Identification Method of Wireless Communication Interference Signals Based on QPSO-SVM

To effectively identify wireless communication interference signals, a quantum particle swarm optimization (QPSO) optimized support vector machine (SVM) wireless communication interference signal recognition method is proposed. Firstly, based on the analysis of wireless interference signals, the edge denoising of wireless suppression interference signals is carried out using a combination of wavelet packet soft and hard thresholding methods. Then, multiple feature parameters such as time-domain kurtosis, frequency-domain kurtosis, and variation domain quartic kurtosis are extracted for normalization, and a multi-dimensional feature vector is formed as the input of SVM to solve the problem of low reliability and low accuracy in identifying interference signals with a single feature. Finally, QPSO optimizes the classification model parameters and uses SVM algorithm to build a multi class binary tree model for identifying wireless interference signals. The results are compared with SVM and PSO optimized SVM methods for wireless interference signal recognition. The experimental results show that QPSO optimized SVM method can improve the recognition rate of wireless interference signals to 92.14%.