Research On Interference Recognition Technology In Frequency Hopping Communication System

In a complex electromagnetic environment,frequency hopping communication systems have good anti-interception,anti-jamming,anti-fading,and multi-access networking capabilities,making this technology widely used in daily civilian communications and defense military communications.However,due to the rapid development of electronic countermeasures,the ways of interference become diverse,and higher requirements have been placed on the reliability and stability of frequency hopping communications.The drawbacks of conventional frequency hopping communication systems are also developing significantly.In recent years,people have continuously optimized and improved the traditional frequency hopping communications system.One is to increase the frequency hopping rate,but high-speed frequency hopping system also has its own limitations.Because the high-frequency hopping signal has a short residence time,low spectrum utilization,and a wide spectrum occupancy,the side lobes cause interference to its adjacent channels.The second is adaptive frequency hopping systems.By adding a real-time channel evaluation module at the receiver,the communication frequency band is evaluated,and the result is fed back to the transmitter,and the frequency hopping pattern is continuously updated to replace the interfered frequency band.However,the real-time performance of this method is poor,and the performance of the system is greatly reduced in the time-varying channel.The third is interference identification technology,which identifies the interference type and its main parameters through effective identification of the interference,and then adopts corresponding anti-interference strategies to achieve the purpose of improving the system's anti-jamming performance.This paper thoroughly studies the interference identification technology and provides support for the anti-jamming strategy of the communication receiver,enhancing the antijamming capability of the frequency hopping communication system.The main research contents and contributions of this article include the following aspects: 1.Based on the characteristics of current interference suppression technologies,eight common types of interference in frequency hopping communication systems are analyzed: No Interference,Broadband Noise Jamming(BNJ),Narrowband Noise Jamming(NNJ),Partial Band Noise Jamming(PBN),Single-Tone Jamming(STJ),Multi-Tone Jamming(MTJ),Swept Frequency Jamming(FSJ),and Pulse Jamming(PJ),establish various mathematical models of interference,and simulate various types of interference signals on Jupyter Notebook and Python.In time domain,frequency domain and time-frequency domain,interference signals are analyzed.2.By reading a large number of documents and simulation experiments,in-depth study of interference signal feature parameters extraction method.Seven characteristic parameters with good classification effect are extracted: energy limit factor,normalized spectrum bandwidth,carrier factor coefficient,normalized spectrum kurtosis,normalized spectrum flatness,time-domain peak-to-average ratio and fractional order Fourier domain energy concentration.All unknown interference signals are represented by 7 kinds of characteristic parameters,to realize the unification of the interference identification algorithm.3.Based on the methodologies of seven characteristic parameters and pattern recognition,two classification strategies are designed.The first is an interference recognition strategy based on BP neural network.This strategy uses the gradient descent algorithm to train and adjust the parameters of the BP neural network to realize the recognition of unknown interference signals.The second is the interference identification strategy based on the decision tree model.Based on the CART tree algorithm,a decision tree is constructed to accurately classify unknown interference signals.The results of simulation experiments show that when the ratio of interference to interference is greater than 15 dB,the accuracy of the two classifiers on the interference signal can reach more than 90%.