Skip to Main Content
In this paper a statistical characterization of the spatial properties of MIMO PLC channels is provided, where the term spatial refers to the multiple input and output ports. Based on a set of channel measurements in the range from 0 to 100 MHz, the channel covariance matrices and their eigenvalues and eigenvectors are analyzed. Spatial eigenvalues can be approximated by uniform random variables while the eigenvectors can be obtained by QR decomposition of a square matrix with i.i.d. Gaussian samples. Finally the consistency between the measured and the generated spatial correlation model for power line channels is proven. Furthermore, the statistical descriptions extracted from the field measurements are used to update a previously proposed MIMO power line channel model.