Research On Interference Identification And Decision-making Technology Of Satellite Communication System

With the rapid development of wireless communication technology today,the electromagnetic environment in which satellite communication systems are located is becoming more and more complex,and communication links are also faced with various interferences.With the in-depth study of electronic countermeasure technology by many scholars,there are more and more interference methods that affect the security of satellite communication,especially in the battlefield environment,it may encounter new types of interference from the enemy or at the same time.Composite interference composed of multiple interferences.In this case,it is difficult for traditional interference identification methods to achieve the desired identification effect.In this thesis,the scene of interference identification is set as an open-set scene with unknown interference and composite interference.Based on deep learning,the interference signal recognition models that can detect unknown interference signals and can identify composite interference signals are respectively studied.In addition,this thesis also designs a satellite communication system with intelligent anti-jamming function,and completes the development of intelligent antijamming software.The main research contents of this thesis are as follows:First,for the situation with a large number of known jamming signal samples,a deep learning-based jamming signal identification method is proposed,which takes the timefrequency graph of the jamming signal as the input data of the network,and trains the network with appropriate parameters.The simulation results show that the method can accurately identify several interference signals studied in this thesis.Secondly,this thesis improves the established interference signal identification model,so that it has the ability to detect unknown signals.This thesis has completed two improvement methods.The first improvement method is to add a confidence estimation branch to the original recognition network,and use this branch to judge whether the current network input interference signal is unknown interference.The second method is to add an unknown signal detection network outside the original identification network,and form an interference identification model together with the original identification network.Experiments show that this method has a high detection ability for unknown signals.Then,in view of the difficulty of compound signal identification,this thesis uses the idea of target detection to convert the identification of compound signals into the detection and identification of individual signals that make up the compound signal.This thesis uses the YOLOv5 algorithm with better detection effect in the current target detection field,uses a single interference signal as the training data to train the YOLOv5 network,and uses the trained network to identify the composite signal.The experimental results show that using this method,it is not only possible to identify which kinds of interference signals the composite interference consists of,but also to detect the key parameters of various interferences that constitute the composite interference,such as center frequency and bandwidth.Finally,this thesis designs and implements a satellite communication system with intelligent anti-jamming capability based on interference cognition,and mainly completes the design and development of the intelligent anti-jamming software.When the system is working normally,the intelligent anti-jamming module will always perform spectrum detection on the communication channel.When the interference signal is detected,it will automatically identify the interference signal and make an anti-jamming decision to ensure the reliable communication of the communication system.