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 .
DETAILED ACTION
This office action is a response to application no. 18/912,922 filed on 10/11/2024.
Claims 1 – 20 are pending and ready for examination.
Priority
This application claims benefit of U.S. Provisional Patent Application No. 63/590,048 filed on October 13, 2023.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 10/17/2024 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
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 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 of this title, 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.
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, 4, 6 – 8, 11, 15 – 17 and 19 are rejected under 35 U.S.C. 103 as being unpatentable over PARK et al. (PARK hereinafter referred to PARK) (US 2023/0319789 A1) in view of YANG et al. (YANG hereinafter referred to YANG) (US 2026/0304185 A1).
Regarding claim 1, PARK teaches a method (Title, METHOD AND APPARATUS FOR CHANNEL STATE INFORMATION FEEDBACK IN COMMUNICATION SYSTEM) for compressing and recovering channel information between edge equipment and a base station ([0002], a channel state information feedback technique for channel state information acquisition, in which a terminal reports measured channel state information to a base station by using an intelligent encoder, and the base station acquires the channel state information by using an intelligent decoder. Here, the encoder and the decoder work as compressing and recovering channel state information between a terminal/ edge equipment and a base station), the method comprising:
partitioning a wireless environment into one or more channel zones ([0027] and [0072], dividing the entire channel into a plurality of subchannels; [0079], the terminal divides downlink channel composed of N RBs into K sub-downlink channels (i.e., subchannels). Here, the plurality of subchannels are one or more channel zones);
decomposing a channel state information (CSI) feedback network into a plurality of subnetwork models ([0072], a method of constructing an intelligent encoder/ decoder for CSI feedback by dividing the entire channel into a plurality of subchannels and combining a subchannel-based neural network encoder/ decoder and a subchannel group-based neural network encoder/decoder; [0084], the base station and the terminal construct a neural network model suitable for each role; [0094], the base station or the network independently configures the first neural network and the second neural network; [0019], generating configuration information of a first neural network encoder, the configuration information of the first neural network encoder including neural network model information for the first neural network encoder; [0020], generating configuration information of a second neural network encoder, the configuration information of the second neural network encoder including neural network model information for the second neural network encoder. Here, the first neural network and the second neural network are considered as a plurality of subnetwork models, as the subchannel-based neural network encoder/ decoder are used for CSI feedback), the plurality of subnetwork models compressing and recovering channels from the one or more channel zones ([0080], the first neural network encoder serves to compress the channel state information of each subchannel defined in the spatial domain into the first latent space; [0081], the second neural network encoder of the terminal acquires channel state information for subchannels constituting each subchannel group, compresses the channel state information into a second latent space; [0083], the first neural network decoder of the base station restores the compressed channel state information for each subchannel; [0086], the second neural network decoder of the base station restores the compressed average channel state information and channel state change amount information, and outputs the average channel state information for each subchannel group. Here, the first and the second neural network encoders work as compressing channels and the first and the second neural network decoders work as recovering channels), each of the plurality of subnetwork models having a CSI encoder (Fig.4 and [0075], a first neural network encoder 411 and a second neural network encoder 412; [0077], the first neural network encoder 411 is composed of a plurality of branch neural network encoders each of which is configured to correspond to each subchannel), each of the plurality of subnetwork models includes one or more model parameters ([0010] and [0088], The neural network model information includes at least one of neural network type information, type information for each layer, information on a number of neurons per layer, or inter-layer connection information. Here, the neural network model information is one or more model parameters);
the plurality of subnetwork models for the one or more channel zones forming a composite CSI feedback model ([0082], the terminal transmits the compressed channel state information for each subchannel group to the base station. Here, the compressed channel state information is a composite CSI feedback); and
compressing and recovering the channel information between the edge equipment and the base station based on the composite CSI feedback model ([0083], the first neural network decoder of the base station restores the compressed channel state information for each subchannel, and generate and outputs the restored channel state information for each subchannel. Then, the base station infers the channel state information for each subchannel from the restored channel state information for each subchannel. Here, the compressing and recovering the channel information is based on the composite CSI feedback).
PARK does not specifically teach
training the plurality of subnetwork models.
However, YANG teaches a method (Title, METHOD PERFORMED BY NODE IN WIRELESS COMMUNICATION SYSTEM AND ELECTRONIC APPARATUS THEREOF) for channel information ([0030], a channel state information acquisition method), the method comprising:
decomposing a channel state information (CSI) feedback network into a plurality of subnetwork models ([0046], The neural network includes a plurality of cascaded sub-neural networks, wherein at least one of the plurality of sub-neural networks is used for acquiring channel state information);
training the plurality of subnetwork models ([0049] and [0160], each of lightweight sub-neural network models is trained independently); and
recovering the channel information between the edge equipment and the base station ([0049], by using the neural network including the plurality of cascaded sub-neural networks, a communication node is enabled to accurately acquire the channel state information through reference signals with lower density and/or lower number distribution).
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to have modified PARK as mentioned above and further incorporate the teaching of YANG. The motivation for doing so would have been to provide a method for using neural network including a plurality of cascaded sub-neural networks, a communication node is enabled to accurately acquire channel state information through reference signals with lower density and/or lower number distribution, thereby reducing a resource overhead of the reference signal, and improving a resource utilization rate and performance of a wireless communication system (YANG, Title and [0049], [0160]).
Regarding claim 4, combination of PARK and YANG teaches all the features with respect to claim 1 as outlined above.
PARK further teaches
wherein the training comprises:
jointly training the CSI encoder and a decoder ([0077], the plurality of branch neural network encoders and decoders having the same weight are configured and trained at once, so that optimization is achieved in terms of performance. Here, the encoder and the decoder are trained at once, i.e. simultaneously/ jointly).
Regarding claim 6, combination of PARK and YANG teaches all the features with respect to claim 1 as outlined above.
PARK further teaches wherein the training comprises:
training multiple decoders and the CSI encoders for the one or more channel zones ([0077], the plurality of branch neural network encoders and decoders having the same weight are trained at once; [0147], base station trains the constructed neural network encoder and neural network decoder using the configured training data set).
Regarding claim 7, combination of PARK and YANG teaches all the features with respect to claim 1 as outlined above.
PARK further teaches
wherein multiple of the one or more channel zones share a decoder or the CSI encoder ([0081], second neural network encoder of the terminal generates subchannel groups by sequentially selecting a predetermined number of subchannels from the plurality of subchannels. Here, the second neural network encoder (i.e. CSI encoder) is shared in subchannel groups, i.e. multiple of the one or more channel zones).
Regarding claim 8, combination of PARK and YANG teaches all the features with respect to claim 1 as outlined above.
PARK further teaches
wherein the one or more channel zones are non-overlapping ([0079] and [0151], the terminal divides the downlink channel composed of N RBs into K sub-downlink channels (i.e., subchannels), each sub-downlink channel is composed of M consecutive RBs. Accordingly, K is N/M and, K, N, and M are positive integers, and N>K. Here, the one or more subchannels/ channel zones composed of M consecutive RBs; therefore, it is obvious that the one or more channel zones are non-overlapping).
Regarding claim 11, combination of PARK and YANG teaches all the features with respect to claim 1 as outlined above.
PARK further teaches
wherein the one or more model parameters differ between each of the plurality of subnetwork models ([0019], neural network model information for the first neural network encoder constituting a first auto-encoder with the first neural network decoder and ….; [0020], neural network model information for the second neural network encoder constituting a second auto-encoder with the second neural network decoder and …. Here, the neural network model information/ model parameters for the first neural network and the second neural network are not same; therefore, the one or more model parameters differ between each of the plurality of subnetwork models).
Regarding claim 15, combination of PARK and YANG teaches all the features with respect to claim 1 as outlined above.
PARK further teaches
wherein the edge equipment comprises cellular devices (Fig.1 and [0060], Each of the plurality of terminals 130-1, 130-2, 130-3, 130-4, 130-5, and 130-6 is referred to as user equipment (UE), terminal equipment (TE )…..; [0061], Each of the plurality of communication nodes 110-1, 110-2, 110-3, 120-1, 120-2, 130-1, 130-2, 130-3, 130-4, 130-5, and 130-6 supports cellular communication. As mentioned in claim 1, terminal is an edge equipment; therefore, the edge equipment comprises cellular devices).
Regarding claim 16, combination of PARK and YANG teaches all the features with respect to claim 1 as outlined above.
PARK further teaches
wherein the one or more channel zones comprise one or more spatial zones ([0080], spatial domain of the subchannels; the first neural network encoder compresses the channel state information for each subchannel in the spatial domain. As mentioned in claim 1, the plurality of subchannels are one or more channel zones; therefore, the one or more channel zones comprise one or more spatial domain/ zones).
Regarding claim 17, PARK teaches (Title, METHOD AND APPARATUS FOR CHANNEL STATE INFORMATION FEEDBACK IN COMMUNICATION SYSTEM) a computing system ([0186], a computer system) for compressing and recovering channel information between edge equipment and a base station ([0002], a channel state information feedback technique for channel state information acquisition, in which a terminal reports measured channel state information to a base station by using an intelligent encoder, and the base station acquires the channel state information by using an intelligent decoder. Here, the encoder and the decoder work as compressing and recovering channel state information between a terminal/ edge equipment and a base station) comprising:
one or more processors ([0188], a microprocessor); and
a memory system comprising one or more non-transitory computer-readable media storing instructions ([0187], store and execute a program command, such as a ROM, RAM or flash memory) that, when executed by at least one of the one or more processors, cause the computing system to perform operations ([0186], computer readable recording medium stores and executes programs or codes which are distributed in computer systems connected through a network and read through computers in a distributed manner; [0188], Some or all of the steps of the method are executed by (or using) a hardware apparatus such as a microprocessor, a programmable computer or an electronic circuit), the operations comprising:
partitioning a wireless environment into one or more channel zones ([0027] and [0072], dividing the entire channel into a plurality of subchannels; [0079], the terminal divides downlink channel composed of N RBs into K sub-downlink channels (i.e., subchannels). Here the plurality of subchannels are one or more channel zones);
decomposing a channel state information (CSI) feedback network into a plurality of subnetwork models ([0072], a method of constructing an intelligent encoder/ decoder for CSI feedback by dividing the entire channel into a plurality of subchannels and combining a subchannel-based neural network encoder/ decoder and a subchannel group-based neural network encoder/decoder; [0084], the base station and the terminal construct a neural network model suitable for each role; [0094], the base station or the network independently configures the first neural network and the second neural network; [0019], generating configuration information of a first neural network encoder, the configuration information of the first neural network encoder including neural network model information for the first neural network encoder; [0020], generating configuration information of a second neural network encoder, the configuration information of the second neural network encoder including neural network model information for the second neural network encoder. Here, the first neural network and the second neural network are considered as a plurality of subnetwork models, as the subchannel-based neural network encoder/ decoder are used for CSI feedback), the plurality of subnetwork models compressing and recovering channels from the one or more channel zones ([0080], the first neural network encoder serves to compress the channel state information of each subchannel defined in the spatial domain into the first latent space; [0081], the second neural network encoder of the terminal acquires channel state information for subchannels constituting each subchannel group, compresses the channel state information into a second latent space; [0083], the first neural network decoder of the base station restores the compressed channel state information for each subchannel; [0086], the second neural network decoder of the base station restores the compressed average channel state information and channel state change amount information, and outputs the average channel state information for each subchannel group. Here, the first and the second neural network encoders work as compressing channels and the first and the second neural network decoders work as recovering channels), each of the plurality of subnetwork models having a CSI encoder (Fig.4 and [0075], a first neural network encoder 411 and a second neural network encoder 412; [0077], the first neural network encoder 411 is composed of a plurality of branch neural network encoders each of which is configured to correspond to each subchannel), each of the plurality of subnetwork models includes one or more model parameters ([0010] and [0088], The neural network model information includes at least one of neural network type information, type information for each layer, information on a number of neurons per layer, or inter-layer connection information. Here, the neural network model information is one or more model parameters);
the plurality of subnetwork models for the one or more channel zones forming a composite CSI feedback model ([0082], the terminal transmits the compressed channel state information for each subchannel group to the base station. Here, the compressed channel state information is a composite CSI feedback); and
compressing and recovering the channel information between the edge equipment and the base station based on the composite CSI feedback model ([0083], the first neural network decoder of the base station restores the compressed channel state information for each subchannel, and generate and outputs the restored channel state information for each subchannel. Then, the base station infers the channel state information for each subchannel from the restored channel state information for each subchannel. Here, the compressing and recovering the channel information is based on the composite CSI feedback).
PARK does not specifically teach
training the plurality of subnetwork models.
However, YANG teaches a method (Title, METHOD PERFORMED BY NODE IN WIRELESS COMMUNICATION SYSTEM AND ELECTRONIC APPARATUS THEREOF) for channel information ([0030], a channel state information acquisition method), the method comprising:
decomposing a channel state information (CSI) feedback network into a plurality of subnetwork models ([0046], The neural network includes a plurality of cascaded sub-neural networks, wherein at least one of the plurality of sub-neural networks is used for acquiring channel state information);
training the plurality of subnetwork models ([0049] and [0160], each of lightweight sub-neural network models is trained independently); and
recovering the channel information between the edge equipment and the base station ([0049], by using the neural network including the plurality of cascaded sub-neural networks, a communication node is enabled to accurately acquire the channel state information through reference signals with lower density and/or lower number distribution).
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to have modified PARK as mentioned above and further incorporate the teaching of YANG. The motivation for doing so would have been to provide a method for using neural network including a plurality of cascaded sub-neural networks, a communication node is enabled to accurately acquire channel state information through reference signals with lower density and/or lower number distribution, thereby reducing a resource overhead of the reference signal, and improving a resource utilization rate and performance of a wireless communication system (YANG, Title and [0049], [0160]).
Regarding claim 19, PARK teaches (Title, METHOD AND APPARATUS FOR CHANNEL STATE INFORMATION FEEDBACK IN COMMUNICATION SYSTEM) non-transitory computer-readable medium storing instructions ([0186], present disclosure is implemented as a computer readable program or code in a computer readable recording medium. The computer readable recording medium includes all kinds of recording apparatus for storing data which are read by a computer system) for compressing and recovering channel information between edge equipment and a base station ([0002], a channel state information feedback technique for channel state information acquisition, in which a terminal reports measured channel state information to a base station by using an intelligent encoder, and the base station acquires the channel state information by using an intelligent decoder. Here, the encoder and the decoder work as compressing and recovering channel state information between a terminal/ edge equipment and a base station) that, when executed by one or more processors of a computing system, cause the computing system to perform operations ([0188], Some or all of the steps of the method are executed by (or using) a hardware apparatus such as a microprocessor, a programmable computer or an electronic circuit), the operations comprising:
partitioning a wireless environment into one or more channel zones ([0027] and [0072], dividing the entire channel into a plurality of subchannels; [0079], the terminal divides downlink channel composed of N RBs into K sub-downlink channels (i.e., subchannels). Here the plurality of subchannels are one or more channel zones);
decomposing a channel state information (CSI) feedback network into a plurality of subnetwork models ([0072], a method of constructing an intelligent encoder/ decoder for CSI feedback by dividing the entire channel into a plurality of subchannels and combining a subchannel-based neural network encoder/ decoder and a subchannel group-based neural network encoder/decoder; [0084], the base station and the terminal construct a neural network model suitable for each role; [0094], the base station or the network independently configures the first neural network and the second neural network; [0019], generating configuration information of a first neural network encoder, the configuration information of the first neural network encoder including neural network model information for the first neural network encoder; [0020], generating configuration information of a second neural network encoder, the configuration information of the second neural network encoder including neural network model information for the second neural network encoder. Here, the first neural network and the second neural network are considered as a plurality of subnetwork models, as the subchannel-based neural network encoder/ decoder are used for CSI feedback), the plurality of subnetwork models compressing and recovering channels from the one or more channel zones ([0080], the first neural network encoder serves to compress the channel state information of each subchannel defined in the spatial domain into the first latent space; [0081], the second neural network encoder of the terminal acquires channel state information for subchannels constituting each subchannel group, compresses the channel state information into a second latent space; [0083], the first neural network decoder of the base station restores the compressed channel state information for each subchannel; [0086], the second neural network decoder of the base station restores the compressed average channel state information and channel state change amount information, and outputs the average channel state information for each subchannel group. Here, the first and the second neural network encoders work as compressing channels and the first and the second neural network decoders work as recovering channels), each of the plurality of subnetwork models having a CSI encoder (Fig.4 and [0075], a first neural network encoder 411 and a second neural network encoder 412; [0077], the first neural network encoder 411 is composed of a plurality of branch neural network encoders each of which is configured to correspond to each subchannel), each of the plurality of subnetwork models includes one or more model parameters ([0010] and [0088], The neural network model information includes at least one of neural network type information, type information for each layer, information on a number of neurons per layer, or inter-layer connection information. Here, the neural network model information is one or more model parameters);
the plurality of subnetwork models for the one or more channel zones forming a composite CSI feedback model ([0082], the terminal transmits the compressed channel state information for each subchannel group to the base station. Here, the compressed channel state information is a composite CSI feedback); and
compressing and recovering the channel information between the edge equipment and the base station based on the composite CSI feedback model ([0083], the first neural network decoder of the base station restores the compressed channel state information for each subchannel, and generate and outputs the restored channel state information for each subchannel. Then, the base station infers the channel state information for each subchannel from the restored channel state information for each subchannel. Here, the compressing and recovering the channel information is based on the composite CSI feedback).
PARK does not specifically teach
training the plurality of subnetwork models.
However, YANG teaches a method (Title, METHOD PERFORMED BY NODE IN WIRELESS COMMUNICATION SYSTEM AND ELECTRONIC APPARATUS THEREOF) for channel information ([0030], a channel state information acquisition method), the method comprising:
decomposing a channel state information (CSI) feedback network into a plurality of subnetwork models ([0046], The neural network includes a plurality of cascaded sub-neural networks, wherein at least one of the plurality of sub-neural networks is used for acquiring channel state information);
training the plurality of subnetwork models ([0049] and [0160], each of lightweight sub-neural network models is trained independently); and
recovering the channel information between the edge equipment and the base station ([0049], by using the neural network including the plurality of cascaded sub-neural networks, a communication node is enabled to accurately acquire the channel state information through reference signals with lower density and/or lower number distribution).
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to have modified PARK as mentioned above and further incorporate the teaching of YANG. The motivation for doing so would have been to provide a method for using neural network including a plurality of cascaded sub-neural networks, a communication node is enabled to accurately acquire channel state information through reference signals with lower density and/or lower number distribution, thereby reducing a resource overhead of the reference signal, and improving a resource utilization rate and performance of a wireless communication system (YANG, Title and [0049], [0160]).
Claims 2, 9, 13 – 14, 18 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over PARK in view of YANG and further in view of GAO (US 2026/0222033 A1).
Regarding claims 2, 18 and 20, combination of PARK and YANG teaches all the features with respect to claims 1, 17 and 19, respectively as outlined above.
PARK further teaches
clustering the downlink channel into the one or more channel zones ([0079] and [0151], terminal divides the downlink channel composed of N RBs into K sub-downlink channels (i.e., subchannels), each sub-downlink channel composed of M consecutive RBs).
PARK does not specifically teach
determining a position of the edge equipment;
collecting a downlink channel dataset; and
clustering the downlink channel dataset based on the position.
However, GAO teaches (Title, METHOD FOR COLLECTING DATA AND COMMUNICATION APPARATUS)
determining a position of the edge equipment ([0185] – [0186], network-side device receives indication information sent by the terminal, where the indication information is used to indicate geographical location of the terminal);
collecting a downlink channel dataset ([0101] and [0103], first data information includes at least one of the following: [0102] forward propagation (FP) data; [0103] estimated downlink channel information. Here, the first data information is the estimated downlink channel information; i.e. a downlink channel dataset. [0063], The CSI generation part model and the CSI recovery part model need to be trained using collected datasets; [0142], a terminal group-specific or area-specific reference signal is designed, enabling terminals within a specific terminal group or area to complete the collection of the first data information based on the reference signal and then send the collected first data information to the network-side device. Here, collection of the first data information is the collected downlink channel dataset); and
clustering the downlink channel dataset into the one or more channel zones ([0505], a group-specific or area-specific reference signal, enabling terminals within the group or area to complete the collection of downlink channel information based on the reference signal. The terminals then send the collected downlink channel information dataset to the network-side device) based on the position ([0512], The grouping or area division method is based on the geographical location of the terminal).
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to have modified combination of PARK and YANG as mentioned in claims 1, 17 and 19 and further incorporate the teaching of GAO. The motivation for doing so would have been to provide a method for collecting data to perform a model processing of a CSI generation model and/or a CSI recovery model or calculate CSI, wherein the base station groups UEs with the same area ID in a specific area and broadcast a group-specific reference signal to the UEs in the group to train a CSI generation part model and a CSI recovery part model specific to the area, thereby improving the effectiveness of CSI compression (GAO, Abstract and [0533]).
Regarding claim 9, combination of PARK and YANG teaches all the features with respect to claim 1 as outlined above.
PARK does not specifically teach
accessing positions of the edge equipment from a network or cloud; and
clustering the edge equipment into the one or more channel zones based on the positions.
However, GAO teaches (Title, METHOD FOR COLLECTING DATA AND COMMUNICATION APPARATUS)
accessing positions of the edge equipment from a network or cloud ([0185] – [0186], network-side device receives indication information sent by the terminal, where the indication information is used to indicate geographical location of the terminal); and
clustering the edge equipment into the one or more channel zones based on the positions ([0505], a group-specific or area-specific reference signal, enabling terminals within the group or area to complete the collection of downlink channel information based on the reference signal; [0512], The grouping or area division method is based on the geographical location of the terminal).
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to have modified combination of PARK and YANG as mentioned in claim 1 and further incorporate the teaching of GAO. The motivation for doing so would have been to provide a method for collecting data to perform a model processing of a CSI generation model and/or a CSI recovery model or calculate CSI, wherein the base station groups UEs with the same area ID in a specific area and broadcast a group-specific reference signal to the UEs in the group to train a CSI generation part model and a CSI recovery part model specific to the area, thereby improving the effectiveness of CSI compression (GAO, Abstract and [0533]).
Regarding claim 13, combination of PARK and YANG teaches all the features with respect to claim 1 as outlined above.
PARK does not specifically teach
wherein partitioning the wireless environment comprises:
clustering data about characteristics of edge equipment; and
partitioning the edge equipment into the one or more channel zones based at least on the characteristics.
However, GAO teaches (Title, METHOD FOR COLLECTING DATA AND COMMUNICATION APPARATUS)
clustering data ([0505], a group-specific or area-specific reference signal, enabling terminals within the group or area to complete the collection of downlink channel information based on the reference signal. The terminals send the collected downlink channel information dataset to the network-side device) about characteristics of edge equipment ([0395]-[0399], the terminal sends indication information to the network-side device, where the indication information is used to indicate at least one of the following: the geographical location of the terminal; the area identifier of the terminal; the time-domain channel characteristics reported by the terminal; or the time-domain correlation of the terminal. Here, the indication information provides characteristics of edge equipment); and
partitioning the edge equipment into the one or more channel zones based at least on the characteristics ([0512], The grouping or area division method is based on the geographical location of the terminal).
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to have modified combination of PARK and YANG as mentioned in claim 1 and further incorporate the teaching of GAO. The motivation for doing so would have been to provide a method for collecting data to perform a model processing of a CSI generation model and/or a CSI recovery model or calculate CSI, wherein the base station groups UEs with the same area ID in a specific area and broadcast a group-specific reference signal to the UEs in the group to train a CSI generation part model and a CSI recovery part model specific to the area, thereby improving the effectiveness of CSI compression (GAO, Abstract and [0533]).
Regarding claim 14, combination of PARK, YANG and GAO teaches all the features with respect to claim 13 as outlined above.
PARK does not specifically teach
wherein the characteristics comprise:
one or more of a position of the edge equipment, signal quality in the one or more channel zones, channel statistics in the one or more channel zones, or the wireless environment in the one or more channel zones.
However, GAO teaches
wherein the characteristics comprise:
one or more of a position of the edge equipment ([0396], geographical location of the terminal), signal quality in the one or more channel zones, channel statistics in the one or more channel zones ([0398], the time-domain channel characteristics), or the wireless environment in the one or more channel zones (Due to alternative language, “one or more” in the claim, examiner did not address all limitations).
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to have modified combination of PARK, YANG and GAO as mentioned in claim 13 and further incorporate the teaching of GAO. The motivation for doing so would have been to provide a method for collecting data to perform a model processing of a CSI generation model and/or a CSI recovery model or calculate CSI, wherein the base station groups UEs with the same area ID in a specific area and broadcast a group-specific reference signal to the UEs in the group to train a CSI generation part model and a CSI recovery part model specific to the area, thereby improving the effectiveness of CSI compression (GAO, Abstract and [0533]).
Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over PARK in view of YANG and further in view of Kwon et al. (Kwon hereinafter referred to Kwon) (US 2023/0145844 A1).
Regarding claim 5, combination of PARK and YANG teaches all the features with respect to claim 1 as outlined above.
PARK does not specifically teach
wherein the training comprises:
training multiple of the CSI encoders per decoder.
However, Kwon teaches (Title, METHOD AND APPARATUS FOR TRANSMITTING CHANNEL INFORMATION BASED ON MACHINE LEARNING)
wherein the training comprises:
training multiple of the CSI encoders per decoder (Fig.8 and [0150], training is performed in a structure of a set including one decoder 831 for the plurality of encoders 811 and 821. When a structure of a set in which the plurality of encoders 811 and 821 are mapped to one decoder 831).
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to have modified combination of PARK and YANG as mentioned in claim 1 and further incorporate the teaching of Kwon. The motivation for doing so would have been to provide a preparation process for utilizing machine learning-based channel information compression and decompression techniques in a wireless communication system, for efficiently transferring channel information using a determined machine learning model, in which one decoder for a plurality of encoders provides an advantage in the number of decoders to be operated by a base station (Kwon, [0005] and [0150]).
Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over PARK in view of YANG and further in view of LI et al. (LI hereinafter referred to LI) (CN 106709875 A, Eng. translation is attached).
Regarding claim 10, combination of PARK and YANG teaches all the features with respect to claim 1 as outlined above.
PARK does not specifically teach
wherein the training comprises: training the plurality of subnetwork models based on an end-to-end learning approach and a mean square error loss function.
However, LI teaches (Title, Compressed Low-resolution Image Restoration Method Based On Combined Depth Network)
wherein the training comprises: training the plurality of subnetwork models based on an end-to-end learning approach (Pg.11: 1st para, network training and model obtaining method, respectively establishing a training mechanism for de-compression distortion sub-network and super-resolution complex atom network, and obtaining the mapping relation through end-to-end learning. Here, network training comprises an end-to-end learning approach) and a mean square error loss function (Pg.7: 2nd para, Training Network: First, respectively establishing loss monitoring for de-compression distortion sub-network and super-resolution complex atom network, and using the common gradient descending method to solve the problem of minimizing loss function. wherein the de-compression distortion sub-network uses the MSE loss function to determine the error. Here, network training comprises mean square error (MSE) loss function).
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to have modified combination of PARK and YANG as mentioned in claim 1 and further incorporate the teaching of LI. The motivation for doing so would have been to provide a compressed low-resolution image recovery method based on combined depth network formed by two sub-networks of cascade structure, in which, predicting the recovered image by network model, improves network prediction performance (LI, Pg.10: last para and Pg.11: 1st para).
Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over PARK in view of YANG and further in view of LIU et al. (LIU hereinafter referred to LIU) (CN 116073867 A, Eng. translation is attached).
Regarding claim 12, combination of PARK and YANG teaches all the features with respect to claim 1 as outlined above.
PARK does not specifically teach
wherein the one or more channel zones comprises:
one or more clusters of scatterers.
However, LIU teaches (Title, Holographic Mixing Pre-coding Method Based On Reconfigurable Intelligent Super-surface)
wherein the one or more channel zones comprises:
one or more clusters of scatterers (Pg.8: 2nd and 3rd para, The channel model : millimeter wave MIMO channel of the temporary communication is viewed as Saleh-Valenzuela (S-V) cluster sparse channel with large scale fading, namely a LoS (Line-of-Sight, sight distance diameter) and Nc (sub-carrier number) is composed of scatterer cluster, Each cluster comprises Np Np NLoS (Non-Line-of-Sight, non-visual distance)).
Therefore, it would have been obvious to one of the ordinary skill in the art before the effective filing date of the claimed invention to have modified combination of PARK and YANG as mentioned in claim 1 and further incorporate the teaching of LIU. The motivation for doing so would have been to provide a method for improving the error bit rate and spectrum efficiency performance of reconfigurable intelligent super-surface auxiliary MIMO-OFDM communication system, and reducing the calculation complexity (LIU, Abstract).
Allowable Subject Matter
Claim 3 is objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Conclusion
The prior arts made of record and not relied upon are considered pertinent to applicant's disclosure.
XIAO (Pub. No. US 2025/0141518 A1) – “MODEL TRAINING METHODS AND APPARATUSES, SAMPLE DATA GENERATION METHOD AND APPARATUS, AND ELECTRONIC DEVICE” discloses model training methods and apparatuses. A model training method includes: a first device generating a plurality of pieces of sample data on the basis of a first codebook of a precoding matrix; the first device training an initial channel state information (CSI) feedback model on the basis of the plurality of pieces of sample data to obtain a CSI feedback meta-model. The CSI feedback meta-model is used for training a target CSI feedback model, and the target CSI feedback model is used for encoding CSI obtained by a signal receiving end and restore the encoded CSI at a signal transmitting end. Further provided in the embodiments of the application are a sample data generation method and apparatus.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ROWNAK ISLAM whose telephone number is (571)272-8009. The examiner can normally be reached on Monday - Friday 8:30 am - 6 pm (EST).
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Michael Thier can be reached on 571-272-2832. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information Regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/ROWNAK ISLAM/
Primary Examiner, Art Unit 2474
Prosecution Insights