Target detection based on matrix information geometry for continuous active sonar
ZHOU Zemin, HUA Xiaoqiang, GUAN Chengyu, ZENG Xinwu*
(College of Meteorology and Oceanology, National University of Defense Technology, Changsha 410073, Hunan, China)
Abstract:
Reverberation is an important interference. Due to the strong correlation between reverberation and transmitted signals, the traditional match processing usually causes severe false alarm. Therefore,it is crucial for signal detection in active sonar to suppress reverberation and decrease the false alarm.Aiming at the problem of constant false rate detection in low signal-to-reverberation ratio (SRR), in this paper, the idea of constructing signal matrix manifold corresponding to distant by using the echo of continuous active sonar is introduced based on the matrix information geometry theory. On this basis, the target detection based on matrix information geometry for continuous active sonar is proposed. The simulation results are based on reverberation statistical model and bistatic reverberation cell scattering model. It is shown that the matrix constant false alarm rate (CFAR) detector is better that of the traditional FFT-CFAR detector under low SRR.
KeyWords:
information geometry; continuous active sonar; constant false alarm rate detection; Kullback-Leibler divergence