Weighted Least Squares Based Spectral Precoder for OFDM Cognitive Radio

This letter presents a novel spectral precoding scheme for orthogonal frequency division multiplexing (OFDM) based cognitive radio (CR) systems. Unlike existing techniques, where the least squares approach leads to a biased solution causing more distortion on some subcarriers than others, the proposed precoder adopts a weighted least squares approach and uses Lawson´s algorithm for obtaining desired weights. Simulation results show that the proposed precoder provides significant BER performance gain without much impact on computational complexity and sidelobe suppression. Moreover, the proposed precoder when combined with decision directed iterative decoding at the receiver provides bit error rate (BER) performance similar to that of an un-suppressed OFDM system.