Detection of DS-SS Signal Using Parallel Filter Banks and High-Order Cumulants

In the intercept of the non-cooperation communication, a new effective method based on the parallel filter banks and high-order cumulants is proposed in this paper. The filter banks split the input signal into many narrow frequency bands, which not only can completely descript unknown signals in the time-frequency domain, but improves the signal-to-noise ratio (SNR). Moreover, Gaussian noises can be suppressed by using the three-order cumulants, which can preserve more useful information and boost the detection performance. Because of merits of the parallel filter banks and high-order cumulants, the combined method is proposed to detect DS-SS signal in this paper. Then we obtain parameters by creating the feature matrix. The simulation indicates that it is effective for this technique to detect DS-SS signal and estimate the carrier frequency and the bandwidth under low SNR.