Delayed decision feed-back sequence estimation OFDM equalization with channel prediction

Abstract

This paper proposes a low-complexity equalizer for orthogonal frequency division multiplexing (OFDM) systems under time-varying frequency-selective fading environments. The equalization uses suboptimum maximum likelihood sequence estimation (MLSE) in cooperation with semi-blind linear prediction (SBLP) in the estimation of the channel frequency responses. When the number of pilot subcarriers is far less than the total subcarrier numbers, it reduces the Mean Square Error in the channel estimation compared to conventional linear interpolation and LMMSE (Linear Minimum Mean Square Error) algorithms. Furthermore, it doesn't require the knowledge of the statistics of the channel, which is necessary in LMMSE. In practical application, these parameters cannot be easily obtained and small estimation errors cause significant performance degradation. Theoretic analysis and simulation results demonstrate that the proposed equalizer can effectively increase the performance of OFDM receivers in time-varying fading channel.