基于回归特征分析的BPSK信号盲处理结果可信性评估

Credibility Evaluation for Blind Processing Results of BPSK Signals by Using Regression Features

针对二相编码(Binary phase shift keying,BPSK)信号盲处理过程中信号调制方式识别及参数估计的可信性评估问题,提出了一种基于线性回归特征分析的处理方法。根据特定处理结果建立参考信号,并将参考信号与原始观测信号进行相关累加取模值,在分析相关累加模值曲线在原序与升序两种不同情形下线性回归差异的基础上,分别定义了排序回归系数及分段回归符号两大特征,用以判别BPSK信号盲处理结果的可信性。计算机仿真结果表明,在低信噪比、缺乏信号先验信息的条件下,本算法可对BPSK信号盲处理结果的可信性进行有效检验。

A credibility test method based on the features of the linear regression is proposed to evaluate the blind processing results of binary phase shift keying（BPSK）signals such as the modulation recognition and parameter estimation.The reference signals are first constructed depending on the certain identified modulation results.Two features,the ordered regression factor and sign of the segments regression,are defined to test the credibility of the processing results by analyzing the differences of linear regression characteristics of the correlation series modulus between the original and the ascending order.Simulation results show that the proposed method can be used to effectively verify the credibility for blind processing results of BPSK signals both at low signal-to-noise ratio and the condition without aprior knowledge for signal parameters.