Exploiting Cell-Free Massive MIMO for Computational Imaging

Wireless sensing is a promising use case of the next-generation wireless networks. In order to achieve accurate sensing with limited radio resources, computational imaging (CI) techniques have been applied to wireless sensing by making use of computational capabilities of wireless networks. With the goal of balancing radio resources and computational capabilities, this paper proposes a cooperative CI algorithm by exploiting the benefits of cell-free massive MIMO (CF mMIMO). According to the distributed architecture of CF mMIMO, each access point (AP) independently obtains a local estimate of the interested target based on the received sensing signals sent by multiple user equipments (UEs) with a Bayesian inference method, and then the central processing unit (CPU) fuses all local estimates from the APs to obtain an accurate image of the target with an optimal combination method. Both theoretical analysis and simulation results validate the effectiveness of the proposed algorithm over baseline algorithms for wireless sensing.