Multi-Gateway Polling for Nanonetworks under Dynamic IoT Backhaul Bandwidth

As an emerging part of the Internet of Things (IoT), Electromagnetic-based Wireless NanoSensor Networks (EM-WNSNs) working in the TeraHertz (THz) band are envisaged to enable applications that demand high sensing resolution. For the high-speed nanonetworks deployed for time critical applications, the limited IoT backhaul bandwidth of the existing infrastructure becomes the bottleneck that deteriorates resource utilization efficiency. On the one hand, the bandwidth demand exceeding the supply results in unnecessary energy consumption on nano-devices due to the overdue data discarded. On the other hand, the excess supply of bandwidth reduces the bandwidth efficiency. To address this problem, the On-demand Efficient (OE) polling and On-demand Probabilistic (OP) polling were proposed. Nevertheless, the existing solutions only apply to the single-gateway scenario whereas deploying multiple gateways is required for achieving high data completeness and energy efficiency. Therefore, in this paper, we propose the Multi-gateway Probabilistic (MP) polling for the multi-gateway EMWNSNs. To achieve high bandwidth efficiency, in MP polling, one master gateway is selected for OP polling with the aggregated bandwidth that reduces the probabilistic uncertainty. Next, all gateways probabilistically receive packets using the bandwidth-based receiving windows. To the best of our knowledge, MP Polling is the first polling solution designed for the multi gateway EM-WNSNs under dynamic backhaul bandwidth.