Signal detection with chaotic neural network

This paper proposes a system model of signal detection in the background of chaotic clutter. In this model, the classical matched filter and single value decomposition (SVD) arithmetic are used to reduce the disturbance of noise. The chaotic neural network based on improvement BP arithmetic and iterative chaotic map with infinite collapses (ICMIC) is used in the approach of the chaotic background clutter and the classical constant false alarm ratio (CFAR) detector is used to ensure CFAR. The results of the analysis and simulation both show this kind of signal detection system based on chaotic neural network has good detecting ability and fine noise immunity.