A CFAR detector based on orthogonal wavelet transform

Radar target constant false alarm rate (CFAR) detection in wavelet domain is elementarily investigated in this paper. Theoretical analysis shows that independent and identically distributed Gaussian noise keeps the independence and variance invariable through the orthogonal wavelet transform. In the proposed detector, orthogonal wavelet transform is applied to radar echo signal and the wavelet coefficients are obtained. And then a squarer detector is utilized. After that, according to the characteristic of Gaussian noise in orthogonal wavelet domain, a cell average-CFAR detector model based on orthogonal wavelet transform is established. The probability equations of false alarm and detection are also deduced, and the influence of the wavelet coefficient length on detection performance is analyzed. Finally, the experiment results demonstrate that the proposed method can detect target under different false alarm rates effectively and adaptively.