DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Receipt is acknowledged of papers submitted claiming the benefit of Application No. KR10-2023-0097058, filed on 07/25/2023, and Application No. KR10-2023-0185088, filed on 12/18/2023, which papers have been placed of record in the file required by 37 CFR 1.55.
Information Disclosure Statement
The information disclosure statement(s) (IDS) submitted on 07/23/2024 and 01/02/2025 have been considered by the examiner.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-5, 11-12, 14-16, 18, and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Jiang et al. (US 2024/0080692 A1, hereinafter Jiang) in view of DEENOO et al. (US 2023/0389057 A1, hereinafter Deenoo).
Regarding claim 1, Jiang teaches a method of a terminal ([0132] referring to FIG. 4, "the process flow 400 illustrates communications between a UE 115-c and a network entity 105-d," UE 115-c read as terminal), the method comprising: receiving a measurement configuration from a serving cell ([Figure 4, 405] and [0134] describes "at 405, the network entity 105-d may transmit a reference signal to the UE 115-c" and "the reference signal may be an SSB, a TRS, a CSI-RS, or some other type of reference signal," i.e. UE 115-c receives a measurement configuration from serving cell 105-d. Furthermore, [0134] describes there may be a plurality of reference signals received by the UE 115-c from one or more network cells); measuring at least one received signal based on the measurement configuration ([Figure 4, 410] and [0135] "at 410, the UE 115-c may measure a multi-dimensional channel response based on the reference signal," i.e. UE 115-c measuring at least one received measurement configuration signal); confirming whether an event corresponding to the measurement configuration has occurred; based on confirming that the event has occurred ([0129] describes the UE 115-b may report the signal strength measurements and the associated beam and cell IDs to the serving cell based on a threshold value, threshold value read as event, i.e. an event being confirmed based on a threshold value being met), generating- -a measurement result set comprising a plurality of measurement results; and in response to an occurrence of the event, transmitting, to the serving cell, a measurement report generated- ([Figure 4, 415] and [0136] "at 415, the UE 115-c may transmit a report that includes a channel measurement vector indicating the multiple measured channel metrics for one or more dimensions of the multi-dimensional channel response," i.e. UE 115-c generates a measurement vector based on the plurality of received reference signals, and in response to a threshold value (event) being met, transmitting to the network entity 105-d a measurement report based on the measurement vector).
Jiang is not relied on for the claim language -a latent vector by encoding- and -based on the latent vector. However, Deenoo teaches [abstract] a method which may be used to aid in handover decisions, see [0044], wherein a WTRU receives a plurality of data from a BS and selects an AI filter to encode the data. Deenoo also teaches -a latent vector by encoding- and -based on the latent vector ([0083] - [0084] describes the ability of a WTRU device to encode/decode a latent vector comprising a plurality of different types of data, see [0028], i.e. the received data may specifically be encoded into a latent vector).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the vector to be specifically a latent vector, which may be used as an input for neural networks as taught by Deenoo, in order to aid in [0019] improving access to one or more communication networks by [0082] enabling the AI component to learn complex behaviors which would otherwise be more difficult using legacy operations/methods such as input methods and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
Regarding claim 2, Jiang is not relied on for the claim language the generating of the latent vector comprises: inputting the measurement result set to a first neural network model corresponding to an encoder trained based on an autoencoder; and obtaining an output of the first neural network model as the latent vector. However, Deenoo teaches the generating of the latent vector comprises: inputting the measurement result set to a first neural network model corresponding to an encoder trained based on an autoencoder ([0108] describes "referring to FIG. 2, the representative procedure 200 may include, for example, a WTRU 120 receiving or obtaining information 220 to perform default functionality using the AI model 210 (e.g., an AI filter (e.g., AI filter 320-1, 320-2 and/or 320-3) may implement weights and biases associated with nodes of a neural network to implement the AI model 210," i.e. each AI filter may include neural networks, and measurements may be used as inputs for said neural networks, wherein the neural networks may correspond to an encoder which may be trained based on an autoencoder, see [0083]); and obtaining an output of the first neural network model as the latent vector ([0117] describes "the AI filter 320-2 may output a first output 330-1 and/or a second output 330-2, as: (1) one or more AI determined packets; and/or (2) one or more parameters/information (e.g., any of: one or more transmission profile parameters/information, one or more next hop parameters/information, one of more sidelink resources/parameters/information, and/or one or more link adaptation parameters/information, among others) which are associated with the inputted packets," i.e. an output from the neural network, which may be "transformed to a lower dimensional latent vector using a DNN based encoder," see [0083]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the vector to be specifically a latent vector, which may be used as an input for neural networks as taught by Deenoo, in order to aid in [0019] improving access to one or more communication networks by [0082] enabling the AI component to learn complex behaviors which would otherwise be more difficult using legacy operations/methods such as input methods and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
Regarding claim 3, Jiang teaches the measurement report comprises the- -vector ([Figure 4, 415] reporting/transmitting to network entity 105-d the measurement vector).
Jiang is not relied on for the claim language based on a second neural network model corresponding to a decoder trained based on the autoencoder being stored in the serving cell, and -latent vector-. However, Deenoo teaches based on a second neural network model corresponding to a decoder trained based on the autoencoder being stored in the serving cell ([0117] referring to FIG. 3, "the first WTRU 102A may include one or a plurality of AI filters 320-1, 320-2 and 320-3," i.e. each AI filter contains a neural network model corresponding to a decoder based on an autoencoder as described in claim 2, i.e. a second neural network model), and -latent vector- ([0083] - [0084] describes the ability of a WTRU device to encode/decode a latent vector comprising a plurality of different types of data, see [0028], i.e. the measurement report vector may be specifically a latent vector).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the vector to be specifically a latent vector, which may be used as an input for neural networks based on an autoencoder as taught by Deenoo, in order to aid in [0019] improving access to one or more communication networks by [0082] enabling the AI component to learn complex behaviors which would otherwise be more difficult using legacy operations/methods such as input methods and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
Regarding claim 4, Jiang teaches transmitting- -to the serving cell ([Figure 4, 415] reporting/transmitting to network entity 105-d the measurement vector).
Jiang is not relied on for the claim language generating the second neural network model; and -the second neural network model-. However, Deenoo teaches as such ([0117] Referring to FIG. 3 WTRU 102A may include one or a plurality of AI filters 320-1, 320-2 and 320-3, wherein each AI filter contains a neural network model, i.e. the capability to generate the second neural network model).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the vector to be specifically a latent vector, which may be used as an input for neural networks based on an autoencoder as taught by Deenoo, in order to aid in [0019] improving access to one or more communication networks by [0082] enabling the AI component to learn complex behaviors which would otherwise be more difficult using legacy operations/methods such as input methods and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
Regarding claim 5, Jiang is not relied on for the claim language transmitting, to the serving cell, capability information comprising first information indicating support for a measurement reporting function based on the autoencoder. However, Deenoo teaches as such ([0111] describes "the WTRU 102A may indicate the capabilities of the WTRU 102A in terms of any of: (1) storage availability, (2) support of various AI filter architectures, (3) parameterizations, and/or (4) processing latency, among others. The WTRU capability may be indicated as a part of a RRC connection request and/or any other RRC message," i.e. any transmission to the serving cell may include capability information, such as parameters or type of AI filter (support for a measurement reporting function), wherein the AI filter may be "an autoencoder (AE) neural network architecture" see [0186]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the terminal device to transmit capability information as taught by Deenoo, in order to aid in [0078] allowing the network controlled AI components in the processing chain to specifically adapt to different contexts (e.g., channel condition, quality of service, WTRU power saving, cell load, interference, and/or WTRU/NW capability, among others, i.e. terminal capabilities) and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
Regarding claim 11, Jiang teaches a method of a serving cell ([Figure 4, 105-d] network entity 105-d may be the BS of the serving cell displayed in FIG. 2), the operating method comprising: transmitting a measurement configuration to a terminal ([0134] describes "at 405, the network entity 105-d may transmit a reference signal to the UE 115-c," i.e. a measurement configuration); receiving, from the terminal, a measurement report corresponding to the measurement configuration- ([0136] describes "at 415, the UE 115-c may transmit a report that includes a channel measurement vector indicating the multiple measured channel metrics for one or more dimensions of the multi-dimensional channel response," i.e. network entity 105-d receives from the terminal 115-c a measurement report that corresponds to the reference signal in operation 405); and controlling a handover operation based on the measurement result set ([0138] referring to FIG. 4 operation 420, "the network entity 105-d may determine whether to initiate a handover procedure for the UE 115-c").
Jiang is not relied on for the claim language -and based on an autoencoder; obtaining a measurement result set comprising a plurality of measurement results by decoding a latent vector included in the measurement report. However, Deenoo teaches -and based on an autoencoder (FIG. 3 depicts AI-filters 320-1, 320-2, 320-3 (each containing a neural network) that create the latent vectors based on an autoencoder, see [0083] and [0186]); obtaining a measurement result set comprising a plurality of measurement results by decoding a latent vector included in the measurement report ([0141] describes "referring to FIG. 4, the representative procedure 400 may include that the WTRU 102 may be configured to determine an AI filter 320-1, 320-2 or 320-3 to apply for building a transport block 430 when the UL grant is received via a base station 114 and/or gNB 180," i.e. the transport block 430, transport block 430 read as measurement result, may include a latent vector (see [0083] - [0084], data results may be encoded/decoded into specifically latent vectors) which may be decoded by the serving cell/BS).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the vector to be specifically a latent vector, which may be used as an input for neural networks based on an autoencoder as taught by Deenoo, in order to aid in [0019] improving access to one or more communication networks by [0082] enabling the AI component to learn complex behaviors which would otherwise be more difficult using legacy operations/methods such as input methods and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
Regarding claim 12, Jiang is not relied on for the claim language the latent vector is generated based on a first neural network model corresponding to an encoder trained based on the autoencoder, and the obtaining of the measurement result set comprises: inputting the latent vector to a second neural network model corresponding to a decoder trained based on the autoencoder; and obtaining an output of the second neural network model as the measurement result set. However, Deenoo teaches the latent vector is generated based on a first neural network model corresponding to an encoder trained based on the autoencoder, and the obtaining of the measurement result set comprises: inputting the latent vector to a second neural network model corresponding to a decoder trained based on the autoencoder (referring to FIG. 4, a plurality of inputs 420-1, 420-2, 420-3 (which may be a latent vector based on the autoencoder as described above and in [0083] and [0186]) are first produced by a first AI-filter (which may include a first neural network) and later used as inputs for an additional AI filter (which may contain a second neural network corresponding to a decoder)); and obtaining an output of the second neural network model as the measurement result set (FIG. 4 depicts the output of the situation described above as "a transport block 430 when the UL grant is received via a base station 114 and/or gNB 180," i.e. obtaining the output from the second neural network model (within AI filter 310-3) as the measurement result set to be sent to the BS 114 of the serving cell).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the vector to be specifically a latent vector, which may be used as an input for neural networks based on an autoencoder as taught by Deenoo, in order to aid in [0019] improving access to one or more communication networks by [0082] enabling the AI component to learn complex behaviors which would otherwise be more difficult using legacy operations/methods such as input methods and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
Regarding claim 14, the claimed limitations of claim are rejected as the same reasons as set forth in claim 5.
Regarding claim 15, Jiang is not relied on for the claim language the capability information further comprises second information indicating the support for a measurement reporting determination function, and the controlling of the handover operation is performed further based on the second information. However, Deenoo teaches as such ([0111] describes "the WTRU 102A may indicate the capabilities of the WTRU 102A in terms of any of: (1) storage availability, (2) support of various AI filter architectures, (3) parameterizations, and/or (4) processing latency, among others. The WTRU capability may be indicated as a part of a RRC connection request and/or any other RRC message," i.e. any transmission to the serving cell may include capability information, such as parameters or type of AI filter (support for a measurement determination function), wherein said capability information may aid in radio resource management decisions or handover decisions, see [0063]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the terminal device to transmit capability information as taught by Deenoo, in order to aid in [0078] allowing the network controlled AI components in the processing chain to specifically adapt to different contexts (e.g., channel condition, quality of service, WTRU power saving, cell load, interference, and/or WTRU/NW capability, among others, i.e. terminal capabilities) and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
Regarding claim 16, Jiang teaches the measurement configuration comprises a configuration related to an ‘A3’ event, and the controlling of the handover operation comprises: determining handover to a target cell based on the second information; and determining the target cell based on the measurement result set ([0129] describes "the UE 115-b may report the signal strength measurements and the associated beam and cell IDs for the two dimensions associated with the strongest channel metric measurements," i.e. the measurement configuration in Figure 4, operation 405 is related to signal quality, the UE will compile a measurement vector to be sent to the serving cell containing the measurements of the candidate cells with the strongest signal qualities in operation 410/415, wherein said signal qualities may be better than the current signal quality of the serving cell, which may result in a handover command being executed (after the reception of operation 420) to one of the candidate cells based on the measurement result set, i.e. an 'A3' event). Furthermore, [0050] specifically discloses “the UE may measure channel metrics, such as signal strength measurements, rank information, spectral efficiency information, or any combination thereof, associated with one or more of the dimensions. The UE may transmit a report to the network entity that indicates a channel measurement vector including the measured channel metrics associated with the multiple spatial dimensions” and “the network entity, the UE, or both may thereby make mobility handover decisions (e.g., beam or cell selections) based on the reported signal strength metrics” i.e. the ability to measure signal strength measurements and cell rank measurements to determine a target cell for cell selection and therefore perform a handover procedure.
Regarding claim 18, Jiang teaches a terminal comprising: a memory configured to store-; and at least one processor- ([0006] "the apparatus for wireless communications at a UE is described. The apparatus may include a processor, memory coupled with the processor, and instructions stored in the memory," also shown in FIG. 6); and determine- -whether to transmit a measurement report to a serving cell ([Figure 4, 410] transmission 415 may not occur until a threshold is met, i.e. determining to transmit the measurement report).
Jiang is not relied on for the claim language -a neural network model trained based on an autoencoder; -configured to: generate, by using the neural network model, a latent vector from a measurement result set comprising a plurality of measurement results for handover, and --based on the latent vector-. However, Deenoo teaches -a neural network model trained based on an autoencoder ([0186] AI filters depicted in FIG. 3 specifically include "an autoencoder (AE) neural network architecture," see [0186], i.e. a neural network model trained based on an autoencoder); -configured to: generate, by using the neural network model (FIG. 3 depicts the use of a plurality of AI filters 320-1, 320-2, 320-1 (each containing a neural network, see [0108]), capable of generating a plurality of outputs, which may be a latent vector as described below), a latent vector from a measurement result set comprising a plurality of measurement results for handover ([0083] - [0084] describes the ability of a WTRU device to encode/decode a latent vector comprising a plurality of different types of data, see [0028], i.e. the measurement report vector may be specifically a latent vector, wherein the management data may be specifically used for aiding in handover decisions, see [0047]), and --based on the latent vector- (described above and in [0083] –[0084]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the vector to be specifically a latent vector, which may be used as an input for neural networks based on an autoencoder as taught by Deenoo, in order to aid in [0019] improving access to one or more communication networks by [0082] enabling the AI component to learn complex behaviors which would otherwise be more difficult using legacy operations/methods such as input methods and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
Regarding claim 20, Jiang teaches the measurement report- ([0136] referring to FIG. 4, "at 415, the UE 115-c may transmit a report that includes a channel measurement vector").
Jiang is not relied on for the claim language -comprises the latent vector. However, Deenoo teaches as such ([0083] - [0084] describes the ability of a WTRU device to encode/decode a latent vector comprising a plurality of different types of data, see [0028], i.e. the AI-filter outputs may be a measurement report vector, which may be specifically a latent vector).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the vector to be specifically a latent vector, which may be used as an input for neural networks as taught by Deenoo, in order to aid in [0019] improving access to one or more communication networks by [0082] enabling the AI component to learn complex behaviors which would otherwise be more difficult using legacy operations/methods such as input methods and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
Claims 6-10, 17, and 19 are rejected under 35 U.S.C. 103 as being unpatentable over Jiang et al. (US 2024/0080692 A1, hereinafter Jiang) and DEENOO et al. (US 2023/0389057 A1, hereinafter Deenoo) as applied in claims above, and further in view of Park et al. (US 2020/0252847 A1, hereinafter Park).
Regarding claim 6, Jiang teaches the transmitting of the measurement report to the serving cell ([Figure 4, 415] transmitting measurement report) and determining- -whether to transmit the measurement report to the serving cell ([Figure 4, 410] determining to transmit the measurement report to the serving cell).
Jiang is not relied on for the claim language -based on the latent vector. However, Deenoo teaches as such ([0083] - [0084] describes the ability of a WTRU device to encode/decode a latent vector comprising a plurality of different types of data, see [0028], i.e. the received data may specifically be encoded into a latent vector).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the vector to be specifically a latent vector, which may be used as an input for neural networks as taught by Deenoo, in order to aid in [0019] improving access to one or more communication networks by [0082] enabling the AI component to learn complex behaviors which would otherwise be more difficult using legacy operations/methods such as input methods and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
The combination of Jiang and Deenoo is not relied on for the claim language comprises: generating an event occurrence reliability value-; comparing the event occurrence reliability value with a reference value; and -based on a result of the comparing-. However, Park teaches [abstract] a method for transmitting configuration parameters between access nodes in a wireless network to aid in handover decisions. Park also teaches comprises: generating an event occurrence reliability value based on the latent vector; comparing the event occurrence reliability value with a reference value; and -based on a result of the comparing- ([0423] describes "the second access node may transmit information of latency and/or information of reliability of the first cell and/or the backhaul link when a configured condition/event occurs based on report configuration parameters (e.g., latency threshold value: 5 ms, reliability threshold value: packet loss rate 0.002) requested by the first access node via a status report request message," i.e. calculating the reliability of a threshold value being met, a plurality of other reliability values are also explained in [0418]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective
filing date of the claimed invention to modify the combination of Jiang and Deenoo to include the ability to compare the reliability value with a reference value, as taught by Park, in order to improve [0249] the ability to fulfill different services having different service requirements (e.g., data rate, latency, reliability), and therefore [0266] increase scheduling flexibility to allow the UE to communicate with different services/cells.
Regarding claim 7, Jiang teaches the plurality of measurement results comprises received signal received powers (RSRP) ([0103] describes "the UE 115-a may measure a signal strength (e.g., RSRP or reference signal received quality (RSRQ)) of the reference signals 225"), -and a candidate target cell for handover ([0103] describes "the UE 115-a and the network entity 105-a may make mobility decisions, such as beam selection, cell selection, handovers, or other mobility decisions, based on the reference signals 225" and [0010] describes "measurement vector indicates the set of multiple measured channel metrics associated with the two or more cell IDs," i.e. a cell ID of the best signal may be a candidate target cell for a handover procedure).
The combination of Jiang and Deenoo is not relied on for the claim language time advances (TA), and precoding matrix indications (PMI) corresponding to the serving cell. However, Park teaches time advances (TA) ([0207] describes "MAC CEs indicating one or more timing advance values for one or more Timing Advance Groups (TAGs)," and [0298] specifically states "a UE may adjust an uplink transmission timing based on a timing advanced command," i.e. the plurality of measurements may include time advances TAs), and precoding matrix indications (PMI) corresponding to the serving cell ([0265] describes "a wireless device 110 may indicate some beam pair quality parameters, comprising at least, one or more beam identifications; RSRP; Precoding Matrix Indicator (PMI)/Channel Quality Indicator (CQI)/Rank Indicator (RI) of a subset of configured beams," i.e. transmissions may include PMI or a plurality of other indicators).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective
filing date of the claimed invention to modify the combination of Jiang and Deenoo to include a reliability value with a reference value, and a plurality of measurement results such as RSRP, TA, and PMI as taught by Park, in order to improve [0228] enabling operations of single-carrier and/or multi-carrier communications, and [0333] enable robust operation.
Regarding claim 8, Jiang is not relied on for the claim language transmitting, to the serving cell, capability information comprising second information indicating support for a measurement reporting determination function. However, Deenoo teaches as such ([0111] describes "the WTRU 102A may indicate the capabilities of the WTRU 102A in terms of any of: (1) storage availability, (2) support of various AI filter architectures, (3) parameterizations, and/or (4) processing latency, among others. The WTRU capability may be indicated as a part of a RRC connection request and/or any other RRC message," i.e. any transmission to the serving cell may include capability information, such as parameters or type of AI filter (support for a measurement determination function)).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the terminal device to transmit capability information as taught by Deenoo, in order to aid in [0078] allowing the network controlled AI components in the processing chain to specifically adapt to different contexts (e.g., channel condition, quality of service, WTRU power saving, cell load, interference, and/or WTRU/NW capability, among others, i.e. terminal capabilities) and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
Regarding claim 9, Jiang teaches the determining of whether to transmit the measurement report to the serving cell- and determining to transmit the measurement report to the serving cell ([Figure 4, 410] transmission 415 may not occur until a threshold is met, i.e. determining to transmit the measurement report to the serving cell).
The combination of Jiang and Deenoo is not relied on for the claim language -comprises, based on the event occurrence reliability value being greater than the reference value. However, Park teaches as such ([0418] describes reliability thresholds being calculated when a separate threshold value is met, i.e. capabilities of determining reliability values being greater than or less than the reference thresholds. [0423] further describes comparing the reliability threshold value with separate threshold/reference values).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective
filing date of the claimed invention to modify the combination of Jiang and Deenoo to include the ability to compare the reliability value with a reference value, as taught by Park, in order to improve [0249] the ability to fulfill different services having different service requirements (e.g., data rate, latency, reliability), and therefore [0266] increase scheduling flexibility to allow the UE to communicate with different services/cells.
Regarding claim 10, Jiang teaches transmitting, to the serving cell, the measurement report- ([Figure 4, 415] transmitting to the serving cell the measurement report).
Jiang is not relied on for the claim language based on a second neural network model corresponding to a decoder trained based on an autoencoder not being stored in the serving cell, the transmitting of the measurement report to the serving cell comprises: inputting the latent vector to the second neural network model; and -comprising an output of the second neural network model. However, Deenoo teaches based on a second neural network model corresponding to a decoder trained based on an autoencoder not being stored in the serving cell, the transmitting of the measurement report to the serving cell (as explained in claim 3, see FIG. 3 and [0117]) comprises: inputting the latent vector to the second neural network model ([0141] referring to FIG. 4, " the plurality of inputs 420-1, 420-2 and 420-3 may input to the selected AI filter 320-3," wherein it is depicted the inputs 420-1, 420-2 and 420-3 are first outputted from other AI filters, i.e. AI filter 310-3 may be the second neural network model, additionally, the data may be compiled as a latent vector as described in [0083] - [0084], i.e. inputting the latent vector into a second neural network model); and -comprising an output of the second neural network model ([0141] transport block 430 in FIG. 4 is the output of the second neural network model).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the vector to be specifically a latent vector, which may be used as an input for neural networks based on an autoencoder as taught by Deenoo, in order to aid in [0019] improving access to one or more communication networks by [0082] enabling the AI component to learn complex behaviors which would otherwise be more difficult using legacy operations/methods such as input methods and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
Regarding claim 17, the claimed limitations of claim are rejected as the same reasons as set forth in claim 7.
Regarding claim 19, Jiang teaches determine- -whether to transmit the measurement report to the serving cell ([Figure 4, 410] transmission 415 may not occur until a threshold is met, i.e. determining to transmit the measurement report).
Jiang is not relied on for the claim language -based on the latent vector-. However, Deenoo teaches as such (described above the WTRU/UE/terminal may comprise the ability to make decisions based on the latent vector, see [0083] - [0084]).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the vector to be specifically a latent vector, which may be used as an input for neural networks as taught by Deenoo, in order to aid in [0019] improving access to one or more communication networks by [0082] enabling the AI component to learn complex behaviors which would otherwise be more difficult using legacy operations/methods such as input methods and therefore [0086] enable a neural network to track evolving conditions for a given task and [0081] implement unsupervised learning.
The combination of Jiang and Deenoo is not relied on for the claim language generate- -event occurrence reliability for an event related to the measurement report generate- and -based on whether the event occurrence reliability is greater than a reference value. However Park teaches generate- -event occurrence reliability for an event related to the measurement report ([0423] describes "the second access node may transmit information of latency and/or information of reliability of the first cell and/or the backhaul link when a configured condition/event occurs based on report configuration parameters (e.g., latency threshold value: 5 ms, reliability threshold value: packet loss rate 0.002) requested by the first access node via a status report request message," i.e. calculating/generating the reliability of a threshold value being met (an event), a plurality of other reliability values are also explained in [0418], wherein said reliability values and corresponding reference values are in response to an event related to “UE measurement reporting and control of the reporting” see [0205]), and -based on whether the event occurrence reliability is greater than a reference value ([0418] describes reliability thresholds being calculated when a separate threshold value is met, i.e. capabilities of determining reliability values being greater than or less than the reference thresholds. [0423] further describes comparing the reliability threshold value with separate threshold/reference values).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective
filing date of the claimed invention to modify the combination of Jiang and Deenoo to include the ability to compare the reliability value with a reference value, as taught by Park, in order to improve [0249] the ability to fulfill different services having different service requirements (e.g., data rate, latency, reliability), and therefore [0266] increase scheduling flexibility to allow the UE to communicate with different services/cells.
Claims 13 is rejected under 35 U.S.C. 103 as being unpatentable over Jiang et al. (US 2024/0080692 A1, hereinafter Jiang) and DEENOO et al. (US 2023/0389057 A1, hereinafter Deenoo) as applied in claims above, and further in view of TOKGOZ et al. (US 2022/0286215 A1, hereinafter Tokgoz).
Regarding claim 13, Jiang is not relied on for the claim language the first neural network model is generated in the terminal, and the second neural network model is generated in the serving cell. However, Deenoo teaches the first neural network model is generated in the terminal ([0076] describes "the WTRU may select an AI filter as a function of resources on which a first transmission was received and apply the data unit to be processed for transmission (e.g., a PDU) or parts thereof as an input to the selected AI filter," i.e. the a WTRU may generate the first neural network model. Furthermore, [0151] describes “the WTRU 102 may include memory (e.g., removeable 130 and/or non-removeable memory 132) to store (1) an AI filter model 210 of the AI filter 320 using a plurality AI nodes,” i.e. the AI filter (which includes the first neural network model) is stored and generated in the terminal itself, also see FIG. 4).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Jiang to include the ability for the first neural network model to be generated in the terminal as taught by Deenoo, in order to improve [0095] latency and/or maximum delay guarantees to [0102] optimize power consumption and performance of the terminal.
The combination of Jiang and Deenoo is not relied on for the claim language the second neural network model is generated in the serving cell. However, Tokgoz teaches [abstract] a method of wireless communication by a network device which includes receiving the location of a UE and neighboring cell information to train a neural network. Tokgoz also teaches the second neural network model is generated in the serving cell ([0092] “training of the neural network may be performed at various nodes. For example, training of the neural network may be performed at a serving cell”, i.e. one of the neural networks may be trained at the serving cell of the UE/terminal).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective
filing date of the claimed invention to modify the combination of Jiang and Deenoo to include the ability of a neural network to be generated in the serving cell, as taught by Tokgoz, in order to [0004] better support mobile broadband Internet access by improving spectral efficiency, lowering costs, improving services, making use of new spectrum, and better integrating with other open standards.
References Cited
Jiang, Jing et al. (2024). Spatial metric based mobility procedures using multi-port mobility reference signals (US 2024/0080692 A1). Filed 2022-09-06.
Deenoo, Yugeswar et al. (2023). Methods, apparatus, and systems for artificial intelligence (ai)-enabled filters in wireless systems (US 2023/0389057 A1). Filed 2021-10-19.
Park, Kyungmin et al. (2020). Base station backhaul link information (US 2020/0252847 A1). Filed 2020-02-06.
Tokgoz, Yeliz et al. (2022). Neural network-based spatial inter-cell interference learning (US 2022/0286215 A1). Filed 2021-03-02.
Other Pertinent References
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Aguirre, Sergio (2024). Flexible configuration of indoor radios to facilitate coexistence with outdoor radios (US 2024/0283626 A1). Filed 2023-02-17. Discloses a method for monitoring interference measurements of a radio in a cellular network and generating a matrix for said measurements. (abstract)
Jung, Ikjoo et al. (2026). Method, communication equipment, processing device, and storage medium for updating knowledge for semantic communication in wireless communication system (US 2026/0156488 A1). Filed 2021-12-02. Discloses a method for updating knowledge for semantic communication in a wireless communication system. (abstract)
Kim, Hyoeun et al. (2023). Artificial intelligence apparatus and method for detecting unseen class items thereof (US 2023/0140893 A1). Filed 2022-10-10. Discloses an AI apparatus for classifying and encoding data received from images. (abstract)
Ali, Samad et al. (2024). Obtaining machine learning (ml) models for secondary method of orientation detection in user equipment (ue) (US 2024/0155553 A1). Filed 2021-03-09. Discloses a method for a BS to receive a feature vector from a UE to be used in a ML training model. (abstract)
Zhang, Rensheng et al. (2022). Clustering cell sites according to signaling behavior (US 2022/0303796 A1). Filed 2021-03-16. Discloses a security management component (SMC) that can determine a neural network (NN) of NNs that can be representative of the cell network based on analysis of a first signal measurement data associated with the cells. (abstract)
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MATTHEW JAMES DWYER whose telephone number is (571)272-5121. The examiner can normally be reached M-F 6 a.m. - 3 p.m. EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Yuwen Pan can be reached at (571) 272-7855. 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.
/MATTHEW JAMES DWYER/Examiner, Art Unit 2649
/JOSHUA L SCHWARTZ/Primary Patent Examiner, Art Unit 2649
Prosecution Insights