Quasi-Deterministic Modeling for Industrial IoT Channels Based on Millimeter Wave Measurements.

The Industrial Internet of Things (IIoT) enables machines to communicate robustly. High reliability, high throughput, and low latency are the critical capabilities of IIoTs, which have posed great challenges to existing wireless solutions for industrial applications. Due to the vast available bandwidth, the emerging millimeter-wave (mmWave) technology is promising to address this bottleneck. However, the propagation behaviors at such high frequencies in the harsh industrial environment have yet been well understood. In this work, extensive measurements have been conducted in a representative industrial application scenario using a 2 GHz wideband directional channel sounder in the 28 GHz mmWave band. By exploiting the measurement with excellent resolution, the multipath components’ (MPCs’) delay-angular space is transformed onto the scatter points (SPs) in the propagation environment. An effective clustering algorithm is then proposed to cluster the SPs without prior knowledge and iterations. Through a geometrical optics analysis, the SP clusters are classified corresponding to the reflectors. By doing this, the cluster-generating reflectors are reduced to a quasi-deterministic (QD) channel model that ensures spatial consistency and MPCs’ stochastic dispersion. Finally, it is shown that the measurement data agrees well with the proposed QD model, indicating the high fidelity of the proposed model.