Adaptive Detection of Weak High-frequency CW Signal for Non-stationary Noise Environments
Li Zeng · Science Technology and Engineering · 2009
The method of time-frequency transformation and classical Kalman filter for detecting high-frequency CW signal have a noticeable decline in performance with heavy and non-stationary background noise.A new adaptive Kalman filter based on ARMA innovation model is designed to detect weak high-frequency CW signal in strong Non-stationary noise environments.The method avoids the shortcomings of classical Kalman filter which requires precise statistical characteristic of noise in system.The ARMA innovation model of CW signals is constructed firstly,according to its state space random signal model.Then with the on-line identification of MA model parameters,the Kalman filter gain is estimated to implement the adpative Kalman filter of CW signals.Simulation studies show this method more efficient and simple in the detection of weak CW signal.The new method can be applied to develop automatic receiver of high-frequency CW telegraph signal.