A Fast Dictionary Learning Algorithm for CSI Feedback in Massive MIMO FDD Systems

Gadamsetty, PK and Hari, KVS (2023) A Fast Dictionary Learning Algorithm for CSI Feedback in Massive MIMO FDD Systems. In: 2023 National Conference on Communications, NCC 2023, 23 - 26 February 2023, Guwahati.

PDF ncc_2023.pdf - Published Version Restricted to Registered users only Download (481kB) | Request a copy

Abstract

In a massive multiple-input multiple-output (MIMO) frequency division duplex (FDD) system, it is required to compress the channel state information (CSI) and feed it back to base station (BS). In this paper, we primarily focus on the compressive sensing (CS)-based feedback design and propose a fast dictionary learning (FDL) algorithm to update the singular vectors of matrices in the K-singular value decomposition (K-SVD) algorithm. The proposed FDL algorithm is a variation of the existing K-SVD algorithm with low computational complexity. Simulation results also show that the proposed method's estimated channel has better normalized mean-squared error (NMSE) performance than the estimated channel using a traditional Discrete Fourier transform (DFT) dictionary. Also, the proposed method's estimated channel has comparable performance with the K-SVD algorithm but with reduced computational complexity ranging from 18 to 45, which is significant. © 2023 IEEE.

Item Type:	Conference Paper
Publication:	2023 National Conference on Communications, NCC 2023
Publisher:	Institute of Electrical and Electronics Engineers Inc.
Additional Information:	The copyright for this article belongs to Institute of Electrical and Electronics Engineers Inc.
Keywords:	Compressed sensing; Computational complexity; Design for testability; Discrete Fourier transforms; Frequency division multiplexing; Learning algorithms; Mean square error; MIMO systems; Singular value decomposition, Channel state information; Channel-state information; Compressive sensing; Dictionary designs; Dictionary learning algorithms; K-singular value decomposition; Massive multiple-input multiple-output; Multiple inputs; Multiple outputs, Channel state information
Department/Centre:	Division of Electrical Sciences > Electrical Communication Engineering
Date Deposited:	25 May 2023 04:20
Last Modified:	25 May 2023 04:20
URI:	https://eprints.iisc.ac.in/id/eprint/81531

Actions (login required)

View Item