Modulation classification in Alpha stable noise

In Alpha stable noise environment, the traditional methods of digital modulation signals classification have the problems of poor performance. A novel classification method based on low order wavelet packet decomposition is proposed to solve this problem. This method extracts the recognition characteristic parameters which are normalized energy in respective bands of the signals db2 after low order wavelet packet decomposition. And then the method uses radial basis function (RBF) neural network as a classifier to achieve digital modulation signals classification. Simulation results show that the proposed method has good performance and robustness in alpha stable noise.