Underlay认知网络中时频重叠OFDM信号的信噪比盲估计

Blind SNR Estimation of Time-Frequency Overlapped OFDM Signals for Underlay Cognitive Radio Networks

针对underlay认知无线电中时频重叠正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)信号的信噪比(Signal to Noise Ratio,SNR)难以估计的问题,提出了一种基于循环自相关特性的时频重叠OFDM信噪比盲估计方法.该方法提取接收信号的循环自相关函数截面中的离散谱线估计出分量信号的总功率,并结合时延和循环频率均为零时的离散谱线估计出时频重叠OFDM信号的信噪比.仿真结果表明,在无需先验信息的情况下,当SNR大于-7d B时,所提方法估计的均方误差小于0.1,并且对分量信号的功率比和频谱重叠率具有稳健性.

In this paper,we propose one cyclic autocorrelation function（ CAF） based method to implement the blind signal-to-noise（ SNR） estimation over the underlay cognitive radio networks. The orthogonal frequency division multiplexing（ OFDM） is adopted here. This SNR estimator utilizes the discrete spectrum of the CAF for the received signals,whose time delay and cyclic frequency are zero. Numerical results show that the proposed method is robust to the power ratio and spectral overlapping of the composite signal without prior information. Moreover,when the SNR is greater than-7 dB,the mean square error（ MSE） for the SNR estimation is less than 0. 1.