A Unified Tensor-Based Joint Communication and Sensing Parameter Estimation for ISAC with Large-Scale User Access

The combination of active user detection (AUD) and integrated sensing and communication (ISAC) can be utilized to realize communication and sensing functionalities over one hardware platform in the ultra-massive machine-type communications (umMTC). However, due to the received signals are typically coupled in both AUD and ISAC, it is hard to obtain the communication and sensing parameters. To solve this problem, the actual channel model is first converted into a unified form through CANDECOMP/PARAFAC decomposition (CPD) to mitigate the interference between communication and radar signals. Then, by utilizing the matrix subspace-based method, the factor matrices are accurately estimated, where the equivalent path parameters can be extracted. Furthermore, to estimate the coupled path parameters, an alternating iterative estimation algorithm is proposed. Simulation results verify the superiority of our proposed joint communication and sensing parameter estimation algorithm in AUD, channel estimation, and radar sensing.