Open AccessIterative receiver with soft‐input‐based‐channel estimation for orthogonal frequency division multiplexing‐interleave division multiple access systems
Author(s) -
Oyerinde Olutayo O.,
Mneney Stanley H.
Publication year - 2014
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/iet-com.2013.1171
Subject(s) - estimator , algorithm , orthogonal frequency division multiplexing , computer science , minimum mean square error , channel (broadcasting) , mean squared error , noise power , fading , bit error rate , multiuser detection , mathematics , decoding methods , statistics , detector , power (physics) , telecommunications , physics , quantum mechanics
This article presents two sets of channel estimation methods that employ soft input from the decoder to enhance channel estimation process for orthogonal frequency division multiplexing ‐interleave division multiple access (OFDM‐IDMA) system. The first channel estimation scheme exploits both time and frequency domains for channel estimation and prediction. The estimator exploits turbo principle while using iterative based sequential linear minimum mean square error (ISLMMSE) for channel transfer function (CTF) estimation, and regularised variable step size normalised least mean square (ℓ 1 ‐VSSNLMS) algorithm for channel impulse response (CIR) prediction. The channel estimation scheme exchanges information with the Multi‐User Detector (MUD), and also employs soft information feedback from the decoder for the enhancement of channel estimation. The second iterative channel estimation scheme, in time domain, is based on regularised noise power estimate‐based variable forgetting factor recursive least square (ℓ 1 ‐NPEVFF‐RLS)‐based CIR estimator. The performances of the proposed estimators are documented through computer simulation. Their comparative performances with other schemes in the literature are presented in this article. From the simulation results, the two proposed channel estimators exhibit better performance in comparison with other schemes in the literature, but with higher computational complexities. However, of the two proposed methods, the ℓ 1 ‐NPEVFF‐RLS‐based CIR estimator that exhibits almost the same performance as the combined ISLMMSE‐based CTF estimator and ℓ 1 ‐VSSNLMS‐based predictor exhibits lower computational complexity.