A Research On Detection And Identification Algorithm Of Radio Co-channel Interference

With the development of radio communication technology,various wireless communication modes have emerged one after another and the anti-interference technology has become more advanced.Interference technologies that disrupt the communication process are also emerging.For the communication process,there will be some phenomenon of maliciously interfering with the communication system.When the interference signal and the communication signal in the same frequency band are superimposed,the signal at the receiving end may demodulate the wrong symbol information,which affects the reception of the normal signal.Therefore,it is necessary to study all kinds of interference that may exist in the communication system and make corresponding countermeasures.When the interference occurs,the detection and identification of the interference can be realized as soon as possible to avoid or reduce the impact in time.In this paper,under the premise of the superposition of communication signals,Gaussian noise and interfering signals,the problem of the detection and classification of co-frequency interference is studied,and the identification of modulation interference under time-frequency aliasing is emphasized.The work is divided into the following three Sections:1.Analyze the common interferences in satellite communications: pulse interference,single frequency interference,frequency sweep interference and modulation interference,give mathematical models of various interference signals,and analyze the time and frequency domains of various interference signals and communication signals superimposed characteristic.2.Propose an interference detection and recognition method based on the combination of multi-domain features of Renyi entropy and constellation,superimpose the communication signal QPSK with the above common interference,add Gaussian white noise,and extract time-domain peaks from the time domain,frequency domain and transform domain Five characteristic parameters: mean ratio,frequency domain peak-toaverage ratio,box dimension,Renyi information entropy,and the number of cluster centers of constellation diagrams,are used to detect and identify the presence of interfering signals.The simulation results show that this method can achieve better results,and the actual data identification effect is slightly lower.3.Modulation recognition algorithm based on cyclic cumulant,modulate and recognize the modulation interference type of time-frequency aliasing.For the colored noise generated by the multi-signal mixed cyclic smooth transformation,combined with morphological filtering to remove noise;for the modulation type interference signal Recognize the problem,use the difference of the theoretical value of the cyclic cumulant at the cyclic frequency of different modulation signals,extract the characteristic parameters based on the second-order cyclic cumulant and the fourth-order cyclic cumulant,and combine the support vector machine classification to realize the identification of the modulated interference signal.Finally,the experimental results are given,and the simulation data can achieve good results,verifying the feasibility of the algorithm.