Research And Implementation Of Blind Demodulation Conventional Communication Signals

This thesis studies the problems of modulation recognition and demodulation of three typical digital communication signals, including Phase-Shift-Keying (PSK), Frequency-Shift-Keying (FSK) and Quadrature Amplitude Modulation (QAM). The simulation system realizes the QAM signal modulation recognition as well as demodulation, and the performance of the designed system is reflected by the probability of bit error. Instead of acquiring prior information such as the carrier frequency and symbol rate, down-conversion and timing synchronization are realized based on estimated parameters. High-order cumulants after calculus of differences and variance of the minimum ring are adopted as key features in modulation recognition, which insures that the modulation recognition is not sensitive to residual frequency offset. Implementation based on Matlab shows that the aforementioned three types of signals have been recognized with an average failure rate less than10%at SNR (signal-to-noise ratio) greater than-15dB. System performance has satisfied the requirements of signal recognition and demodulation in practical non-cooperation digital communication systems...