Distributed Detection for Centralized and Decentralized Millimeter Wave Massive MIMO Sensor Networks

Multi-sensor millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) wireless sensor networks (WSNs) relying on both distributed (D-MIMO) and centralized (C-MIMO) configurations are conceived. Hybrid combining based low complexity fusion rules are constructed for the fusion center (FC) for both D-MIMO and C-MIMO systems employing a partially connected structure (PCS) and a fully connected structure (FCS), respectively. The decision rules are based on the transmission of local binary sensor decisions and also take into account the accuracy of local detection at the individual sensors. Closed-form analytical expressions are derived for the probabilities of false alarm and correct detection to analyze the system's performance. Furthermore, the asymptotic distributed detection (DD) performance corresponding to both antenna architectures is analyzed in the large-scale antenna regime along with the pertinent power scaling laws. Additionally, digital signaling matrices are designed for enhancing the system performance. Our simulation results quantify the performance gains of the proposed architectures, which closely match the analytical results.