Iterative Decision Feedback Equalization Using Online Prediction

TL;DR

This paper introduces an online prediction method for estimating decision feedback reliability in static-filter SISO MMSE FIR decision feedback equalizers, enhancing detection performance for high-data-rate applications.

Contribution

It proposes a novel online prediction approach for feedback reliability estimation, improving static-filter DFE performance over existing methods.

Findings

Improved detection performance with the new online prediction method.

Static-filter DFE remains effective for high-spectral efficiency.

Finite-length and asymptotic analysis support the approach.

Abstract

In this article, a new category of soft-input soft-output (SISO) minimum-mean square error (MMSE) finite-impulse response (FIR) decision feedback equalizers (DFEs) with iteration-wise static filters (i.e. iteration variant) is investigated. It has been recently shown that SISO MMSE DFE with dynamic filters (i.e. time-varying) reaches very attractive operating points for high-data rate applications, when compared to alternative turbo-equalizers of the same category, thanks to sequential estimation of data symbols [1]. However the dependence of filters on the feedback incurs high amount of latency and computational costs, hence SISO MMSE DFEs with static filters provide an attractive alternative for computational complexity-performance trade-off. However, the latter category of receivers faces a fundamental design issue on the estimation of the decision feedback reliability for filter computation. To address this issue, a novel approach to decision feedback reliability estimation through online prediction is proposed and applied for SISO FIR DFE with either a posteriori probability (APP) or expectation propagation (EP) based soft feedback. This novel method for filter computation is shown to improve detection performance compared to previously known alternative methods, and finite-length and asymptotic analysis show that DFE with static filters still remains well-suited for high-spectral efficiency applications.