New efficient optimal FIR equalizer design

Publication Date

4-1-2024

Document Type

Article

Publication Title

Physical Communication

Volume

63

DOI

10.1016/j.phycom.2024.102298

Abstract

Finite impulse response (FIR) filters are commonly used as linear equalizers in vehicular communication systems due to their advantages including linear-phase property, stability, and implementation efficiency. As the perfect equalizer is actually the inverse of a channel filter, it would inevitably possess a mixing-phase transfer function very often. However, there hardly exists any systematic approach governing how to design an FIR equalizer by the optimal approximation of the perfect equalizer subject to a target signal-to-interference ratio (SIR). Therefore, in this correspondence, we make the first-ever attempt to formulate a new optimization problem for designing an FIR equalizer to determine the minimum FIR-equalizer length subject to a target (maximum allowable) SIR. We propose a novel algorithm to solve this optimization problem. The conventional approach is to apply the least-squares (LS) method to determine the optimal FIR coefficients given an FIR-equalizer length; then the SIRs resulting from various equalizer lengths will be compared to determine the minimum among them in compliance with the target SIR. In other words, the conventional FIR-equalizer design approach is simply based on the exhaustive trial-and-error analysis. In this work, we propose a novel efficient FIR-equalizer design algorithm, which can determine the minimum equalizer length subject to a maximum allowable SIR in a fast manner. The advantage of our proposed new FIR-equalizer design algorithm is justified by the run-time study in comparison with the conventional scheme.

Keywords

Finite impulse response (FIR) equalizer, Inverse system, Signal-to-interference ratio (SIR), Vehicular communication system, ℒ approximation-error 2

Department

Applied Data Science

Link to Full Text

Share

COinS