Research On Interference Mitigation Of Power Line Communication

Power line communication(PLC)is a communication technology that utilizes existing power line networks for data transmission.PLC has always been highly valued for its advantages such as no new wires,network with electricity and extensive coverage.However,the power-line is not designed for communication.When it is used to transmit high frequency signals,the noise will cause burst error code,which seriously affect the communication performance.In order to realize reliable communication on powerline,the noise mitigation technology is studied in this thesis.The main work in this thesis is as follows:Firstly,referring to the European Telecommunications Standards Institute measurement campaign,noise in four scenarios is measured for the purpose of studying noise characteristics in power-lines.Based on the measured noise,time-frequency domain characteristics and time-variability are analyzed and statistical properties are studied.Measured noise is compared with the noise model.Secondly,the filter for suppressing pulse oscillation and trailing is simulated and fabricated,and its trailing suppression performance is verified.Traditionally,automatic gain control mainly deals with signal fluctuation caused by the near-far effect,without considering the situation that impulse noise is much larger than useful signal.However,the power spectral density of impulse noise in PLC is 10～40 d B higher than that of useful signal.In view of this situation,combined with a new fast power estimation method,an improved automatic gain control that detects and deeply saturates impulse noise is proposed.The experimental results demonstrate that improved method improves output SNR and reduces adverse effects of impulse noise in the analog domain.Thirdly,adaptive threshold calculation method for nonlinear preprocessing is investigated.In view of the shortcomings of adaptive threshold calculation methods based on estimating useful signal average power,estimating noise characteristic parameter and received signal peak-to-median ratio,an adaptive threshold calculation method based on threshold iteration and received signal PAPR is proposed.The proposed improved method detects whether there is impulse noise according to power ratio of adjacent OFDM symbols.If impulse noise is detected,the impulse samples are continuously removed in an iterative way to reduce the average power of the received signal,and then the nonlinear preprocessing threshold is determined adaptively in combination with the PAPR of received signal.If no impulse noise is detected,the OFDM demodulation is directly performed to reduce processing delay and loss of useful signal.The simulation results demonstrate that,compared with existing methods,improved adaptive threshold determination method has similar Bit error rate(BER)performance under weak pulse scenarios,while BER performance is improved by about 2 d B under strong pulse scenarios.