Abstract

Turbo equalization improves communication system performance by iteratively exchanging information between soft-input soft-output (SISO) equalizer and SISO channel decoder. The trellis-based maximum a posteriori probability (MAP) algorithm serves as the optimum SISO equalizer for turbo equalization. However, MAP algorithm is unsuitable for systems with large modulation constellation size and severe inter-symbol interference (ISI) due to its prohibitively high computational complexity. In this paper, an improved SISO block decision feedback equalizer (BDFE) is proposed for low complexity turbo equalization. Unlike other sub-optimum equalizers which perform symbol by symbol detection, the proposed equalizer generates the soft output for each data bit by collecting information from a sequence of samples as in MAP algorithm. The sequence-based equalization is enabled by using not only soft a priori input from channel decoder, but also hard a priori information obtained from BDFE in previous iteration. The combination of soft a priori information and hard a priori information renders better performance with less iterations compared to other sub-optimum algorithms. In addition, the computational complexity of the proposed algorithm is on the same order as conventional SISO BDFE algorithm, and is much lower compared to the trellis-based MAP algorithm.

Department(s)

Electrical and Computer Engineering

Sponsor(s)

National Science Foundation (U.S.)
United States. Office of Naval Research

Keywords and Phrases

Block Decision Feedback Equalization; Iterative Equalization; Sequence Detection; Turbo Equalization

International Standard Serial Number (ISSN)

1536-1276

Document Type

Article - Journal

Document Version

Final Version

File Type

text

Language(s)

English

Rights

© 2008 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.

Publication Date

01 Jan 2008

Share

 
COinS