Dynamic peak-based threshold estimation method for mitigating impulsive noise in power-line communication systems

Impulsive noise (IN) is one of the most dominant factors responsible for degrading the performance of power-line communication systems. One of the common techniques for mitigating IN is blanking which is applied at the front end of the receiver to zero the incoming signal when it exceeds a certain threshold. Determining the optimal blanking threshold (OBT) is, however, key for achieving the best performance. Most reported work to find the OBT is based on the availability of the long-term characteristics of IN at the receiver. In this paper, we consider orthogonal frequency-division multiplexing (OFDM)-based power-line communications and propose a method for finding the OBT without requiring any knowledge about the IN. We show that there is a direct relationship between the OBT and the peak-to-average power value of the OFDM symbol and utilize it to identify the OBT. The results reveal that the proposed technique not only eliminates the need to prior knowledge about the characteristics of IN but also achieves a gain between 0.5-2.5 dB depending on the accuracy of the signal peak-to-average estimate. It will also be shown how the performance of the proposed method can be further enhanced by employing some basic signal per processing at the transmitter.