Kalman noncoherent detection of CPFSK signals

The paper addresses the problem of detecting Continuous Phase Frequency Shift Keying, (CPFSK) signals embedded in Gaussian noise. The paper introduces a description of CPFSK signals using a state-space model. Based on this model, a noncoherent detection method for CPFSK signals that uses an Extended Kalman Filter was developed. The performances of the detection method were tested on Minimum Shift Keying, (MSK) and Gaussian Minimum Shift Keying, (GMSK) modulated signals embedded in Gaussian noise. The results leaded to the conclusion that for moderately low SNR's, the Extended Kalman Filtering (EKF) algorithm provides satisfactory, even good results in noncoherent demodulation of CPFSK signals.