基于单通道信号及NLR算法的OFDM系统频偏估计

发布时间：2018-05-21 21:08

本文选题：单通道信号 + 载波频率偏移　；参考：《仪器仪表学报》2014年12期

【摘要】：提出了一种新的OFDM系统频偏估计方法,该方法首先利用相关检测技术从单路的接收信号中获得独立分量分析(ICA)技术所需的多路混合信号,相比于传统的估计方法减少了接收通道;然后使用非参数似然比(NRL)算法来实现对混合信号的分离,同时利用Hilbert变换估计OFDM系统的载波频偏(CFO)。NLR算法不需要假设信号的非高斯统计特性,且与MMI-ICA相比,NLR-ICA可以显著改善分离信号的信号干扰比(SIR)。与传统的频偏估计方法相比较,该方法不需要借助同步训练符号,不仅节省了信道带宽,而且克服了信道参数对CFO估计的影响,因此具有一定的优越性。仿真结果表明,该方法具有良好的特性。NLR算法的实现不依赖于信号的概率分布,克服了传统ICA算法对信号分布要求,有利于对通信信号的分离。
[Abstract]:In this paper, a new frequency offset estimation method for OFDM systems is proposed. In this method, the multi-channel mixed signals required by the independent component analysis (ICA) technique are obtained from a single received signal by correlation detection technique. Compared with the traditional estimation method, the received channel is reduced, and then the nonparametric likelihood ratio (NRL) algorithm is used to separate the mixed signals. At the same time, the Hilbert transform is used to estimate the carrier frequency offset of the OFDM system without assuming the non-Gao Si statistical characteristics of the signal, and compared with the MMI-ICA, the NLR-ICA can significantly improve the signal interference ratio of the separated signal. Compared with the traditional frequency offset estimation method, this method not only saves channel bandwidth, but also overcomes the influence of channel parameters on CFO estimation, so it has some advantages. The simulation results show that the realization of the method does not depend on the probability distribution of the signal. It overcomes the requirement of the traditional ICA algorithm for the signal distribution and is advantageous to the separation of the communication signal.
【作者单位】：集美大学信息工程学院;
【基金】：福建省科技计划重点项目(2013H0035) 厦门市科技计划项目(3502Z20123028)资助
【分类号】：TN929.53

【参考文献】