PAPR Reduction In OFDM System And Implementation On FPGA

Orthogonal Frequency Division Multiplexing( OFDM) is a parallel transmission technique that divides a high rate data into many low rate data stream. Compared with the traditional single-carrier technique, OFDM has higher spectrum efficiency by using orthogonal sub-carriers to transmit the information. OFDM also has high capability of anti-multipath fading and anti-ISI, what makes it become hot research of new generation of mobile communication technology.As one of the main problems of OFDM system, high Peak-to-Average Power Ratio(PAPR) hinders the development of OFDM technique. Since each OFDM signal is the sum of many independent signals modulated onto sub-carriers, the OFDM signal with high PAPR will be generated when the independent signals have same or similar phases. It will cause in-band distortion and out-band interference to send such a signal with large dynamic range into nonlinear devices. As a result, the performance of OFDM system will become worse.To solve the problem of high PAPR, the paper detailedly discusses four typical algorithms to reduce PAPR of OFDM system:Clipping, Companding Transform, Slective Mapping, Partial Transmit Sequence. This paper firstly introduces Clipping and points out the traditional Clipping algorithm with a problem of peak regrowth.Then Clipping Filtering algorithm and Repeat Clipping Filtering algorithm are discussed. Based on principle analysis and Matlab simulation, this paper points out the PAPR performance of RCF is obvious but its BER performance is not good enough. Then this paper analyzes Companding Transform algorithm. Based on study of μ principle Companding Transform algorithm, this paper simulates μ principle algorithm on Matlab. Results show that this algorithm is although simple in implementation but with a serious out-band interference to gain BER. To avoid the decrease on BER performance, SLM and PTS are analyzed in detailed. The Matlab simulation indicates that both of them will increase the complexity of system. Finally considering performance of algorithm and complexity of implementation, this paper designs a PAPR sheme of RCF and implements it on FPGA.