Bottom-Up Statistical PLC Channel Modeling—Part II: Inferring the Statistics

Abstract

We investigate the statistical behavior of power-line communication (PLC) channels. It is inferred from the results obtained with a statistical bottom-up channel simulator (described in Part I of this paper) that uses an in-home topology model derived from the observation of wiring practices and norms. It computes channel transfer functions via the application of transmission-line theory. The comprehensive study includes the analysis of the statistics of the path-loss profile, the average channel gain, the root-mean-square delay spread, and the channel capacity. We highlight the dependency on topological information as the network layout area, the intensity of outlets, the backbone length, etc. We furthermore propose a channel classification based either on average capacity or topological information (e.g., the belonging of outlets to rooms served by the same derivation box). We show that the developed channel simulator constitutes a powerful theoretical framework for the generation and analysis of statistically representative channels with a strong connection to physical reality and close match to the results obtained in the measurements campaigns.