# US 12,771,419 B2 · Two stage machine learning based channel state feedback > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,771,419 B2 - Document ID: US12771419B2 - Status: Granted - Kind: B2 - Application number: 18/335,726 - Filed: 2023-06-15 - Published: 2024-12-19 - Granted: 2026-10-20 - Earliest claimed: 2023-06-15 - Assignees: QUALCOMM Incorporated - Applicants: QUALCOMM Incorporated - Inventors: Pavan Kumar Vitthaladevuni; Jay Kumar Sundararajan; Taesang Yoo; Krishna Kiran Mukkavilli; Naga Bhushan - Web page: https://patentgrep.com/patent/US12771419B2 ## CPC - H04B 7/0626: {Channel coefficients, e.g. channel state information [CSI]} - H04W 8/24: Transfer of terminal data - H04B: TRANSMISSION - H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72) ## Abstract Methods, systems, and devices for wireless communication are described. A user equipment (UE) may select a non-Discrete Fourier Transform (non-DFT) codebook of a set of non-DFT codebooks associated with a channel state feedback message. The UE may determine a set of singular vectors associated with the non-DFT codebook based on a first machine learning model, where the set of singular vectors corresponds to a subspace associated with the non-DFT codebook. The UE may compress the non-DFT codebook, the set of singular vectors, or both, based on a second machine learning model. The UE may transmit the channel state feedback message including the compressed non-DFT codebook, the compressed set of singular vectors, or both. ## Claims ### Claim 1 A user equipment (UE) for wireless communications, comprising: one or more memories storing processor-executable code; and one or more processors coupled with the one or more memories and individually or collectively operable to execute the code to cause the UE to: select a non-Discrete Fourier Transform (non-DFT) codebook of a set of non-DFT codebooks associated with a channel state feedback message; determine a set of singular vectors associated with the non-DFT codebook based at least in part on a first machine learning model, wherein the set of singular vectors corresponds to a subspace associated with the non-DFT codebook; compress the non-DFT codebook, the set of singular vectors, or both, based at least in part on a second machine learning model; and transmit the channel state feedback message comprising the compressed non-DFT codebook, the compressed set of singular vectors, or both. ### Claim 2 (depends on 1) The UE of claim 1, wherein the one or more processors are individually or collectively further operable to execute the code to cause the UE to: transmit a UE capability message indicating that a machine learning feedback mode is supported at the UE, wherein to determine the set of singular vectors associated with the non-DFT codebook, to compress the non-DFT codebook, to compress the set of singular vectors, or a combination thereof is based at least in part on the UE capability message. ### Claim 3 (depends on 1) The UE of claim 1, wherein the one or more processors are individually or collectively further operable to execute the code to cause the UE to: receive control signaling indicating a reporting configuration associated with the compressed non-DFT codebook, the compressed set of singular vectors, or both, based at least in part on a UE capability message, wherein to transmit the compressed non-DFT codebook, the compressed set of singular vectors, or both, is based at least in part on the reporting configuration. ### Claim 4 (depends on 3) The UE of claim 3, wherein, to transmit the compressed non-DFT codebook, the compressed set of singular vectors, or both, the one or more processors are individually or collectively operable to execute the code to cause the UE to: jointly transmit the compressed non-DFT codebook and the compressed set of singular vectors based at least in part on the reporting configuration. ### Claim 5 (depends on 3) The UE of claim 3, wherein, to transmit the compressed non-DFT codebook, the compressed set of singular vectors, or both, the one or more processors are individually or collectively operable to execute the code to cause the UE to: separately transmit the compressed non-DFT codebook and the compressed set of singular vectors based at least in part on the reporting configuration. ### Claim 6 (depends on 3) The UE of claim 3, wherein the one or more processors are individually or collectively further operable to execute the code to cause the UE to: determine a first periodicity associated with reporting the compressed non-DFT codebook based at least in part on the reporting configuration; and determine a second periodicity associated with reporting the compressed set of singular vectors based at least in part on the reporting configuration, wherein to transmit the compressed non-DFT codebook is based at least in part on the first periodicity associated with reporting the compressed non-DFT codebook, and wherein to transmit the compressed set of singular vectors is based at least in part on the second periodicity associated with reporting the compressed set of singular vectors. ### Claim 7 (depends on 6) The UE of claim 6, wherein the first periodicity associated with reporting the compressed non-DFT codebook is different than the second periodicity associated with reporting the compressed set of singular vectors. ### Claim 8 (depends on 7) The UE of claim 7, wherein the first periodicity associated with reporting the compressed non-DFT codebook is less than the second periodicity associated with reporting the compressed set of singular vectors. ### Claim 9 (depends on 3) The UE of claim 3, wherein, to receive the control signaling, the one or more processors are individually or collectively operable to execute the code to cause the UE to: receive a radio resource control message indicating the reporting configuration associated with the compressed non-DFT codebook, the compressed set of singular vectors, or both. ### Claim 10 (depends on 3) The UE of claim 3, wherein, to receive the control signaling, the one or more processors are individually or collectively operable to execute the code to cause the UE to: receive a medium access control-control element indicating the reporting configuration associated with the compressed non-DFT codebook, the compressed set of singular vectors, or both. ### Claim 11 (depends on 3) The UE of claim 3, wherein, to receive the control signaling, the one or more processors are individually or collectively operable to execute the code to cause the UE to: receive downlink control information indicating the reporting configuration associated with the compressed non-DFT codebook, the compressed set of singular vectors, or both. ### Claim 12 (depends on 1) The UE of claim 1, wherein the compressed non-DFT codebook corresponds to a first quantity of bits, and wherein the compressed set of singular vectors correspond to a second quantity of bits different than the first quantity of bits. ### Claim 13 (depends on 12) The UE of claim 12, wherein the first quantity of bits is greater than the second quantity of bits. ### Claim 14 (depends on 1) The UE of claim 1, wherein the one or more processors are individually or collectively further operable to execute the code to cause the UE to: determine a change in a parameter associated with the UE, wherein the parameter comprises a speed associated with the UE, a location associated with the UE, an altitude associated with the UE, or a combination thereof, wherein to transmit the compressed non-DFT codebook, the compressed set of singular vectors, or both, is based at least in part on the change in the parameter associated with the UE. ### Claim 15 (depends on 14) The UE of claim 14, wherein the one or more processors are individually or collectively further operable to execute the code to cause the UE to: transmit first control signaling indicating a request to adjust a periodicity associated with reporting the compressed non-DFT codebook, the compressed set of singular vectors, or both, based at least in part on the change in the parameter associated with the UE; and receive second control signaling indicating an adjusted periodicity associated with reporting the compressed non-DFT codebook, the compressed set of singular vectors, or both, based at least in part on the first control signaling, wherein to transmit the compressed non-DFT codebook, the compressed set of singular vectors, or both, is based at least in part on the adjusted periodicity. ### Claim 16 A method for wireless communications by a user equipment (UE), comprising: selecting a non-Discrete Fourier Transform (non-DFT) codebook of a set of non-DFT codebooks associated with a channel state feedback message; determining a set of singular vectors associated with the non-DFT codebook based at least in part on a first machine learning model, wherein the set of singular vectors corresponds to a subspace associated with the non-DFT codebook; compressing the non-DFT codebook, the set of singular vectors, or both, based at least in part on a second machine learning model; and transmitting the channel state feedback message comprising the compressed non-DFT codebook, the compressed set of singular vectors, or both.