Research Construction And Application Of Radio Regulatory Knowledge Graph

The wide application of radio technology and business in various industries and fields of the national economy is inseparable from the escort of radio regulatory work.Because of the openness of wireless communication,interference devices can easily reduce the signal-to-interference-noise ratio through radio interference,disrupting wireless communication,and the sharp increase in the number of radio transmitting devices makes the electromagnetic environment in the air more complex.As a result,effective regulatory data management,as well as knowledge mining and exploration,are essential study fields in current radio regulation.Knowledge Graph(KG)has been widely applied in numerous engineering sectors in recent years and has achieved good use benefits as a new technical means to actualize a huge amount of semantic integration and interactive operation.This work investigates the creation and use of a radio regulatory knowledge graph based on this.The following are the key contents and contributions:(1)Radio regulatory knowledge graph construction.The construction of radio regulatory knowledge graph is studied from the three modules of radio transmitting equipment type approval,radio jamming monitoring and Wi Fi signal monitoring.The radio regulatory data is obtained by means of web crawler,data simulation and equipment collection,respectively,and the radio regulatory knowledge graph is successfully constructed by using the property graph model and triplet model defined based on EES model({Entity,Event,Scene}).(2)Radio regulatory data mining and knowledge reasoning.Use the centrality algorithm,similarity algorithm and community detection algorithm commonly used in ALGO to conduct data mining for radio transmission equipment model approval;use the correlation between the CSI(Channel State Information)of the WiFi signal and human behavior to build a SNN(Siamese Neural Networks)model for learning and training the CSI data features in two different scenarios,manned and unmanned,realize the prediction of indoor personnel intrusion in the case of small samples;based on the type approval of radio transmission equipment and radio jamming monitoring,the knowledge reasoning method based on logical rules is used to obtain new entities and relationships in the graph.(3)Design and develop a radio regulatory experiment system.The radio regulatory experimental system is designed and developed on the Django framework,and the B/S architecture is adopted to provide users with five functional services: knowledge graph construction,data mining and knowledge reasoning,knowledge base question answering(KBQA),graph visualization and cloud management.The work presented in this thesis enables efficient organization and management of radio regulatory data,information,and knowledge,as well as the creation of value-added data,decision-making support for radio regulatory work,and new technical means and solutions for management departments,all of which contribute to the goal of improving radio regulatory.