Millimeter-Wave Sensing and Mapping: A Low-Rank CP Decomposition Approach

Integration sensing and communication (ISAC) has been envisaged as a promising candidate for 6G wireless communications. In this paper, we address a user-centric environment sensing and mapping on millimeter-Wave (mmWave) orthogonal frequency division multiplexing (OFDM) systems. By leveraging the inherent sparse characteristics of mmWave channels, the received signal in terms of the single-bounce reflected paths is formulated by a low-rank third-order tensor that admits a tensor rank decomposition. A structured CANDECOMP/PARAFAC (CP) decomposition based method is then developed to extract the sensing channel parameters for environment mapping. Simulation results show that the proposed method can achieve a centi-meter accuracy of localization and mapping.