OFDM de-noising with RLS adaptive filter

Orthogonal Frequency Division Multiplexing (OFDM) is designed to improve the spectral efficiency and combat multipath fading by dividing the wide band frequency selective fading channel into many narrow flat sub-channels. The noise in the wireless communication channel may damage the OFDM transmission signal and reduce the signal-to-noise ratio (SNR) at the receiver, which limits the bit error rate (BER) performance and data rate. The adaptive filter has been widely used in random signal processing to deal with random signal and noise. In this paper, we propose a scheme using recursive least square (RLS) adaptive filter to achieve OFDM de-noising in the wireless communication system. RLS filter is an efficient fast convergence data filtering method, which can track the changes of the signal and noise by using exponential weighting and a forgetting factor, and self-adjust the current filter parameters to reach the best filtering performance. Simulation results show that the larger filter length results in better de-noising performance, and the RLS filter can remove the noise of OFDM signal efficiently.