This note corrects a recurrent erratum in Section VII of our paper “An MTL-based channel model for indoor broadband mimo power line communications”[1]. Errors are concerned with the quantification of the spatial correlation of MIMO channels by means of the condition number of the channel matrix. According to the definition given in expression (25) of the paper, the higher the condition number, the higher the spatial correlation. However, Section VII is written as if high condition numbers correspond to low spatially correlated channels. Hence,whenever it is said that a given fact reduces the spatial correlation or, equivalently, that it yields lower spatially correlated channels, it should say exactly the opposite, i.e, the referred fact increases the spatial correlation or yields higher spatially correlated channels; and vice versa.

The recent release of indoor power line communication (PLC) standards with multiple-input multiple-output (MIMO) capabilities has led to the need for channel models that reflect the multiconductor nature of the grid. This paper proposes a model, based on the multiconductor transmission line theory, along with a random PLC channel generator operating in the frequency range from 1 to 100 MHz. Its distinctive features are the modeling of three key elements: the asymmetry in the layout caused by the derivations to the light switches, the impedance of the devices connected to the three wires of the grid, and the variable distance between the conductors in loose wiring deployments. The influence of these elements in the spatial correlation between the streams of the MIMO links is studied and compared with measurements.