Design of bistable system based nonlinear detector

In this paper, a novel nonlinear detector based on bistable system is developed to detect signals in a large group of non-Gaussian noise with heavy tail. Assuming that the functional form of the probability density function (PDF) of the noise is unknown, we use the bistable system (BS) as a nonlinear limiter. The BS parameters are chosen by matching two nonlinearities. One is the BS nonlinearity, which is derived by the proposed PNE algorithm. The other is the objective nonlinear function, which is derived by the proposed GI algorithm. Experiment results show that proposed detector provides a superior performance compared to a linear detector and very close to that of the locally optimal (LO) detector.