Comparison of Channel State Information Estimation Using SLM and Clipping -based PAPR Reduction Methods

Abstract Channel estimation is a crucial issue in orthogonal frequency-division multiplexing (OFDM) as well as in all multicarrier systems. However, OFDM suffers from a major setback, the peak-to-average power ratio (PAPR). PAPR can be solved using a number of available techniques in literature, such as coding, active constellation extension, amplitude clipping, and selected mapping. The coding approach presents a disadvantage, represented by redundant data that significantly reduce the bit rate. The active constellation extension is an effective method; however, it requires higher transmission power. The clipping method is the simplest, but it produces high bit error rate (BER) degradation. Selected mapping (SLM) is the best among the available methods; however, it sends several bits as side information. In this study, we compare the clipping and SLM methods and show how the channel state information (CSI) estimation is affected in both techniques. Simulation results show that the SLM method is more effective than the clipping technique. The BER significantly increases when the clipping method is used because of the inaccurate estimation of CSI when the high peaks are clipped, such as in the case of the inserted pilots.