Research On Impulsive Noise Mitigation And Channel Estimation Of Power Line Communication System

With the rapid development of society,the pace of intelligent construction is accelerating,especially in the field of smart grid and smart home.Communication technology plays an extremely important role in intelligent construction.Power Line Communication(PLC)has the advantages of wide distribution and low deployment cost,and has become one of the core intelligent communication technologies.However,existing power lines are mainly used for power transmission.Compared with dedicated communication systems,power line channels exhibit more complex channel characteristics and noise characteristics.In order to improve communication quality,it is important to conduct research on impulsive noise suppression technology and channel estimation technology for power lines.The main research contents and results are as follows:Firstly,the characteristics of power line channel and noise are studied.According to its characteristics of typical multipath effect and frequency selective fading,the multipath propagation model is used to model it.The sources and characteristics of power line noise are studied,and the modeling representation is given.Through the analysis and comparison,it can be seen that high power burst impulsive noise interferes most with PLC.Secondly,the nonlinear impulsive noise suppression method is studied.Aiming at the characteristics of the peak-to-average power ratio of the OFDM power line system signal,an impulsive noise suppression method based on companding peak value and Blanking-Clipping is proposed.Several typical methods of nonlinear impulsive noise suppression are compared and analyzed,and a method is proposed to improve the Blanking-Clipping performance by reducing the signal peak-to-average ratio based on the peak of companding.This method is based on the estimation of the threshold of the companding peak,reduces the peak-to-average power ratio of the signal,enhances the distinction between impulsive noise and signal,and performs more accurate impulsive noise detection and suppression.The simulation results show that the method in this thesis shows better performance in the Signal to Implusive Noise Ratio(SINR)conditions,both the signal output signal to noise ratio and the bit error rate.Finally,impulsive noise suppression and channel estimation in OFDM system are studied jointly.Firstly,the principle,advantages and disadvantages of traditional channel estimation methods are analyzed.Finally,the channel estimation method based on DFT is selected among the iterative channel estimation and impulsive noise suppression methods.Aiming at the shortcomings of threshold-based nonlinear impulsive noise suppression methods that usually need to achieve a compromise between the success rate of IN detection and the false alarm rate.This thesis proposes a method of IN detection assisted by deep neural network(DNN)and reconstruction and elimination of residual impulsive noise at the receiver after applying adaptive threshold to preprocess in at the transmitter.Simulation results show that the proposed method has faster convergence speed,better performance of channel estimation mean square error and system error rate.