Adaptive de-noising in power line communication using multi-resolution analysis

De-noising is a crucial task in power line communication (PLC) for home automation. This paper presents a novel adaptive de-noising technique that is based on the multi-resolution analysis (MRA) of the wavelet transform for X10, which is a communication protocol designed for sending signals over low-voltage AC wiring. The proposed method establishes a look-up table to detect the noise level of 60 Hz signals. The proposed method is robust against variations of amplitude of a 60 Hz signal. According to the above identified noise level, the noise imposed upon 120 kHz bursts in X10 can be mitigated by using the proper decomposition and reconstruction scales of MRA.