Work

Enhanced OFDM Channel Estimation via DFT-Based Precomputed Matrices

Year: 2025

Type: article

Abstract: Orthogonal Frequency Division Multiplexing (OFDM) modulation currently dominates the physical layer design in modern transmission systems. Its primary advantage is the simple reconstruction of channel frequency response (CFR). However, the Least Squares (LS) algorithm commonly used here is prone to significant estimation errors due to noise interference. A promising and relatively simple alternative is a DFT-based strategy that uses a pre-computed refinement/correction matrix to improve estimation performance. This paper investigates two implementation approaches for CFR reconstruction with pre-computed matrices. Focusing on multiplication operations, a threshold number of active subcarriers was identified at which these two implementations exhibit comparable numerical complexity. A numerical performance factor was defined and a detailed performance analysis was carried out for different guard interval (GI) lengths and the number of active subcarriers in the OFDM signal. Additionally, to maintain channel estimation quality irrespective of GI length, a channel impulse response (CIR) energy detection procedure was introduced. This procedure refines the results of the symbol synchronization process and, by using the circular shift property, preserves constant values of the precomputed matrix coefficients without system performance loss, as measured by Bit Error Rate (BER) and Mean Square Error (MSE) metrics.

Source: Electronics

DOI: 10.3390/electronics14173378

Authors: Grzegorz Dziwoki, Jacek Izydorczyk, Marcin Kucharczyk

Institution: Silesian University of Technology

Language: English

Topic: Advanced Wireless Communication Techniques

Subfield: Electrical and Electronic Engineering

Field: Engineering

Domain: Physical Sciences

Open Access status: gold

Estimated APC paid: $2,896

Funder: Ministry of Science and Higher Education of the Russian Federation