An OFDM modulation recognition algorithm based on spectrum analysis

This paper presents the feature of energy distribution parameters for power spectrum. Then based on the spectrum analysis, an OFDM recognition method is proposed using the energy distribution parameter P, the kurtosis coefficient of the power spectrum and the power spectrum parameter R. Power spectrum is preprocessed to eliminate the effect of signal bandwidth, sampling rate, carrier frequency and FFT points on the threshold of identification. This method does not require any prior knowledge (such as symbol rate, carrier frequency, etc.), and the IF signal can be directly processed. Simulation results show that the extracted parameters have good performances in both AWGN and Rayleigh fading channel. The recognition accuracy is up to 97% with a signal to noise ratio (SNR) at 10dB.