Temporal Exemplar Channels In High-Multipath Environments

Industrial wireless plays a crucial role in cyber-physical system (CPS) advances for the future vision of smart manufacturing. However, industrial wireless environments are different from each other and are different from home and office environments. Hence, industrial wireless channel modeling is essential for the development of industrial wireless systems. Moreover, millimeter-wave (mmWave) wireless bands have a high potential to be used for the high data-rates required for industrial automation reliability, with multiple antennas envisioned to mitigate the high path loss. As a result, in this work, we introduce a machine learning (ML) based exemplar extraction approach on mmWave wireless spatial-channel measurements. The proposed approach processes the measured power-angle-delay-profiles to cluster them into a number of groups with respect to the angle of arrival. Then, an exemplar power-delay-profile (PDP) is extracted to represent each group. The resulting set of exemplars provide a tractable way to conduct mmWave industrial wireless systems testing and evaluation by compactly representing various feature groups. This allows the assessment of wireless equipment over the exemplars without the need to test over all of the different instances of wireless channel paths.