Blind Distributed Detection in MIMO Networks

This work considers a blind detection problem in the distributed multiple-input multiple-output (MIMO) networks, i.e., wireless sensor, co-operative, IoT networks, etc., without knowing the encoding schemes of users, or sensors, and channel state information (CSI) between users and fusion center (FC). Each user observes the source phenomenon, encodes their observations, and transmit to the FC using same time-frequency channel resource, i.e., coherent multiple-access channel (MAC). Exploiting encoding and decision vectors across users to be orthogonal, a novel maximum-likelihood (ML) criteria-based constrained alternating least squares (CALS) algorithm is proposed for the blind distributed detection problem in the MIMO networks. Finally, simulation results illustrate the efficacy of the proposed CALS detection algorithm.