研究未知复杂信号环境,在正交频分复用(OFDM)盲子载波频偏估计中,常规ESPRIT算法进行估计,要求在高信噪比、加性高斯白噪声环境下,否则估计性能下降.为了使信道提高性能,提出一种基于高阶累积量的ESPRIT估计OFDM载波频偏的方法,改善了在现实复杂应用环境下性能下降的问题.由于高阶累积量能有效抑制平稳噪声和非平稳噪声,算法充分利用基带信号四阶累积量的对焦切片重构了含有频偏的新矩阵,再从特征值分解中可得到相应估计值.不仅解决了ESPRIT算法应用环境单一,对信噪比要求高等问题,且在上述噪声背景下,与同类算法相比较,分辨能力和估计精度大大提高,通过仿真实验验证了算法的有效性,并可推广应用.%Presently, in blind carrier frequency offset estimation over OFDM, when applying common ESPRIT algorithm, the signals with high SNR and AWGN channel are required to achieve good performance. In this paper, a novel ESPRIT algorithm based on higher-order cumulants to estimate carrier frequency offset for OFDM system is presented. The modified method can be used in varies of complicated channel. Due to the fact that the high order cumulants can suppress the stationary and non stationary Gaussian noise, the method sufficiently made use of the diagonally slice of four order cumulants of base band signals to construct the new matrix which included carrier frequency offset, CFO was obtained from the decomposition of above matrix. It solved the problem that ESPRIT algorithm can not estimate complicated channel and demanded high-signal to SNR, meanwhile, its spectrum resolving power is obviously improved compared with other algorithm. The proposed method can also popularized on OFDM signal with 64 QAM and high order PSK modulation. And simulation proves the effectiveness of the proposed method.
展开▼
