Modulation Recognition Of Dense Multiple Modulation Mixed Signal

Accurate and efficient interaction between information is the ultimate goal of communication,and modulation mode is the most important characteristic parameter of communication signal.Only when the modulation type of the signal is known accurately,the subsequent demodulation and the analysis of the contained information can be realized.Because modern wireless communication technology is in a dynamic and high-speed development state,the actual communication environment becomes more and more complex because of the appearance of unknown signals.The communication model also gradually evolves from single signal to multi signal structure.The existence of interference signal will bring some difficulties to signal preprocessing,which greatly improves the difficulty of signal modulation recognition in non-cooperative communication degree.At present,most of the recognition algorithms are still at the level of theoretical research,and the high complexity makes many algorithms difficult to be applied in practical engineering.Therefore,how to achieve accurate reconnaissance and effective classification of multi modulation mixed signals in dense environment,and how to innovate and improve on the basis of existing automatic identification technology to adapt to the changing recognition requirements,has very important practical significance.In the dense high-order amplitude and phase signal communication environment,the preprocessing process is easily affected by the spectrum interference of adjacent signals,and the degree of aggregation of the demodulated signal is poor.At this time,the recognition performance of the traditional feature extraction algorithm based on ideal demodulation decreases rapidly.For dense high-order amplitude phase mixed signals,this thesis proposes a modulation recognition algorithm based on the track characteristics of demodulation constellation.The proposed algorithm demodulates the signal based on the high-order amplitude and phase blind receiver,constructs the constellation trajectory features according to the amplitude curvature distribution and trajectory information of the constellation,and establishes the recognition template of the corresponding feature domain for classification,so that the unique constellation structure information of the signal can be fully utilized.Theoretical analysis and simulation results show that the performance of the proposed method is better than the traditional recognition algorithm based on constellation amplitude information,especially in the non-ideal demodulation environment of dense signals.Furthermore,the recognition of classical analog digital mixed signal in ECM is studied.It is found that the anti-noise performance of the instantaneous characteristics of the signal is poor,while the high-order moment has good anti-noise performance,but it has high requirements for timing synchronization.At the same time,the cyclic spectrum characteristics of the analog signal are relatively weak and there is no stable constellation feature,which leads to the low differentiation of the modulation recognition algorithm based on the above features in the mixed signal.In view of this phenomenon,this thesis proposes a modulation recognition algorithm based on multi-dimensional characteristics of all digital receiver.The blind receiver demodulates the signal,jointly demodulates the number of quadrants of constellation distribution,the distribution distance of frequency offset tracking data and the matching error of baseband data template,which makes full use of all kinds of demodulation output information.Sufficient theoretical basis and a variety of simulation results show that the algorithm has better performance than the widely used modulation recognition algorithm based on cyclic spectrum and high-order moment,and its implementation difficulty is lower.Finally,a set of electromagnetic situation awareness simulation software system in dense signal environment is developed.In the wide-band and complex electromagnetic environment where high-order amplitude and phase signals and classical analog-to-digital signals coexist,a modulation recognition scheme in dense signal environment is proposed.The triple receiver is used to extract the required characteristic parameters of 10 kinds of signals.The overall calculation is simple and the complexity of software system is reduced.On the platform of MATLAB simulation software,the model of the scheme is built and its recognition function is verified.The signal used in the software system test is provided by China Electronics Technology Group.The test results show that the recognition scheme designed in this thesis can complete the correct identification of signals in dense and mixed environment,and its recognition rate can meet the application requirements in practical engineering.