Detailed Action
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 7/30/2026 has been entered.
Claims 1, 3-13, and 15-20 are pending and ready for examination.
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 Examiner has reviewed the Applicant’s arguments submitted on 7/28/2026 in their entirety. Applicant’s argument are moot based on a new prior art rejection comprising Lin (US 20210158145). Lin as furnished in the new prior art rejection teaches the subject matter provided in Applicant’s newly amended independent claims.
The Examiner notes prior to the effective filing date of the claimed invention:
incorporating inference with respect to machine learning was a known technique to one of ordinary skill in the art as evidenced by Ryden (US 20240187991) (see e.g. Fig. 6)
Utilizing energy profiles was a known technique to one of ordinary skill in the art as evidenced by Mathur (US 2021/0376604) :
[0015] The method may comprise: [0016] determining a set of execution strategies, wherein an execution strategy specifies, for respective tasks of a processing pipeline, an associated processing device for executing said task, [0017] determining an energy profile associated with the respective execution strategies as a function of said energy demand data, wherein the energy profile associated with an execution strategy specifies, for each processing device, the energy that would be consumed in case of applying the execution strategy, [0018] selecting an execution strategy as a function of said battery availability data and of said energy profiles.
Sarferaz (US 2021/0241168) teaches:
[0070] Consumer 510 can monitor, at step 515, the use of the deployed ML model. During monitoring, consumer 510 can evaluate whether the deployed ML model is outputting sufficient results (e.g., evaluate the quality of productively used models at runtime). Monitoring provides consumers 510 with a comprehensive view of the ML model. The ML model contains a central cockpit (or centralized component) for various ML scenarios and can provide consumer 510 with various indications such as model status, accuracy KPIs, state of inference calls, or amount of processed data. In some cases, the deployed ML model can experience technical issues or connectivity issues. Consumer 510 can monitor for such issues. There may come a time when the trained ML model no longer outputs accurate results and requires re-training. The ML models are consuming historical data to learn from, but as time progresses, new data is generated that can be used to re-train the ML model. In case of issues, alerts can be provided to consumer 510 (e.g., via a UI) to inform the consumer 510 to act. In some variations, solution proposals can be recommended to administrators for resolving issues as well as statistics (e.g., number of errors, resources consumption or costs).
Barta (US 2012/032947) teaches energy profiles ([0049], [0065]/
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 1, 4- 8, 10, 12-13, and 16-20 are rejected under 35 USC 103 as being unpatentable over 3GPP TR 28.908, December 22 in view of Lin (US 20210158145)
Regarding claim 1, 3GPP discloses an apparatus for use by a communication network element or communication network function acting as an artificial intelligence, Al, machine learning, ML, management service consumer, the apparatus comprising
at least one processing circuitry (3GPP; 3GPP discloses a network computing environment which requires processing circuitry (e.g. processor, ASIC, etc.) to be inherently present), and
at least one memory for storing instructions that, when executed by the at least one processor, cause the apparatus at least (3GPP; 3GPP discloses a network computing environment which requires memory storing instructions to be inherently present)
to request an AI/ML energy consumption parameter from an AI/ML management service producer offering services related to at least one Al/ML entity (3GPP;
PNG
media_image1.png
311
878
media_image1.png
Greyscale
see e.g. Section 5.8.2.1 comprising Figure 5.8.2.1-1 which shows that the consumer may wish to obtain AI/ML capabilities to determine how to use them (for the consumer’s needs, e.g., for its intent targets or other automation targets;
see e.g. Section 5.8.3 “... capabilities of an ML entity as a decision described as a triplet <object(S), parameters, metrics> with the entries respectively indicating the object types for which the ML entity an undertake optimization or control ...”
The Examiner note capability is equivalent to energy consumption related parameter)
to receive, from the Al/ML management service producer, the requested Al/ML energy consumption parameter (3GPP;
see e.g. Section 5.8.2.1 comprising Figure 5.8.2.1-1 which shows that the consumer may wish to obtain AI/ML capabilities to determine how to use them (for the consumer’s needs, e.g., for its intent targets or other automation targets;
Section 5.8.2.1 teaches the consumer use the capability information to make decisions aligned with their intent or automation targets which is the processing or usage of the energy consumption parameter
see e.g. Section 5.8.3
The triplet structure <object(s), parameters, metrics> in Section 5.8.3 directly maps to object types, parameters (i.e. energy consuption related configuration parameters, as explicitly supported earlier in Section 5.8.2.1 and Figure 5.8.2.1-1. Section 5.8.3 requires that the producer deliver such information as part of the triplet of objects, parameters, and metrics. Because the consumer obtains this report “to determine how to use” the capability (Section 5.8.2.1), the act of determining the requested energy-consumption parameter for at least one mL entity is inherently present in the disclosed workflow) , and
to process the Al/ML energy consumption parameter for deriving an energy saving strategy considering the Al/ML energy consumption related parameter (3GPP;
Sections 5.8.2.1, 5.8.3, 5.8.4, 5.8.5
PNG
media_image2.png
564
1136
media_image2.png
Greyscale
PNG
media_image3.png
1049
1104
media_image3.png
Greyscale
PNG
media_image4.png
242
1105
media_image4.png
Greyscale
The Examiner notes the determination is within the context of energy consumption
See e.g. Section 5.10 AI/ML Configuration Management: 5.10.1 Description
“... use AI to formulate energy saving solutions ...”
See e.g. 5.10.2.5 “... per the performance of the energy efficiency result by execution of the inference output ...”
Under the Broadest Reasonable Interpretation, the claimed invention is anticipated. The cited figures and sections disclose that the consumer evaluates AI/ML capability data comprised of configuration and performance metrics (e.g., Z in the triplet), which inherently include parameters that may relate to energy consumption. These metrics are used by the ML entity to “optimize through its actions” which is functionally equivalent to processing the energy parameter to derive an energy saving strategy.
Furthermore, Section 5.8.5 confirms that the AI/ML capability data (triplet X,Y,Z) is returned to the consumer either via discovery MNS or the direct reading from the MNI, thereby supporting the consumer’s processing behavior. The strategy derivation in inherently present, as the entire framework is designed to inform the consumer’s decisions and optimization behavior, consisted with the claimed function.)
3GGP strongly suggests but does not expressly disclose:
wherein the AI/ML energy consumption parameter indicates energy consumption of the AI/ML entity per phase for at least one phase in a lifecycle of the AI/ML entity including a data collection phase, a model training phase, a testing phase and an inference phase, and
when processing the AI/ML energy consumption parameter for deriving the energy saving strategy considering the AI/ML energy consumption parameter, to consider information related to at least one of network performance or user targets.
Lin discloses:
wherein the AI/ML energy consumption parameter indicates energy consumption of the AI/ML entity per phase for at least one phase in a lifecycle of the AI/ML entity including a data collection phase, a model training phase, a testing phase and an inference phase (Lin; Lin teaches the determination of an energy loss associated with each processing layer of the ML model according to whether the respective layer is processed or skipped ([0047]), Fig. 1). Thus rather than merely providing an individual energy value, Lin characterizes the ML model’s energy consumption across its layer-execution configuration according to which portions of the model are executed, thereby providing the claimed AI/ML energy consumption profile. Line further expressly defines E(W,G) as an energy cost, where W represents the base-model parameters and G represents gating parameters controlling execution of portions of the model ([0053]), thereby providing the corresponding Ai/ML energy consumption metric. Lin further teaches processing layers to generate an inference ({0082]) and determines the energy loss based on the subset of the processing layers used to process the input data ([0084]). Accordingly, because Lin’s energy consumption profile characterizes which portions of the AI/ML model are executed and its energy metric quantifies the corresponding energy cost/loss, and those processing layers are executed t generate the inference, the AI/ML energy consumption parameter indicates energy consumption of the AI/ML entity for at least the in inference phase of the AI/ML entity’s lifecycle, as claimed; see e.g. Fig. 3)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Lin’s scheme. The motivation being the combined solution provides for implementing a known technique resulting in increased efficiencies of energy data processing.
3GPP in view of Lin discloses:
when processing the AI/ML energy consumption parameter for deriving the energy saving strategy considering the AI/ML energy consumption parameter, to consider information related to at least one of network performance or user targets (The combined solution as Lin teaches determining an energy loss associated with operation of the ML model and optimizing the model based on the energy loss ([0051]- [0055], [0084] [0085]), such that the AI/ML energy consumption information is processed to reduce the energy cost associated with operation of the ML model. 3GPP (Section 5.2.1 and Fig. 5.8.2.1-1 further teach that the AI/ML MnS Consumer obtains AI/ML capabilities to determine how to use the AI/ML entity for the consumer’s needs, including its “intent targets”, while Section 5.3 teaches that the capability of the ML entity is characterized by objects the ML entity optimizes or controls, parameters it optimizes or controls, and metrics it optimizes through its actions. Accordingly, in the combined solution, when Lin’s AI/ML energy consumption information is processed to derive the energy saving optimization, 3GPP’s consumer target information is additionally considered in determining how the AI/ML entity is to be used and optimize, thereby deriving the energy savings strategy considering both the AI/ML energy consumption parameter and information related to the claimed user targets (Figure 5.8.21-1))
Regarding claim 4, 3GPP in view of Lin discloses The apparatus according to claim 1, wherein the Al/ML energy consumption metric is defined per Al/ML capability including a definition of an object or object types for optimization or control (3GPP; Section 5.83 defines the capability of an ML entity for a decision using the triplet and identifies the object(s)/object type(S) that the ML entity can optimize or control), configuration parameters on an object or object types (3GPP; The triplet separately identifies the configuration parameters on the object(s)/object type(s) that the ML entity can optimize or control. The configuration parameters identify the parameters of that object that are optimized/controlled), and network metrics being optimized (3GPP; Section 5.8.3 “... network metrics ...’), and per Al/ML scope indicating a validity range (3GPP; Section 5.7.2.2 “... validity scope ...”)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Lin’s scheme. The motivation being the combined solution provides for implementing a known technique resulting in increased efficiencies of energy data processing.
Regarding claim 5, 3GPP in view of Lin discloses the apparatus according to claim 1, wherein the Al/ML energy consumption metric is defined per Al/ML entity (3GPP;
See e.g. Section 5.17.2.1 “... quantitative metric ... ML entity performance by evaluating the effect on ML entity ...”)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Lin’s scheme. The motivation being the combined solution provides for implementing a known technique resulting in increased efficiencies of energy data processing
Regarding claim 6, 3GPP in view of Lin discloses the apparatus according to claim1, wherein the Al/ML energy consumption metric is defined per managed element or managed entity each comprising one or more Al/ML entities (3GPP; The metrics may be based on a managed environment (3GPP;
See e.g. Section 5.4.2.2 “ ... operator may wish to control and manage ...ML entity ...”);
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Lin’s scheme. The motivation being the combined solution provides for implementing a known technique resulting in increased efficiencies of energy data processing
Regarding claim 7, 3GPP in view of Lin disclose the apparatus according to claim 1, Styles does not expressly disclose wherein the AI/ML energy consumption metric is defined separately for AI/ML related signaling concerning at least one of transmission of data to be processed by an AI/ML model, or transmission of data comprising an AI/ML model.
The above feature is considered to be Applicant Admitted Prior Art and it would have been obvious to one of ordinary skill in the art to implement this feature. The motivation being the combined solution provides for increased efficiencies in energy consumption optimization..
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Lin’s scheme. The motivation being the combined solution provides for implementing a known technique resulting in increased efficiencies of energy data processing
.
Regarding claim 8, 3GPP in view of Lin discloses The apparatus according to claim 1, wherein the Al/ML energy consumption metric is defined according to a requested energy consumption requirement provided from the consumer to the producer (3GPP; Per independent claim 1).
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Lin’s scheme. The motivation being the combined solution provides for implementing a known technique resulting in increased efficiencies of energy data processing
Regarding claim 10, 3GPP in view of Lin disclose The apparatus according to claim 1, to decide on at least one of a recommendation regarding activation/deactivation of the AI/ML entity a recommendation regarding switching over towards another AI/ML based solution based on performance or energy consumption, a recommendation regarding switching over towards a less energy demanding solution being different to an AI/ML based solution, a recommendation is derived regarding switching over towards a more energy demanding solution in order to improve performance of AI/ML solution, and a recommendation regarding a scheduling of a re-training of AI/ML entity (Because the combined solution applies AI/ML techniques to evaluate performance and energy-related metrics (Lin) and operates within an AI/ML framework that includes training-related functionality and signaling (3GPPP), the control of the AI/ML entity necessarily includes decision regarding its training and updating. Accordingly, when the system determines actions for the AI/ML entity based on those evaluated conditions, it would have been obvious for those actions to include a recommendation regarding scheduling of re-training of the AI/ML entity as part of managing and improving the model based on the same evaluated performance and energy conditions as training is part of machine learning. Furthermore execution of training by a processor necessarily occurs under instruction controlled timing/order)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Lin’s scheme. The motivation being the combined solution provides for implementing a known technique resulting in increased efficiencies of energy data processing
Regarding claim 12, 3GPP in view of Lin discloses The apparatus according to claim 1, wherein the instructions further cause the apparatus to send, to the AI/ML management service producer, at least one of a result of the processing the AI/ML energy consumption parameter for deriving the energy saving strategy or configuration data for configuring the energy saving strategy at the AI/ML management service producer (The combined solution as Lin processes the AI/ML energy consumption information to generate the energy aware optimization result, while 3GPP Section 5.8.2.1 and Fig 5.8.2.1-1 provide the AI/ML management framework in which the AI/ML producer/ML entity through requests and reporting concerning the AI/ML entity and its capabilities. Thus, when Lin’s energy consumption processing is incorporated into 3GPP’s consumer/producer management interaction, the resulting energy saving processing result is information exchanged by the consumer with producer for management of hted AI/ML entity. Accordingly, the combined solution provides for sending to the AI/ML management service producer the result of processing the AI/ML energy consumption parameter for deriving the energy saving strategy as claimed))
Regarding claim 13, claim 13 comprises the same and/or similar subject matter as claim 1 and is considered an obvious variation; therefore it is rejected under the same rationale.
Regarding claim 16, claim 16 comprises the same and/or similar subject matter as claim 4 and is considered an obvious variation; therefore it is rejected under the same rationale.
Regarding claim 17, claim 17 comprises the same and/or similar subject matter as claim 5 and is considered an obvious variation; therefore it is rejected under the same rationale.
Regarding claim 18, claim 18 comprises the same and/or similar subject matter as claim 6 and is considered an obvious variation; therefore it is rejected under the same rationale.
Regarding claim 19, claim 19 comprises the same and/or similar subject matter as claim 7 and is considered an obvious variation; therefore it is rejected under the same rationale.
Regarding claim 20, claim 20 comprises the same and/or similar subject matter as claim 8 and is considered an obvious variation; therefore it is rejected under the same rationale.
Claims 3 and 15 are rejected under 35 USC 103 as being unpatentable over 3GPP in view of Lin and in further view of Thotan (US 2014/0324240)
Regarding claim 3, 3GPP in view of Lin discloses the apparatus according to claim 1, 3GPP does not expressly disclose wherein the Al/ML energy consumption metric is expressed in a form of numerator / denominator, wherein the numerator indicates one of
a complexity related value indicating a computational complexity of an Al/ML model, an energy consumption related value indicating an energy consumption of an Al/ML model under predefined operation conditions, or an environmental related value indicating a carbon emission when operating an Al/ML model, and
the denominator indicates one of or a combination of per sample, per entire training process, per Al/ML related performance gain, per network KPI gain, per amount of saved energy, or per inference.
However in analogous art Thotan discloses:
wherein the Al/ML energy consumption metric is expressed in a form of numerator / denominator, wherein the numerator indicates one of
a complexity related value indicating a computational complexity of an Al/ML model,
the denominator indicates one of or a combination of per sample (Thotan;
see e.g. [0121] “ ... HMM ... computational complexity per data sample)
Therefore it would have been prima facie obvious before the effective filing date of the claimed invention to incorporate Thotan’s metric. The motivation being the combined solution provides for implanting a known technique providing for optimization of resources within the ML environment.
.
Regarding claim 15, claim 15 comprises the same and/or similar subject matter as claim 3 and is considered an obvious variation; therefore it is rejected under the same rationale.
Claim 9 is rejected under 35 USC 103 as being unpatentable over 3GPP in view of Lin and in further view of Khalil et al. “Energy Efficiency in 5G Communications – Conventional to Machine Learning Approaches”, April 2020
Regarding claim 9, 3GPP in view of Lin disclose the apparatus according to claim 1, 3GPP does not expressly disclose wherein the AI/ML energy consumption profile indicates an expected energy consumption during at least one phase of the lifecycle of an AI/ML entity and includes at least one of an indication of a complexity of the AI/ML entity, an indication of energy consumption of the AI/ML entity when operated under predefined operation conditions, an indication of an alternative energy saving solution based on an AI/ML model, an amount of data required for inference, an amount of data required for model training purposes, an indication of a required periodicity for re-training of the AI/ML entity, an indication of a requirement for data signaling regarding model training purposes.
However in analogous art Khalil discloses:
an indication of energy consumption of the AI/ML entity when operated under predefined operation conditions (Khalil; Khalil monitors energy efficiency within a predefine environment;
PNG
media_image5.png
658
677
media_image5.png
Greyscale
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Khalils’s scheme. The motivation being implanting a known technique provides for increased efficiencies in networking transmissions.
Claim 11 is rejected under 35 USC 103 as being unpatentable over 3GPP in view of Lin and in further view of Bellamkonda (US 20220232399)
Regarding claim 11, 3GPP in view of Lin disclose The apparatus according to claim 10, 3GPP does not expressly disclose:
wherein the instructions further cause the apparatus when processing the AI/ML energy consumption parameter for deriving the energy saving strategy considering the AI/ML energy consumption parameter, to consider information related to network performance in a tradeoff between an improvement in network performance and an energy consumption when applying the AI/ML entity, or to consider information related to a performance of the AI/ML entity in a tradeoff between AI/ML entity performance improvement and an energy consumption when improving the AI/ML entity
Bellamkonda discloses:
to consider information related to network performance in a tradeoff between an improvement in network performance and an energy consumption (Bellamkonda;
see e.g. [0015], [0020], [0022], [0023], [0030], [0035], [0036])
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Bellamkonda’s scheme. The motivation being the combined solution provides for implementing a known technique.
3GPP in view of Lin and in further view of Bellamkonda disclose:
wherein the instructions further cause the apparatus when processing the AI/ML energy consumption parameter for deriving the energy saving strategy considering the AI/ML energy consumption parameter, to consider information related to network performance in a tradeoff between an improvement in network performance and an energy consumption when applying the AI/ML entity ((The combined solution because 3GPP already discloses AI/ML processing using parameters and performance metrics to optimize outcomes and Bellamkonda discloses evaluating energy consumption related parameters together with QoS/KPI metrics to select andn prioritize actions based on an optimization score. Thus the combined system necessarily performs a tradeoff between network performance (e.g. QoS/KPIs) and energy consumption when determining actions for the AI/ML entity, which reads on considering information related to network performance in a tradeoff with energy consumption when applying or improving the AI/ML entity)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Bellamkonda with 3GPP, the motivation being the combined solution provides for implementing a known technique resulting in increased efficiency and improve optimization of AI/ML based decision making by incorporating energy consumption-related parameters together with network performance information within a unified optimization framework.
Any inquiry concerning this communication or earlier communications from the Examiner should be directed to TODD L. BARKER whose telephone number is (571) 270 0257. The Examiner can normally be reached on Monday through Friday, 7:30am to 5:00pm.
If attempts to reach the Examiner by telephone are unsuccessful, the Examiner's supervisor Vivek Srivastava can be reached on (571) 272 7304.
/TODD L BARKER/Primary Examiner, Art Unit 2449
