ML model, so that the UE can separately evaluate and report readiness based on the indicated CSI metrics of different inactive ML models for being activated to potentially improve performance.
Regarding claim 13, Cheng in view of Leng and Kim and Sun disclose the method of claim 12.
Cheng fails to disclose but Leng further teaches wherein the CSI report further includes indices of a predetermined number of best models and their corresponding monitored metrics ([[0111]-[0113] discloses that assistance information for CSI feedback reported by the UE may include the best N (i.e. a predetermined number of) beams with the N highest L1-RSRP/L1-SINR, and evaluation offsets or prediction accuracy based on comparing the N best beams with the N highest L1-RSRP/L1-SINR to the N predicted values, and that evaluation results are reported along with the CSI feedback.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have the method of claim 12, as disclosed by Cheng in view of Leng and Kim and Sun, wherein the CSI report further includes indices of a predetermined number of best models and their corresponding monitored metrics, as further taught by Leng. The motivation to do so would have been to have a method for a UE to transmit to a base station ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE for a CSI prediction use case for beam management, receive a configuration for ML models and features for the UE to run, monitor and for which to report assistance information and indicating which ML models are inactive and for which of the inactive ML models the UE should perform performance monitoring using identifiers for each inactive ML model to indicate the best inactive ML beam predicting models based on comparing RSRP/SINR measured metrics to predicted values, so that the UE can separately evaluate and report readiness and the corresponding ML beam predicting evaluations of different inactive ML models for being activated to potentially improve performance.
Claims 14 & 15 are rejected under 35 U.S.C. 103 as being unpatentable over Cheng et al. (WO 2023/211343)(herein after “Cheng”) in view of Leng et al. (US 2024/0098533)(herein after “Leng”) and Kim et al. (US 2024/0214840)(herein after “Kim”) and Sun et al. (US 2025/0254549)(herein after “Sun”) as applied to claim 12, and further in view of Abdelghaffar et al. (US 2025/0167851)(herein after “Abdelghaffar”).
Regarding claim 14, Cheng in view of Leng and Kim and Sun disclose the method of claim 12.
Cheng fails to disclose but Abdelghaffar further teaches wherein the CSI report is counted towards a CSI processing unit (CPU) limit ([0022] & [0118] disclose counting a number of CSI processing units (CPUs) for processing a CSI report based on the report quantity towards a maximum CPU count limit.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have the method of claim 12, as disclosed by Cheng in view of Leng and Kim and Sun, wherein the CSI report is counted towards a CSI processing unit (CPU) limit, as further taught by Abdelghaffar. The motivation to do so would have been to have a method for a UE to transmit to a base station ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE for a CSI prediction use case for beam management, receive a configuration for ML models and features for the UE to run, monitor and for which to report assistance information and indicating which ML models are inactive and for which of the inactive ML models the UE should perform and report CSI feedback performance monitoring, wherein central processing units (CPUs) are counted for each CSI report towards a maximum CPU count limit, so that if the next CSI feedback report increasing the CPU count above the maximum CPU count limit, the UE can drop the measuring and reporting of certain metrics in future CSI reports to reduce processing load of the CSI reporting in order to avoid not being able to send a CSI report due to processing limitation of the UE.
Regarding claim 15, Cheng in view of Leng and Kim and Sun and Abdelghaffar disclose the method of claim 14.
Cheng fails to disclose but Abdelghaffar further teaches therein the CPU limit is based on a UE capability or a predetermined value ([0098] discloses that the total (i.e. maximum) number of CPUs corresponds to the UE capability.).
Therefore, it would have been obvious to someone having ordinary skill in the art prior to the effective filing date of the claimed invention to have the method of claim 14, as disclosed by Cheng in view of Leng and Kim and Sun and Abdelghaffar, wherein the CPU limit is based on a UE capability or a predetermined value, as further taught by Abdelghaffar. The motivation to do so would have been to have a method for a UE to transmit to a base station ML models, functionalities and dependencies amongst the ML models that can maximally be supported by the UE for a CSI prediction use case for beam management, receive a configuration for ML models and features for the UE to run, monitor and for which to report assistance information and indicating which ML models are inactive and for which of the inactive ML models the UE should perform and report CSI feedback performance monitoring, wherein central processing units (CPUs) are counted for each CSI report towards a maximum CPU count limit based on a processing capability of the UE, so that if the next CSI feedback report increasing the CPU count above the maximum CPU count limit, the UE can drop the measuring and reporting of certain metrics in future CSI reports to reduce processing load of the CSI reporting in order to avoid not being able to send a CSI report due to processing limitation of the UE’s capabilities.
Conclusion
The following prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Wang et al. US (2026/0222837) discloses a Method and Apparatus of Supporting Artificial Intelligence.
Ali et al. (US 2025/0056211) discloses Capability Information Transmission.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JAMES P SEYMOUR whose telephone number is (571)272-7654. The examiner can normally be reached M-F 8-5 EST.
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, Nishant Divecha can be reached at 571-270-3125. 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.
/JAMES P SEYMOUR/Examiner, Art Unit 2419
/Nishant Divecha/Supervisory Patent Examiner, Art Unit 2419
Prosecution Insights