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 . This action is in response to the communication filed on 01/05/2026. Claims 1-20 are pending in this application.
Response to Arguments
Applicant’s arguments filed 01/05/2026 have been fully considered but they are not persuasive. Applicant argues:
a.
Applicant states that “However, Bedekar does not disclose the information that the RAN node feeds back to the controller, the resources used by the RAN node for running the neural network configured by the controller, nor the information for adjusting the parameters of the neural network configured by the controller. Bedekar cannot adjust the parameters of the neural network configured by the controller. Therefore, Bedekar does not disclose the second processing resource information, or the target algorithm recommendation information, as recited by amended claim 1 (Reply, page 16).”
a.
Examiner respectfully disagrees. Bedekar exemplifies the interaction process between a RAN node and a controller node in FIG. 3 for initial determination of the neural network configuration (step 318) and updating the neural network configuration (step 326) which is applied in the RAN node. Further on, paragraph [0072] of Bedekar corresponding to FIG. 2 uses two “subsequently” clauses to disclose the interaction process of the controller node providing the RAN node the neural network configuration, the RAN node updating the measurement information and its currently available resources information to the controller node, and then the controller node providing the updated neural network configuration to the RAN node. And the neural network configuration may include a type of neural network and/or neural network parameters or settings. Therefore, Bedekar continues to teach the amended independent claims 1, 11 and 17.
Response to Amendment
The claim objection to claim 7 is now withdrawn in view of the claim amendments.
Claim Rejections - 35 USC § 102
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 the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(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 1-20 is/are rejected under 35 U.S.C. 102 (a)(2) as being anticipated by Bedekar et al. (US 20210385682 A1, priority dated 02/19/2019; hereinafter Bedekar).
For Claim 1, Bedekar teaches an information interaction method, comprising:
sending, by a first communication device (Bedekar exemplifies RAN node 212 in FIG. 3), first processing resource information to a second communication device (Bedekar discloses the RAN node sending RAN capabilities information to the controller node, as exemplified in step 312 of FIG. 3; Examiner notes that the limitation “a second communication device” is not referred to in the remaining parts of Claim 1 or in the dependent claims of Claim 1, and is not differentiated from the later recited limitation “at least one target communication device,” the limitation “a second communication device” is therefore mapped to the Controller 210 exemplified in FIG. 3; ¶ 0073 “… At 312, the RAN node 212 sends to controller 210, its RAN capabilities, which may include neural network support information, e.g., to indicate the RAN node's capabilities or support for neural networks, for example. Thus, the RAN capabilities at 312 may include, for example, neural network support information, e.g., which may include neural network capability information and hardware information for the RAN node. Thus, for example, at 312, the RAN capabilities may include or indicate, by way of example, a list of supported neural network (NN) types, a list of functions (e.g., RAN functions) that can be supported or performed by the RAN node 212 with a neural network, and/or a list of hardware capabilities (e.g., hardware features) of the RAN node …”); and
performing, by the first communication device, information interaction with at least one target communication device (Bedekar exemplifies Controller node 210 in FIG. 3) according to the first processing resource information (Bedekar discloses the controller node exchanging information with the RAN node, determining and transmitting the types and configuration of neural network to the RAN node based on the RAN capabilities information, as exemplified in steps 314-326 of FIG. 3; ¶ 0076 “… at 316, the RAN node 212 may send controller 210 hardware availability information that indicates an availability of the hardware (e.g., processors or processor cores, memory, ...) of the RAN node to be used for neural network processing …”; ¶ 0078 “… At 320, the controller 210 sends to RAN node 212, neural network configuration information that indicates the configuration of the neural network for the RAN node 212 (e.g., including the neural network type and/or settings or parameters of the neural network) …”; also see the disclosure corresponding to the steps in FIG. 3),
wherein the first processing resource information is used for indicating at least one of the following:
available resource information of the first communication device for running a target algorithm, the target algorithm comprising at least one of a neural network algorithm, an artificial intelligence algorithm, and a machine learning algorithm (Bedekar, FIG. 3; ¶ 0073 “… At 312, the RAN node 212 sends to controller 210, its RAN capabilities, which may include neural network support information, e.g., to indicate the RAN node's capabilities or support for neural networks, for example. Thus, the RAN capabilities at 312 may include, for example, neural network support information, e.g., which may include neural network capability information and hardware information for the RAN node. Thus, for example, at 312, the RAN capabilities may include or indicate, by way of example, a list of supported neural network (NN) types, a list of functions (e.g., RAN functions) that can be supported or performed by the RAN node 212 with a neural network, and/or a list of hardware capabilities (e.g., hardware features) of the RAN node …”); and
a target algorithm set supported by the first communication device (Bedekar discloses a list of supported NN types in ¶ 0073, as exemplified in step 312 of FIG. 3),
wherein the performing, by the first communication device, information interaction with at least one target communication device according to the first processing resource information further comprises:
sending, by the first communication device, first interaction information to the at least one target communication device according to configuration information of the target algorithm, wherein the first interaction information comprises at least one of second processing resource information (Bedekar discloses the RAN node sending the updated available resource information (e.g. hardware resources) to the controller node according to the currently applied neural network configuration; FIG. 2, FIG. 3; ¶ 0072 “… Also for example, subsequently, the RAN node 212 may continue to provide to Controller 210, the measurement information as requested (e.g., in a format requested, specific types of requested measurement information and/or with specific processing or pre-processing before sending the measurement information), e.g., which may include data elements or information describing updated conditions or metrics or measurements of (or with respect to) a cell and/or UEs within a cell(s) at the RAN node 212, or an indication of currently available resources (e.g., hardware resources) of the RAN node (such as an amount of processor or memory resources available for neural network processing), for example …”; ¶ 0079 “… Also, at 322, the RAN node 212 may send controller 210 updated hardware availability information that indicates an updated availability of the hardware (e.g., processors or processor cores, memory, ...) of the RAN node to be used for neural network processing …”) and target algorithm recommendation information,
the second processing resource information being used for indicating resource information of the first communication device for running the target algorithm (Bedekar discloses the RAN node sending the updated available resource information (e.g. hardware resources) to the controller node according to the currently applied neural network configuration; FIG. 2, FIG. 3; ¶ 0072, ¶ 0079).
For Claim 2, Bedekar teaches the method according to claim 1, wherein the performing, by the first communication device, information interaction with at least one target communication device according to the first processing resource information comprises:
receiving, by the first communication device, the configuration information of the target algorithm from the at least one target communication device (Bedekar discloses the RAN node receiving the NN configuration information from the controller node, as exemplified in steps 318 and 320 of FIG. 3; ¶ 0077 “… At 318, controller 210 determines a configuration of a neural network for the RAN node 212 (e.g., including selecting a neural network type and determining one or more parameters or settings of the neural network), based on the measurement information or data received from the RAN node 212. Thus, for example, at 318, the controller 210 may make an initial determination, for at least one RAN function, of a type of neural network to use, and a configuration of the neural network (e.g., inputs, output(s), input layers, hidden layers, number of neurons, activation functions, weights, and/or other neural network settings or parameters …”; ¶ 0078 “… At 320, the controller 210 sends to RAN node 212, neural network configuration information that indicates the configuration of the neural network for the RAN node 212 (e.g., including the neural network type and/or settings or parameters of the neural network) …”), wherein the configuration information comprises at least one of the following:
at least one expected time period for running the target algorithm;
an input type of the target algorithm (Bedekar discloses the configuration of the neural network including the inputs information in ¶ 0077, as exemplified in step 318 of FIG. 3);
preprocessing information corresponding to input information of the target algorithm;
different target parameters corresponding to neural network models of a same target algorithm type (Bedekar discloses various neural network settings or parameters including weights in ¶ 0077, as exemplified in step 318 of FIG. 3); and
a neural network model library.
For Claim 3, Bedekar teaches the method according to claim 2, further comprising:
obtaining update indication information from the at least one target communication device, wherein the update indication information is used for indicating to update the configuration information (Bedekar, FIG. 3; ¶ 0080 “… At 324, controller 210 may determine an updated (or adjusted) neural network configuration (e.g., which may include an updated neural network type and/or one or more adjusted or updated neural network settings or parameters), e.g., based on the updated or changed measurement information, data or hardware availability information …”; ¶ 0081 “… At 326, the controller 210 sends to RAN node 212, updated or adjusted neural network configuration information that indicates the updated or adjusted configuration of the neural network for the RAN node 212 (e.g., including the updated or adjusted neural network type and/or the updated or adjusted settings or parameters of the neural network for the RAN node 212 …”); and
updating the configuration information of the target algorithm according to the update indication information (Bedekar, FIG. 3; ¶ 0081 “… At 326, the controller 210 sends to RAN node 212, updated or adjusted neural network configuration information that indicates the updated or adjusted configuration of the neural network for the RAN node 212 (e.g., including the updated or adjusted neural network type and/or the updated or adjusted settings or parameters of the neural network for the RAN node 212 …”).
For Claim 4, Bedekar teaches the method according to claim 1, wherein the second processing resource information comprises at least one of the following:
a target algorithm supported or not supported by the first communication device;
whether the first communication device supports the target algorithm according to a predefined time period;
a time period required for running the target algorithm by the first communication device (Bedekar, FIG. 3, FIG. 4; ¶ 0107 “… RAN node may provide to the Controller over the API (e.g., one or more of): 1) A list of supported types of neural networks whose execution is supported by the RAN node … 2) A list of RAN functions for which it can execute a neural network to make decisions … 3) A list of HW capabilities descriptors describing the HW capabilities of the RAN node for executing neural networks, … available time cycles for executing time-critical tasks based on NN …”);
an expected time period corresponding to running the target algorithm by the first communication device; and
a target parameter of the target algorithm supported by the first communication device;
wherein the target algorithm recommendation information comprises at least one of the following:
recommendation information of the first communication device on the target parameter of the target algorithm;
revision information of the first communication device on the target parameter of the target algorithm; and
information about whether the first communication device supports the target algorithm (Bedekar discloses the RAN node 212 sending controller 210 updated hardware availability information in ¶ 0079, which is implicitly information about whether the RAN node can support the target algorithm).
For Claim 5, Bedekar teaches the method according to claim 4, wherein the sending first interaction information comprises: sending at least one of the second processing resource information and the target algorithm recommendation information in a case that a preconfigured trigger condition is met (Bedekar discloses the RAN node sending updated hardware availability information based on the instructions from the controller node; FIG. 3; ¶ 0079 “… At 322, the RAN node 212 may send to controller 210 updated measurement information, e.g., as requested by the instructions from the controller 210 … Also, at 322, the RAN node 212 may send controller 210 updated hardware availability information that indicates an updated availability of the hardware (e.g., processors or processor cores, memory, ...) of the RAN node to be used for neural network processing … ”);
wherein the trigger condition comprises at least one of the following:
a quantity of available resources of the first communication device is less than a preset quantity of resources;
available resources of the first communication device are incapable of supporting running of an indicated target algorithm;
the first communication device cannot perform processing in a manner in the first processing resource information or the second interaction information;
the available resources of the first communication device change or a range of changes in the available resources exceeds a threshold (Bedekar, FIG. 3; ¶ 0080 “… hardware resources at the RAN node 212 available for neural network processing may be significantly less than before, e.g., thus causing the controller 210 to change or adjust settings or parameters (e.g., to use fewer neurons or layers) of a neural network at the RAN node 212, or possibly to cancel the use of such neural network at the RAN node. …”; ¶ 0101 “… 2. RAN-to-Controller API: to enable RAN to provide to the controller: A. capabilities of the RAN, such as HW capabilities or the types of neural network (NN) supported or the types of functions for which NN can be used; B. data related to the NN or the operation of the relevant functions, such as metrics or measurements, as requested by the Controller according to the mode (periodic/event-driven, or streaming/batch-mode etc.) …”); and
the first communication device needs to simultaneously run a plurality of target algorithms.
For Claim 6, Bedekar teaches the method according to claim 1, wherein the first processing resource information comprises at least one of the following:
an activation function type supported by the first communication device (Bedekar, ¶ 0107 “… RAN node may provide to the Controller over the API (e.g., one or more of): 1) A list of supported types of neural networks whose execution is supported by the RAN node … This may further include parameters for each type of neural networks, such as: … type of activation functions supported …”);
an operation type supported by the first communication device (Bedekar, FIG. 3; ¶ 0073 “… at 312, the RAN capabilities may include or indicate, by way of example, a list of supported neural network (NN) types, a list of functions (e.g., RAN functions) that can be supported or performed by the RAN node 212 with a neural network, and/or a list of hardware capabilities (e.g., hardware features) of the RAN node …”);
a conversion relationship between available resources used between different operations supported by the first communication device; whether the first communication device supports different specific operations or hyperoperations within a same time unit; a combination relationship between the different specific operations supported by the first communication device; a quantity of operation times supported by the first communication device within a first target time period, the first target time period comprising at least one time unit; a relationship between the quantity of operation times supported within the first target time period and a quantity of operation times supported within a second target time period, the second target time period comprising at least one time unit, and the time unit comprised in the second target time period being different from the time unit comprised in the first target time period; a time period required for completing a specific operation of a specific target by the first communication device; information about the hyperoperation supported by the first communication device; information about a neural network model supported by the first communication device; a model size; a quantity of quantization bits of a target algorithm model; a quantity of layers of the model; a maximum quantity of neurons at each layer of the model; a running time period of each neural network algorithm; a running time period of a plurality of neural network algorithm combinations; recommendation information of the first communication device on the target parameter of the target algorithm; and support information of the first communication device for a pre-agreed target algorithm; indication information of a running priority of the target algorithm; a time period from configuration to taking effect of the neural network model; a time period required for switching between at least two neural network models; and whether a corresponding operation is runnable during switching between the at least two neural network models.
For Claim 7, Bedekar teaches the method according to claim 1, wherein the sending, by the first communication device, first processing resource information comprises: sending, by the first communication device, the first processing resource information in a case that a preconfigured trigger condition is met (Bedekar discloses the RAN node sending updated hardware availability information based on the instructions from the controller node; FIG. 3; ¶ 0079 “… At 322, the RAN node 212 may send to controller 210 updated measurement information, e.g., as requested by the instructions from the controller 210 … Also, at 322, the RAN node 212 may send controller 210 updated hardware availability information that indicates an updated availability of the hardware (e.g., processors or processor cores, memory, ...) of the RAN node to be used for neural network processing … ”);
wherein the trigger condition comprises at least one of the following:
a quantity of available resources of the first communication device is less than a preset quantity of resources;
available resources of the first communication device are incapable of supporting running of an indicated target algorithm;
the first communication device cannot perform processing in a manner in the first processing resource information or the second interaction information;
the available resources of the first communication device change or a range of changes in the available resources exceeds a threshold (Bedekar, FIG. 3; ¶ 0080 “… hardware resources at the RAN node 212 available for neural network processing may be significantly less than before, e.g., thus causing the controller 210 to change or adjust settings or parameters (e.g., to use fewer neurons or layers) of a neural network at the RAN node 212, or possibly to cancel the use of such neural network at the RAN node. …”; ¶ 0101 “… 2. RAN-to-Controller API: to enable RAN to provide to the controller: A. capabilities of the RAN, such as HW capabilities or the types of neural network (NN) supported or the types of functions for which NN can be used; B. data related to the NN or the operation of the relevant functions, such as metrics or measurements, as requested by the Controller according to the mode (periodic/event-driven, or streaming/batch-mode etc.) …”); and
the first communication device needs to simultaneously run a plurality of target algorithms.
For Claim 8, Bedekar teaches the method according to claim 1, further comprising:
sending, by the first communication device, at least one of second interaction information and third processing resource information in a case that a preconfigured trigger condition is met (Bedekar discloses the RAN node sending updated hardware availability information based on the instructions from the controller node, the information sent at steps 316 and/or 322 may be considered as “second interaction information” or “third processing resource information”; FIG. 3; ¶ 0079 “… At 322, the RAN node 212 may send to controller 210 updated measurement information, e.g., as requested by the instructions from the controller 210 … Also, at 322, the RAN node 212 may send controller 210 updated hardware availability information that indicates an updated availability of the hardware (e.g., processors or processor cores, memory, ...) of the RAN node to be used for neural network processing … ”),
wherein the second interaction information is used for indicating a sharing capability of a plurality of target algorithms for an NPU (Bedekar, FIG. 3; ¶ 0076 “… Also, at 316, the RAN node 212 may send controller 210 hardware availability information that indicates an availability of the hardware ( e.g., processors or processor cores, memory, ... ) of the RAN node to be used for neural network processing …”), and
the third processing resource information is used for indicating an available computing resource (Bedekar, FIG. 3; ¶ 0079 “… Also, at 322, the RAN node 212 may send controller 210 updated hardware availability information that indicates an updated availability of the hardware (e.g., processors or processor cores, memory, ...) of the RAN node to be used for neural network processing … ”);
wherein the trigger condition comprises at least one of the following:
a quantity of available resources of the first communication device is less than a preset quantity of resources;
available resources of the first communication device are incapable of supporting running of an indicated target algorithm;
the first communication device cannot perform processing in a manner in the first processing resource information or the second interaction information;
the available resources of the first communication device change or a range of changes in the available resources exceeds a threshold (Bedekar, FIG. 3; ¶ 0080 “… hardware resources at the RAN node 212 available for neural network processing may be significantly less than before, e.g., thus causing the controller 210 to change or adjust settings or parameters (e.g., to use fewer neurons or layers) of a neural network at the RAN node 212, or possibly to cancel the use of such neural network at the RAN node. …”; ¶ 0101 “… 2. RAN-to-Controller API: to enable RAN to provide to the controller: A. capabilities of the RAN, such as HW capabilities or the types of neural network (NN) supported or the types of functions for which NN can be used; B. data related to the NN or the operation of the relevant functions, such as metrics or measurements, as requested by the Controller according to the mode (periodic/event-driven, or streaming/batch-mode etc.) …”); and
the first communication device needs to simultaneously run a plurality of target algorithms.
For Claim 9, Bedekar teaches the method according to claim 1, wherein the performing, by the first communication device, information interaction with at least one target communication device according to the first processing resource information comprises:
obtaining first indication information from the at least one target communication device; and determining, according to the first indication information, a target algorithm used by the first communication device (Bedekar, FIG. 3; ¶ 0078 “… At 320, the controller 210 sends to RAN node 212, neural network configuration information that indicates the configuration of the neural network for the RAN node 212 (e.g., including the neural network type and/or settings or parameters of the neural network) …”),
wherein the first indication information comprises at least one of the following:
indication information of a running priority of the target algorithm;
a time period from configuration to taking effect of the neural network model (Bedekar, FIG. 3, FIG. 4; ¶ 0107 “… RAN node may provide to the Controller over the API (e.g., one or more of): 1) A list of supported types of neural networks whose execution is supported by the RAN node … 2) A list of RAN functions for which it can execute a neural network to make decisions … 3) A list of HW capabilities descriptors describing the HW capabilities of the RAN node for executing neural networks, … available time cycles for executing time-critical tasks based on NN …”);
a time period required for switching between at least two neural network models; and
whether a corresponding operation is runnable during switching between the at least two neural network models.
For Claim 10, Bedekar teaches the method according to claim 1, wherein the performing, by the first communication device, information interaction with at least one target communication device according to the first processing resource information comprises: obtaining second indication information from the at least one target communication device, wherein the second indication information is used for indicating a target algorithm used by the first communication device (Bedekar, FIG. 3; ¶ 0078 “… At 320, the controller 210 sends to RAN node 212, neural network configuration information that indicates the configuration of the neural network for the RAN node 212 (e.g., including the neural network type and/or settings or parameters of the neural network) …”; see also ¶ 0081).
For Claim 11, Bedekar teaches an information interaction method, comprising:
obtaining, by a second communication device (Bedekar exemplifies Controller node 210 in FIG. 3), first processing resource information sent by a first communication device (Bedekar discloses the controller node obtaining RAN capabilities information from the RAN node 212, as exemplified in step 312 of FIG. 3; ¶ 0073 “… At 312, the RAN node 212 sends to controller 210, its RAN capabilities, which may include neural network support information, e.g., to indicate the RAN node's capabilities or support for neural networks, for example. Thus, the RAN capabilities at 312 may include, for example, neural network support information, e.g., which may include neural network capability information and hardware information for the RAN node. Thus, for example, at 312, the RAN capabilities may include or indicate, by way of example, a list of supported neural network (NN) types, a list of functions (e.g., RAN functions) that can be supported or performed by the RAN node 212 with a neural network, and/or a list of hardware capabilities (e.g., hardware features) of the RAN node …”); and
performing, by the second communication device, interaction with a target communication device (Bedekar exemplifies RAN node 212 in FIG. 3) according to the first processing resource information, the target communication device comprising at least one of the first communication device and a third communication device (Bedekar discloses the controller node exchanging information with the RAN node, determining and transmitting the types and configuration of neural network to the RAN node based on the RAN capabilities information, as exemplified in steps 314-326 of FIG. 3; ¶ 0076 “… at 316, the RAN node 212 may send controller 210 hardware availability information that indicates an availability of the hardware (e.g., processors or processor cores, memory, ...) of the RAN node to be used for neural network processing …”; ¶ 0078 “… At 320, the controller 210 sends to RAN node 212, neural network configuration information that indicates the configuration of the neural network for the RAN node 212 (e.g., including the neural network type and/or settings or parameters of the neural network) …”; also see the disclosure corresponding to the steps in FIG. 3),
wherein the first processing resource information is used for indicating at least one of the following:
available resource information of the first communication device for running a target algorithm, the target algorithm comprising at least one of a neural network algorithm, an artificial intelligence algorithm, and a machine learning algorithm (Bedekar, FIG. 3; ¶ 0073 “… At 312, the RAN node 212 sends to controller 210, its RAN capabilities, which may include neural network support information, e.g., to indicate the RAN node's capabilities or support for neural networks, for example. Thus, the RAN capabilities at 312 may include, for example, neural network support information, e.g., which may include neural network capability information and hardware information for the RAN node. Thus, for example, at 312, the RAN capabilities may include or indicate, by way of example, a list of supported neural network (NN) types, a list of functions (e.g., RAN functions) that can be supported or performed by the RAN node 212 with a neural network, and/or a list of hardware capabilities (e.g., hardware features) of the RAN node …”); and
a target algorithm set supported by the first communication device (Bedekar discloses a list of supported NN types in ¶ 0073, as exemplified in step 312 of FIG. 3),
wherein the performing, by the second communication device, interaction with a target communication device according to the first processing resource information further comprises:
obtaining, by the second communication device, first interaction information sent by the target communication device, wherein the first interaction information comprises at least one of second processing resource information (Bedekar discloses the RAN node sending the updated available resource information (e.g. hardware resources) to the controller node according to the currently applied neural network configuration; FIG. 2, FIG. 3; ¶ 0072 “… Also for example, subsequently, the RAN node 212 may continue to provide to Controller 210, the measurement information as requested (e.g., in a format requested, specific types of requested measurement information and/or with specific processing or pre-processing before sending the measurement information), e.g., which may include data elements or information describing updated conditions or metrics or measurements of (or with respect to) a cell and/or UEs within a cell(s) at the RAN node 212, or an indication of currently available resources (e.g., hardware resources) of the RAN node (such as an amount of processor or memory resources available for neural network processing), for example …”; ¶ 0079 “… Also, at 322, the RAN node 212 may send controller 210 updated hardware availability information that indicates an updated availability of the hardware (e.g., processors or processor cores, memory, ...) of the RAN node to be used for neural network processing …”) and target algorithm recommendation information,
the second processing resource information being used for indicating resource information of the first communication device for running the target algorithm (Bedekar discloses the RAN node sending the updated available resource information (e.g. hardware resources) to the controller node according to the currently applied neural network configuration; FIG. 2, FIG. 3; ¶ 0072, ¶ 0079).
For Claim 12, Bedekar teaches the method according to claim 11, wherein the performing, by the second communication device, interaction with a target communication device according to the first processing resource information comprises:
sending, by the second communication device, configuration information of the target algorithm (Bedekar discloses the controller node sending the NN configuration information to the RAN node, as exemplified in steps 318 and 320 of FIG. 3; ¶ 0077 “… At 318, controller 210 determines a configuration of a neural network for the RAN node 212 (e.g., including selecting a neural network type and determining one or more parameters or settings of the neural network), based on the measurement information or data received from the RAN node 212. Thus, for example, at 318, the controller 210 may make an initial determination, for at least one RAN function, of a type of neural network to use, and a configuration of the neural network (e.g., inputs, output(s), input layers, hidden layers, number of neurons, activation functions, weights, and/or other neural network settings or parameters …”; ¶ 0078 “… At 320, the controller 210 sends to RAN node 212, neural network configuration information that indicates the configuration of the neural network for the RAN node 212 (e.g., including the neural network type and/or settings or parameters of the neural network) …”), wherein the configuration information comprises at least one of the following:
at least one expected time period for running the target algorithm;
an input type of the target algorithm (Bedekar discloses the configuration of the neural network including the inputs information in ¶ 0077, as exemplified in step 318 of FIG. 3);
preprocessing information corresponding to input information of the target algorithm;
different target parameters corresponding to neural network models of a same target algorithm type (Bedekar discloses various neural network settings or parameters including weights in ¶ 0077, as exemplified in step 318 of FIG. 3); and
a neural network model library.
For Claim 13, Bedekar teaches the method according to claim 11, wherein the second processing resource information comprises at least one of the following:
a target algorithm supported or not supported by the first communication device;
whether the first communication device supports the target algorithm according to a predefined time period;
a time period required for running the target algorithm by the first communication device (Bedekar, FIG. 3, FIG. 4; ¶ 0107 “… RAN node may provide to the Controller over the API (e.g., one or more of): 1) A list of supported types of neural networks whose execution is supported by the RAN node … 2) A list of RAN functions for which it can execute a neural network to make decisions … 3) A list of HW capabilities descriptors describing the HW capabilities of the RAN node for executing neural networks, … available time cycles for executing time-critical tasks based on NN …”);
an expected time period corresponding to running the target algorithm by the first communication device; and
a target parameter of the target algorithm supported by the first communication device;
wherein the target algorithm recommendation information comprises at least one of the following:
recommendation information of the first communication device on the target parameter of the target algorithm;
revision information of the first communication device on the target parameter of the target algorithm; and
information about whether the first communication device supports the target algorithm (Bedekar discloses the RAN node 212 sending controller 210 updated hardware availability information in ¶ 0079, which is implicitly information about whether the RAN node can support the target algorithm).
For Claim 14, Bedekar teaches the method according to claim 11, wherein the performing, by the second communication device, interaction with a target communication device according to the first processing resource information comprises:
sending first indication information (Bedekar, FIG. 3; ¶ 0078 “… At 320, the controller 210 sends to RAN node 212, neural network configuration information that indicates the configuration of the neural network for the RAN node 212 (e.g., including the neural network type and/or settings or parameters of the neural network) …”), wherein the first indication information comprises at least one of the following:
indication information of a running priority of the target algorithm;
a time period from configuration to taking effect of the neural network model (Bedekar, FIG. 3, FIG. 4; ¶ 0107 “… RAN node may provide to the Controller over the API (e.g., one or more of): 1) A list of supported types of neural networks whose execution is supported by the RAN node … 2) A list of RAN functions for which it can execute a neural network to make decisions … 3) A list of HW capabilities descriptors describing the HW capabilities of the RAN node for executing neural networks, … available time cycles for executing time-critical tasks based on NN …”);
a time period required for switching between at least two neural network models; and
whether a corresponding operation is runnable during switching between the at least two neural network models.
For Claim 15, Bedekar teaches the method according to claim 11, further comprising: sending second indication information, wherein the second indication information is used for indicating a target algorithm used by the first communication device (Bedekar, FIG. 3; ¶ 0078 “… At 320, the controller 210 sends to RAN node 212, neural network configuration information that indicates the configuration of the neural network for the RAN node 212 (e.g., including the neural network type and/or settings or parameters of the neural network) …”; see also ¶ 0081).
For Claim 16, Bedekar teaches the method according to claim 12, further comprising: sending update indication information, wherein the update indication information is used for indicating to update the configuration information (Bedekar, FIG. 3; ¶ 0080 “… At 324, controller 210 may determine an updated (or adjusted) neural network configuration (e.g., which may include an updated neural network type and/or one or more adjusted or updated neural network settings or parameters), e.g., based on the updated or changed measurement information, data or hardware availability information …”; ¶ 0081 “… At 326, the controller 210 sends to RAN node 212, updated or adjusted neural network configuration information that indicates the updated or adjusted configuration of the neural network for the RAN node 212 (e.g., including the updated or adjusted neural network type and/or the updated or adjusted settings or parameters of the neural network for the RAN node 212 …”).
For Claim 17, the claim is substantially similar to claim 1 and therefore is rejected for the same reasoning set forth above. Additionally, Bedekar teaches a communication device, comprising a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing (Bedekar, FIG. 7; ¶ 0168 “… The wireless station also includes a processor or control unit/entity (controller) 1004 to execute instructions or software and control transmission and receptions of signals, and a memory 1006 to store data and/or instructions …” ).
For Claim 18, the claim is substantially similar to claim 2 and therefore is rejected for the same reasoning set forth above.
For Claim 19, Bedekar teaches the communication device according to claim 17, wherein the second processing resource information comprises at least one of the following:
a target algorithm supported or not supported by the first communication device;
whether the first communication device supports the target algorithm according to a predefined time period;
a time period required for running the target algorithm by the first communication device (Bedekar, FIG. 3, FIG. 4; ¶ 0107 “… RAN node may provide to the Controller over the API (e.g., one or more of): 1) A list of supported types of neural networks whose execution is supported by the RAN node … 2) A list of RAN functions for which it can execute a neural network to make decisions … 3) A list of HW capabilities descriptors describing the HW capabilities of the RAN node for executing neural networks, … available time cycles for executing time-critical tasks based on NN …”);
an expected time period corresponding to running the target algorithm by the first communication device; and
a target parameter of the target algorithm supported by the first communication device;
wherein the target algorithm recommendation information comprises at least one of the following: recommendation information of the first communication device on the target parameter of the target algorithm;
revision information of the first communication device on the target parameter of the target algorithm; and
information about whether the first communication device supports the target algorithm (Bedekar discloses the RAN node 212 sending controller 210 updated hardware availability information in ¶ 0079, which is implicitly information about whether the RAN node can support the target algorithm).
For Claim 20, the claim is substantially similar to claim 11 and therefore is rejected for the same reasoning set forth above. Additionally, Bedekar teaches a communication device, comprising a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the information interaction method according to claim 11 (Bedekar, FIG. 7; ¶ 0168 “… The wireless station also includes a processor or control unit/entity (controller) 1004 to execute instructions or software and control transmission and receptions of signals, and a memory 1006 to store data and/or instructions …” ).
Citation of Pertinent Prior Art
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure is listed below, thank you:
i. Wang et al. (US 20210064996 A1) teaches neural network formation configuration feedback for wireless communications. A network entity (base station 120, core network server 302) determines a set (sets 2002, 2104) of deep neural network reference signals to send on a physical channel for measuring performance of deep neural networks, and/or a set of neural network formation configurations (sets 2004, 2106). The network entity communicates indications (indications 2030, 2102) of the set of neural network formation configurations and the set of deep neural network reference signals to a user equipment (UE 110). The network entity initiates transmission of each deep neural network reference signal of the set, and directs the user equipment to measure (outputs 2032, 2034, 2036, 2124, 2128, 2132) performance of a set of deep neural networks formed using the set of neural network formation configurations, and select one of the set of neural network formation configurations (Abstract).
ii. YANG et al. (US 20230245000 A1) teaches that model evaluation methods and apparatuses are described, which may be applied to systems such as 5G and vehicle-to-everything (V2X). In an example model evaluation method, an inference device determines an inference output of a machine learning model based on an inference input of the machine learning model, where the machine learning model is determined based on first configuration information. The inference device sends a first message to a measurement device, where the first message includes a measurement object, the measurement object is determined based on the inference output, and the first message is used to request a measurement result of the measurement object. The inference device receives the measurement result from the measurement device, where the measurement result corresponds to the inference output. The inference device sends an evaluation result to an update device, where the evaluation result is determined based on the inference output and the measurement result (Abstract).
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ZONGHUA DU whose telephone number is (408)918-7596. The examiner can normally be reached Monday - Friday 8 AM - 5 PM PST.
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, John Follansbee can be reached on (571) 272-3964. 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.
/Z.D./Examiner, Art Unit 2444
/SCOTT B CHRISTENSEN/Primary Examiner, Art Unit 2444
Prosecution Insights