作者: Zaihe Yu , Y.Q. Shi , Wei Su
DOI: 10.1109/MILCOM.2003.1290361
关键词: Discrete-time Fourier transform 、 Mathematics 、 Frequency domain 、 Short-time Fourier transform 、 Fractional Fourier transform 、 Spectral density estimation 、 Non-uniform discrete Fourier transform 、 Amplitude and phase-shift keying 、 Artificial intelligence 、 Multidimensional signal processing 、 Pattern recognition
摘要: The existing decision-theory based classifiers for M-ary frequency shift keying (MFSK) signals have assumed that there is some prior knowledge of the transmitted MFSK signal parameters; while feature-based limitations such as their thresholds are signal-to-noise-ratio-dependent (SNR-dependent). In this paper, we investigate useful properties amplitude spectrum signals. Using these classification criteria, a fast Fourier transform classifier (FFTC) has been developed. FFTC algorithm practical since it only requires reasonable received signal. It found works well in classifying 2-FSK, 4-FSK, 8-FSK, 16-FSK, and 32-FSK when SNR>0dB. also gives good estimation deviation