Recognition of communication signal modulation based on SAA-SVM

Support vector machine has a wide range of applications in the communications signal modulation recognition, its parameters directly affect the recognition results, but lack of proper selection methods. In this paper, the simulated annealing algorithm has been utilized for optimization of the parameters C and g of support vector machine classifier. Compared with genetic algorithm, which is a traditional method of performing parameter searching, the rate of recognition of the proposed method increased by 3.58% and optimization time reduced by 27.7%. The results suggest that recognition of communication signal modulation based on SAA-SVM is accurate and feasible.