低峰均值低密度奇偶校驗碼之搜尋演算法

Abstract

Orthogonal Frequency Division Multiplexing (OFDM) is popular modulation scheme for wireless system. Because OFDM has advantages of multipath-delay-spread tolerance, immunity to frequency selective fading channel and high spectral efficiency, many wireless systems have adopted the OFDM technique as a choice. However, Peak-to-average ratio (PAPR) is a main drawback of OFDM. PAPR problem is caused by Inverse Discrete Fourier Transform (IDFT). IDFT makes the OFDM symbols with high PAPR. High PAPR increases the complexity and the cost of an OFDM system. So far, many various PAPR reduction schemes have been published. Most of them have high online computation complexity and must modify the architecture of OFDM system. We advocate an idea to solve PAPR problem in this thesis. The idea satisfies to use offline computation to replace online computation without modifying the architecture. We want to reduce the PAPR by finding out low PAPR codes. However, the solution space is too large, so exhaustive search is not efficient. We want to search low PAPR codes with search algorithm, so an efficient criterion is needed. We explain the criterion in our thesis. The target code of the thesis is the Low-Density Parity-Check (LDPC) code. The LDPC codes have received a lot of attention recently because of there excellent performance and low decoding complexity. Finally, some simulated result is presented in the thesis.