RLS channel estimation with superimposed training sequence in OFDM systems

Abstract

In this paper, A Recursive Least Squares (RLS) channel estimator with improved decision-directed algorithm (referred as DDA2-RLS) is proposed based on the superimposed training sequence in orthogonal frequency division multiplexing (OFDM) systems. The DDA2-RLS is exploited to further eliminate the interference driven by the superimposed unknown information data. Then, the theoretical analysis for DDA2-RLS algorithm with superimposed training sequence that uses the constant Pseudo-Noise (PN) sequence is given. It is shown that the proposed DDA2-RLS algorithm can improve the channel estimation performance compared with the original RLS and decision-directed algorithm (DDA) RLS algorithms. Simulations results demonstrate the effectiveness of the proposed DDA2-RLS, and the performance is close to the theoretical analysis compared with original RLS and DDA-RLS algorithms.