Research On Co-channel Interference Recognition Based On WiFi Channel State Information

In the indoor environment,many wireless devices share the same unlicensed 2.4G ISM frequency band,which causes them to be easily interfered by signals from the same radio frequency during communication,and the communication performance drops sharply.Therefore,effective detection and identification of radio frequency interference sources is of great significance to the indoor planning and deployment of wireless devices.However,the current traditional RSSI-based wireless radio frequency interference detection and identification methods have low accuracy.To this end,this thesis proposes a co-channel interference detection and identification scheme based on WiFi channel state information(Channal State Information,CSI)technology.The main work of this thesis is as follows:First,construct a channel state information interference model,use the cyclostationary characteristics to perform interference discrimination on the model,and distinguish the interfered and non-interfered CSI signals by the cyclic spectrogram.The variance is used as the variation index to select the optimal sub-carrier,feature extraction in the time domain and frequency domain,and the RNN-LSTM algorithm is used to classify and identify the interference signal.Then,a co-channel interference recognition software was designed.The functional modules of the software include database design,interference detection module,classification and recognition module,and visual interface.Among them,the database is used to store data information such as interference signal data and recognition results;the interference detection module is used to obtain interference signal dataset;the classification recognition module is used to obtain classification results;the visualization interface is used to display the recognition results,including signal acquisition and graphic display,results display,control module,status bar module five parts;using C#and MATLAB mixed programming method to complete the realization of the overall function of the same frequency interference recognition software.Finally,this article builds a test platform for software function and performance test verification,which are tested in Zig Bee,Bluetooth,and WiFi interference environments.Experimental results show that the recognition rates under Zig Bee,Bluetooth,and WiFi interference environments are 95.35%,95.69%,and 96.43%,respectively,which are better than traditional RSSI-based wireless interference detection and recognition methods,and achieve the expected results.