Singularity Intensity Function Analysis of Autoregressive Spectrum and Its Application in Weak Target Detection Under Sea Clutter Background

Weak target detection based on fractal analysis under rada sea clutter background is an open problem. The existing methods, using single fractal dimension and Hurst exponent in time domain or Fourier domain, are not applicable under low signal-to-clutter ratio (SCR) conditions. Since autoregressive (AR) spectrum has the advantage of high-frequency resolution over the Fourier spectrum in sea clutter analysis, while multifracal theory is an extension of single fractal analysis. Therefore, we combined the multifractal analysis with AR spectrum estimate theory. Since singularity intensity function is an important parameter to describe a multifractal set, this paper proposed a weak target detection method based on singularity intensity function of AR spectrum under the sea clutter background. Then real S-band data sets are used to analyze the singularity intensity function of AR spectrum, and the results show that the AR singularity intensity function between sea clutter and targets has different value range interval. Finally, the singularity intensity function width of AR spectrum is taken as a statistical test for weak target detection. Compared to the existing fractal methods method, the proposed target detection method improves the detection probability over 20% in low SCR condition.