Granted Patent B2

US 12,700,908 · App. 18/748,613 · Granted Aug 4, 2026

Inventor: Alexander Chiskis (Tel Aviv, IL)

Assignee: Sequans Communications SA

H04B7/0663H04B7/0626H04W24/10

Abstract

Claims (43)

1 . A wireless communication system, comprising:

a User Equipment (UE) configured to:

calculate downlink (DL) channel-related information,

preprocess the DL channel-related information using a unitary matrix to generate post-processing information,

compress the post-processing information to generate transmission data, and

communicate the transmission data over a communication channel; and

a Base Station configured to:

receive the transmission data,

decompress the received transmission data to reconstruct the post-processing information, and

construct a DL precoder utilizing the reconstructed post-processing information;

wherein the preprocessing of the DL channel-related information further includes using diagonal unitary matrices to facilitate compression of the post-processing information.

2 . The wireless communication system of claim 1 , wherein the unitary matrix is constructed to optimize at least one of: DL communication performance, reduce compressed size, or balance both DL communication performance and compressed size.

3 . The wireless communication system of claim 2 , wherein optimization of DL communication performance uses at least one of: a metric related to DL communication performance, DL capacity, power-related metrics, cosine similarity-related criteria (CS), and a reconstruction error metric, wherein the reconstruction error metric relates to a difference between UE post-processing based information and an expected base station reconstructed post-processing based information, and wherein the optimization uses unitary matrix components directly or via a barrier function method.

4 . The wireless communication system of claim 1 , wherein the unitary matrix is configured to be dynamically adjustable based on feedback from the base station.

5 . The wireless communication system of claim 1 , wherein the compression of the post-processing information by the UE includes a lossy compression algorithm to reduce an amount of transmission data.

6 . The wireless communication system of claim 1 , wherein preprocessing the DL channel-related information using a unitary matrix performed by the UE includes an application of a Fourier Transform (FT) to the DL channel-related information after using the unitary matrix.

7 . The wireless communication system of claim 1 , wherein the base station is configured to perform a DL channel quality estimation based on the reconstructed post-processing information and adjust the base station's operational parameters accordingly to optimize system performance.

8 . The wireless communication system of claim 1 , wherein the system is compatible with a 5G Standard, 5G Advanced Standard, or 6G Standard, and includes a mechanism to switch between standard and modified compression modes using a communicated system bit indicator.

9 . The wireless communication system of claim 1 , wherein an input to a CSI-compression engine incorporates precoder matrix factorization, with the precoder being W n =W (A)n W (B)n .

10 . The wireless communication system of claim 1 , wherein the preprocessing includes a unitary modification to set initial parameters for artificial intelligence/machine learning (AI/ML) algorithms, facilitating faster learning and re-learning.

11 . A wireless communication system, comprising:

a User Equipment (UE) configured to:

calculate downlink (DL) channel-related information,

preprocess the DL channel-related information using a unitary matrix to generate post-processing information, compress the post-processing information to generate transmission data, and communicate the transmission data over a communication channel; and

a Base Station configured to:

receive the transmission data,

decompress the received transmission data to reconstruct the post-processing information, and

construct a DL precoder utilizing the reconstructed post-processing information, wherein:

the preprocessing of the DL channel-related information by the UE includes modifying input matrix columns using an auxiliary synchronization vector (ASV) approach, wherein the ASV is constructed based on one of:

a per-layer strategy to modify columns of the input matrix wherein the ASV is optimized to achieve at least one of improved DL Communication performance, improved compression, or a balance between DL communication performance and compression,

an averaged correlation channel matrix, using the best eigenvectors of an averaged correlation matrix or best eigenvectors of a Singular Value Decomposition (SVD) of the input matrix, or

columns of the input matrix, selected based on a predetermined strategy, or

modifying columns using at least one of a Component Sum Approach (CSA) or a Weighted CSA (WCSA) wherein weights for the WCSA are optimized based on amplitudes of the input columns.

12 . A wireless communication system, comprising:

a User Equipment (UE) configured to:

calculate downlink (DL) channel-related information,

preprocess the DL channel-related information using a unitary matrix to generate post-processing information,

compress the post-processing information to generate transmission data, and

communicate the transmission data over a communication channel; and

a Base Station configured to:

receive the transmission data,

decompress the received transmission data to reconstruct the post-processing information, and

construct a DL precoder utilizing the reconstructed post-processing information, wherein the DL precoder incorporates interference treatment by constructing the precoder from eigenvectors of a correlation matrix of a modified channel.

Assignments (1)

Continuity (1)

References Cited (10)