System Distortion Model for the Cross-Validation of Millimeter-Wave Channel Sounders

Because millimeter-wave directional channel measurements are time-consuming and expensive to collect, there is considerable interest in combining measurement data obtained with different channel sounders in order to yield more comprehensive datasets. The simplest way to verify that the results obtained with these different instruments in a given environment are comparable would be to transport the channel sounders to that environment, collect and process measurement data, and then compare the results. Because this is rarely feasible, we propose an alternative method that is much more practical. It involves: 1) Generating an ideal three-dimensional channel impulse response that corresponds to a scenario of interest, 2) Degrading the ideal response by applying a distortion model that capture the factors that limit the spatio-temporal resolution and dynamic range of each channel sounder, and 3) Applying the multipath component (MPC) extraction techniques used by the channel sounder to the distorted response. After the last step, one will observe: a) correctly estimated, b) incorrectly estimated, c) missing, and d) spurious MPCs. Discrepancies between the ideal and distorted responses will be readily apparent and the performance of the channel sounders can be easily compared in a given environment. The effort required to fully characterize the three-dimensional patterns of the transmitting and receiving antennas is considerable and further work is required to determine the corresponding accuracy requirements.