Intelligent Anti-Jamming Decision Method Of Communication Based On Knowledge Graph

In the past 20 years,wireless communication has been widely used in human society.However,due to the openness of wireless channel,wireless communication systems are also increasingly subject to human jamming.Conventional anti-jamming communication technology represented by extended spectrum communication has improved the anti-fixed jamming performance of wireless communication system to a certain extent.In recent years,in order to deal with dynamic and intelligent jamming,communication anti-jamming technology based on heuristic and machine learning has been developed,which improves the anti-jamming ability against specific jamming patterns to a certain extent.However,these algorithms have high complexity,and it is difficult to deal with a variety of jamming patterns.Once the jamming parameters or patterns change,the system needs to learn again.In this context,this paper proposes an intelligent communication anti-jamming decision method based knowledge graph,which improves the adaptability of wireless communication system to complex jamming environment by representation,storage and efficient utilization of anti-jamming knowledge.Specifically,the paper includes the following three aspects:1.To solve the problem that historical experience and existing anti-jamming knowledge are lack of efficient storage method,and are difficult to use to improve the efficiency of anti-jamming decision,this paper proposes a construction method of anti-jamming knowledge graph.Firstly,the anti-jamming knowledge was transformed into RDF data form,and anti-jamming knowledge graph was generated based on Neo4 j graph database,which represent the anti-jamming knowledge structurally.Construction results show that the generated anti-interference knowledge graph can be used to represent anti-jamming knowledge efficiently.The anti-jamming knowledge graph can be expanded from different angles.The relationship between different entities in anti-jamming knowledge can be displayed intuitively,which contributes to using the relationship between anti-jamming knowledge entities for reasoning,and providing knowledge base for decision making based on anti-jamming knowledge graph.2.To solve the problem that unknown jamming patterns lack reasoning starting point in the anti-jamming knowledge graph,this paper proposes a time-frequency domain feature extraction algorithm for unknown jamming patterns.By calculating the characteristic parameters in the time-frequency domain of the unknown jamming pattern,we judge whether the pattern has the specific time-frequency domain characteristics,and establish a rule-based relationship between the characteristics of the unknown jamming patterns and the entities in the anti-jamming knowledge graph.Simulation results show that the feature extraction algorithm of unknown jamming patterns can accurately identify and extract the time-frequency domain features of the unknown patterns,which provides a basis for reasoning and decision based on anti-jamming knowledge graph in wireless communication system under unknown jamming conditions.3.Aiming at the problem of efficient reasoning using anti-jamming knowledge graph,this paper proposes an anti-jamming decision algorithm based on knowledge graph.To the known patterns,the algorithm searches the anti-jamming strategies directly from the anti-jamming knowledge graph,calculates the communication parameters according to the anti-jamming strategies,and gives the decision.To the unknown patterns,the algorithm searches the corresponding anti-jamming strategies after extracting the time-frequency domain features and calculates the communication parameters of the anti-jamming strategies.Finally,the anti-jamming decision is generated through strategy fusion.Based on the representation and utilization of existing anti-jamming knowledge,low complexity anti-jamming decision has been realized.On the basis of simulation verification,we developed a prototype machine to test the algorithm.The simulation result shows that the proposed algorithm has excellent adaptability and robustness to the errors existing in the jamming perception results.Prototype system test shows that the algorithm has excellent real-time performance and accuracy.