Font Size: a A A

The Research On Recognition Of Communication Modulation Signal Based On Artificial Bee Colony Algorithm

Posted on:2017-04-26Degree:MasterType:Thesis
Country:ChinaCandidate:S H WuFull Text:PDF
GTID:2308330491951605Subject:Circuits and Systems
Abstract/Summary:PDF Full Text Request
Communication signals modulation recognition technology is widely used in military and civil fields. In this paper, according to the research and analysis of the digital signals modulation, a new modulation recognition method based on artificial bee colony algorithm is proposed, and an improved algorithm is proposed to optimize the traditional artificial bee colony algorithm. The major work in this paper is summarized as below:(1) According to the characteristic that two and higher order cumulants of Gaussian noise are zero, three feature vectors based on higher order cumulants are constructed to avoid the effects of noise. To solve the problem of nonlinearity and non-stationary in non-cooperative communication system, Hilbert-Huang transform is proposed and optimized to process this type of signal. Finally,united feature module based on higher order cumulants and Hilbert-Huang transform is constructed.(2) A modulation recognition method based on support vector machine whose penalty factor and kernel function parameters are optimized by artificial bee colony algorithm is proposed. The recognition rate of support vector machine is as the fitness value of artificial bee colony algorithm,and the simulation results show that the recognition rate is 89.2% when SNR is-3 dB, as well as over 99% when SNR is more than 4 dB.(3) To solve the problem of slow convergence speed and nonuniform distribution of the initial food source of traditional artificial bee colony algorithm, an improved bee colony algorithm based on two-dimensional uniform design and Euclidean-distance is proposed. Two-dimensional uniform design which is used to make the food source uniformly distribute in the solution space can improve the global search ability of the algorithm, and Euclidean-distance which is used in constructing new food source can improve the efficiency of optimization. The simulation results show that the convergence speed of improved bee colony algorithm is improved and this method has better recognition rate. And compared with traditional artificial bee colony algorithm, the recognition rate of improved bee colony algorithm is increased by 3.7%.
Keywords/Search Tags:modulation recognition, higher-order cumulants, Hilbert-Huang transform, artificial bee colony, support vector machine
PDF Full Text Request
Related items