Research On Modulation Identification Of OFDM Signals Based On Higher Order Cumulant

The increasing of the modulation types and the requirements of the interoperability between different systems promote the continuous development of the modulation identification techniques. Orthogonal Frequency Division Multiplexing (OFDM), which has the advantages of the high bandwidth utilization, the anti-multipath capability and the high-speed data transmission, has been widely used in the communication field. Therefore, the identification of the OFDM signals is gaining more and more attentions.Firstly, the paper introduces the definition, the history and the development of the signal identification techniques and the principles, the advantages and the disadvantages of the OFDM technique. Then, the paper focuses on the identification of the OFDM signals without any prior-knowledge based on the higher order cumulants in the low SNR and the multipath Rayleigh fading channels. For the channels which are low-SNR and multipath Rayleigh fading, the paper proposes the technique of using the odd-order cumulants as the feature to achieve the classification between OFDM and the single-carrier signals. The algorithms based on the 3th-order and the 5th-order cumulants have been simulated and the results show that the identification of the OFDM signals based on the 3th-order cumulant has a better performance and a certain degree of anti-multipath capability. Combining with the support vector machines (SVM), the algorithm based on the 3th-order cumulant achieves a higher correct identification rate. Additionly, for the problem that the correct rate of the identification of different single-carrier signals in low SNR environments is low, a module of wavelet denoising has been added to process the signals before the acquirement of feature vectors. Computer simulations have been made and the results show that the method effectively improves the identification performance.