A Low-Complexity SLM-Based Peak-to-Average Power Ratio Reduction Technique Using Reed-Solomon Coding for OFDM Systems

• Main Author: 林新智
• Other Authors: 王晉良
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/37089960370928335318

碩士 === 國立清華大學 === 通訊工程研究所 === 95 === Orthogonal frequency division multiplexing (OFDM) is an efficient transmission scheme for multicarrier systems. Its major advantages include high data rate and the immunity to multipath fading channels. One of the major problems associated with OFDM signals is the high peak-to-average power ratio (PAPR), which may cause distortion when the OFDM signals pass through a nonlinear amplifier. A number of techniques have been proposed for PAPR reduction. These techniques can be roughly classified as the signal distortion methods, the coding methods, and the multiple signal representation (MSR) methods. All these methods have their own drawbacks, such as signal distortion, limitation of the number of subcarriers, requirement of side information, and high computational complexity.In this thesis, we propose a new selective mapping (SLM)-based method for PAPR reduction. The proposed method does not suffer the drawbacks mentioned above. By sacrificing a little part of error correcting capacity of channel coding, the proposed SLM-based method generates a set of candidate signals by adding the symbol block after channel coding with a set of different random error patterns. Then the candidate signal with the lowest PAPR is chosen for transmission. To further reduce the complexity of the proposed method, the set of random error patterns are generated in advance and the corresponding IFFT output sequences are prestored in a ROM. Then, the candidate signals of the proposed method can be produced by only one IFFT computation and some extra additions. It is shown that the computational complexity of the proposed SLM-based method is much lower than the ordinary SLM method that needs a set of U IFFT computations, where U is the number of candidate signals. Computer simulation results demonstrate that the proposed method has a reasonable and acceptable performance loss in terms of PAPR reduction and symbol error rate, as compared to the ordinary SLM scheme.