Multi-View Millimeter-Wave Imaging Over Wireless Cellular Network

Millimeter-wave (mmWave) imaging over wireless networks is one of the potential technologies in the design of integrated sensing and communication (ISAC) systems. To achieve complete and accurate sensing of the large-scale complex environment, multiple views from different user equipments (UEs) and base stations (BSs) in a wireless network should be fully and cooperatively exploited. In this paper, based on the uplink channels of the wireless cellular network, we propose a multi-view mmWave imaging architecture. In the proposed architecture, a single BS centrally or multiple BSs jointly process the transmitted data of UEs. Taking into account the complex physical propagation characteristics of mmWave in the environment, especially the occlusion effect, we exploit the multi-view sensing of the environment from various UEs and BSs. To solve the multi-view sensing problem for the considered model, we propose a generalized-approximate-message-passing-based multi-view sparse vector reconstruction (GAMP-MVSVR) algorithm to obtain the imaging results. In the proposed algorithm, a multi-layer factor graph is proposed to describe the data receiving and sending relationship, as well as the occlusion effect of mmWave propagation. The sum-product algorithm (SPA) is used to iteratively solve the imaging result. Specifically in each iteration, the occlusion relationship between the target object in the environment is recalculated according to the proposed occlusion detection rule, and in turn, used to estimate the scattering coefficients of the target objects. Simulation results verify the effectiveness of the proposed algorithm.