Non-Gaussian Noise Suppression and its Application to Power-Line Communications

%en a power-line is used as the transmission medium, the information-bearing waveform is usually subjected to high levels of non-Gaussian noise and interference. As a result, direct sequence spread spectrum (DSSS) is a common choice for the signalling format. The power-line channel, however, is power and bandwidth limited, so as not to interfere with electrical equipment. Consequently, this has the overall efJkct of limiting the attainable processing gain of power-line based DSSS communication systems. Interference mitigation strategies, therefore, have to be included within the system architecture to attain the low bit error rates required by emerging power-line applications. In this paper, we detail a mathematical framework, using joint timePequency analysis, for effective mitigation of non-Gaussian noise and inteverence. Following this) we propose two preprocessing techniques which complement the processing gain of a DSSS system by discriminating between the information-bearing signal and inteflerence, and then works to suppress the interfevence prior to despreading.