基于MUSIC算法的OFDM信号识别方法

摘要

针对多径信道下传统的OFDM信号识别方法存在循环前缀较短时估计性能不高、所需OFDM符号数过多等问题,提出一种基于MUSIC算法的OFDM信号识别方法.该方法首先分析了OFDM信号和单载波信号的结构特点,然后对两类信号的自相关矩阵进行奇异值分解,提取奇异值矩阵,最后根据奇异值矩阵中较大非零奇异值的个数实现OFDM信号的识别.仿真实验表明,该方法仅需较少的OFDM符号就可以实现短循环前缀OFDM信号的识别,且识别性能优于传统方法.%Aiming at the poor performance and the large required OFDM symbol number of conventional identification methods for the short cyclic prefix OFDM signal,we proposed a identification method for OFDM signal based on MU-SIC algorithm.The method first analyzes the structure characteristic of OFDM signal,then implements singular value decomposition for the autocorrelation matrix and extracts all the singular value matrix.Finally,the algorithm identifies the OFDM signal according to the number of larger non-zero singular values of the matrix.The experimental results shows that the proposed algorithm just need less OFDM symbols to be able to effectively identify the short cyclic prefix OFDM signal and has better performance than the conventional algorithm.