Multi-Modal Sensing-Aided Channel Prediction for 6G mmWave Massive Antenna Systems

As a key technology to meet the explosively growing data rate demands for 6G wireless networks, millimeter wave (mmWave) massive antenna systems have received significant attention recently. To fully leverage the advantages of massive antenna systems, acquiring accurate channel information at the base station (BS) is crucial but such is challenging due to the short coherence time of mmWave channels. In this paper, we propose a multi-modal sensing-aided channel prediction (MSCP) technique to acquire real-time channel information. The core idea of MSCP is to exploit both vision sensing and radio frequency (RF) signaling to identify the positions of mobile devices and humans, as well as the positions, orientations, and materials of surrounding wireless environments (e.g., walls and ground). Since the vision sensor can capture sensing images quickly, we can analyze the surrounding environment seamlessly and accurately, thereby generating an accurate channel prediction. From the numerical results, we demonstrate that the proposed technique achieves more than 77.1 % improvement in channel prediction accuracy over the conventional RF signal-based channel prediction techniques.