Improved fisher linear discriminant analysis for feature extraction of unintentional modulation on pulse by combining ambiguity function with wavelet transform

The unintentional modulation on pulse (UMOP) features produced by radar emitter are similar to the fingerprint of a person, which plays a significant role in radar emitter recognition.In this paper, we propose an improved Fisher discriminant analysis method for feature extraction of UMOP by combining ambiguity function and wavelet transform. In this method, the radar emitter signals are firstly mapped onto the time-delay and frequency-offset plane via ambiguity function. We extract the diagonal slice of signal in the ambiguity function plane, and obtain wavelet coefficients of this signal through wavelet transform. Then, an improved Fisher discriminant rule is used to achieve feature selection and feature space reduction. Experiments on simulated and measured data verify the effectiveness of the proposed method.