Identification of frequency non-selective fading channels using decision feedback and adaptive linear prediction

In this paper we propose an algorithm for tracking the phase and amplitude of frequency nonselective fading channels. In contrast to Irvine-McLane's (see IEEE Journal on Selected Areas in Communications, vol.10, no.8, p.1289-1299, 1992) symbol-aided plus decision-directed (SADD) tracker which mainly uses training symbols for channel estimation, the new algorithm exploits data-bearing-bearing symbol correlation of fading channels. This algorithm combines decision feedback and adaptive linear prediction (DFALP) by using tentative coherent detection and the least mean square (LMS) algorithm. Both simulation results and theoretical analysis demonstrate that the performance of this algorithm compares favorably with the SADD algorithm in terms of decision delay, mean-square error and bit-error-rate. The algorithm is also applied to a system similar to EIA/TIA recommendation IS-54 and has achieved 3 dB gain using the same channel coding and similar interleaving. >