Modulation Recognition For Signals In The Alpha-stable Distribution Noise

Modulation recognition for communication signal has a significant application in the military and civilian fields.How to identify the modulation mode of the communication signal has been a problem to be solved urgently.With the advancement of the information process,the electromagnetic space is complex and varied,and the signal modulation way is varied.The channel environment is filled with all kinds of interference noise,and the received signal is uncertain due to the interference of various kinds of noise.In general,it is assumed that the noise in the channel environment is consistent with the Gaussian distribution.However,in the actual wireless channel,such noise often has significant strong pulse characteristics,and the traditional Gaussian distribution assumption can not be described correctly.After the majority of experts argued that the stability of the Alpha distribution of such strong impulse noise described the most appropriate.Therefore,it is very necessary to study how to identify the modulation method of communication signal in the environment of Alpha stable distribution noise,which is of great practical significance.The purpose of this paper is to study the modulation and recognition of communication signals in Alpha stable distributed noise environment.The main work is as follows:(1)The four definitions of Alpha stable distribution are studied,and carefully analyzed the characteristics of each parameter.The infinite series approximation method is used to calculate the probability density function,and three common cases of the distribution are given.Compared with the simulation,it is found that many modulation recognition methods under the traditional Gaussian hypothesis are obviously weakened in the noise environment because there is no second order statistic for the stability of the alpha stable distribution.(2)The modulation recognition algorithm based on fractional lower order statistics is studied.First introduced the concept of several common fractional low order moments.Then,a method for estimating the stability of Alpha stable distribution based on zero-order statistics is proposed.The simulation results show that the method has a good estimation effect.Finally,the concept of fractional low order cyclic spectrum is introduced,and five correlation coefficients are defined.A modulation recognition algorithm based on fractional low-order cyclic correlation coefficient is proposed.By using the difference of five different decision thresholds,the classification and recognition of common digital signals are realized by decision tree classifier.Simulation can be obtained,the method can better distinguish the signal modulation.(3)The modulation recognition algorithm based on multifractal spectrum is studied.First introduced the related concepts of fractal theory.Secondly,the descriptionof the multifractal spectrum is realized based on the Hausdorff dimension and the generalized dimension.Then,a method of extracting the multifractal spectrum feature of the signal based on the correlation integral method is proposed,and the relevant characteristic parameters of the signal are obtained by using the correlation integral and the generalized dimension.The modulation recognition algorithm based on multifractal spectrum is proposed.The characteristics of multiple fractal spectrum of signal and noise are simulated and analyzed,and the modulation recognition of signal is realized by the difference of five characteristic parameters of signal.Simulation results show that this algorithm can distinguish the modulation of the signal and is better than the single dimension.(4)The digital signal modulation and identification system is designed and implemented.Aiming at the three important steps of digital signal generator,signal existence detection and signal recognition trainer,the corresponding GUI interface is designed respectively.The modulation and recognition system of digital signal is designed,which can be used to realize the modulation and recognition of common digital signals conveniently and effectively.The modulation and recognition system of digital signal is designed.