Study Of Power Line Carrier Communication Signals Detection And Identification

In recent years,with the rapid development of China's power industry,power network has also been widely used.In the power line channels,due to the complex noise and multipath fading,it is a great challenge to transmit data.In the power line communication(PLC)signal detection and recognition research,power line communication channel characteristics of the study is essential.Based on the analysis of multipath fading characteristics of PLC channel,this paper improves the multipath fading model,which makes it more able to reflect the fading characteristics.The paper studies the feature extraction method and classifier model of classification recognition.And it uses the wavelet transform coefficient amplitude,high order cumulant and kurtosis as the characteristic parameters,and support vector machine as classifier to realize the classification of PLC signal.In this paper,the improved particle swarm algorithm is combined with the support vector machine classifier to realize the primary classification of the PLC signal.In other words,the signals of MASK,MFSK,MPSK and 16 QAM are divided into {2ASK,2PSK},4ASK,{4PSK,8PSK,MFSK} and 16 QAM.The inter-class signals classification is achieved by the combination of high order cumulant,wavelet transform,amplitude variance,and the improved kurtosis algorithm.Finally,the wavelet transform coefficient amplitude histogram is used to classify and identify the intra-class signals.Through the study of classification characteristic parameters and classifier,this paper improves the inter-class and intra-class recognition algorithms of PLC signals,and designs a signal classification recognizer for power line channel.The simulation results indicate that the correct recognition rate of the signal is more than 88% when signal to noise ratio(SNR)is 5d B,which shows that the improved recognition algorithm has better recognition performance.