Research on Time Domain Filtering Based on Choi-Williams Distribution about Time-Phase Modulation

Abstract

Based on the non-stationary characteristics of the time-phase modulation signal, in this paper, Choi-Williams distribution time-domain filter method is put forward to solve the problem that the bad aggregation performance of time-domain analysis and poor detection performance of the system in the time-phase modulation signal analysis. First the best input parameter of the time-phase modulation signal is determined by the analyses of cyclo-stationarity and power spectral characteristics and the effect of the phase transition time, transition angle and other parameters on the power spectrum. Second the Choi-Williams transform method is used to get the relationship between the Choi-Williams time-frequency distribution and the phase mutation angle, the window function length, the carrier frequency and other parameters. Theoretical analysis and simulations show that: the time domain filtering based on Choi-Williams time-frequency distribution can convert the phase mutation characteristics of time-phase modulation signal into amplitude information by which we can detect TPM signal. The performance comparison is simulated between the detection method of this paper and the traditional filtering method, and the error rate of method in this paper is 1–2 dB lower than the traditional method.