Throughput optimization in backscatter-assisted wireless-powered underground sensor networks for smart agriculture

Wireless underground sensor networks (WUSNs) using wirelessly-connected buried sensors enable smart agriculture through real-time soil sensing, timely decision-making, and precise remote operation. Energy harvesting technology is adopted in WUSNs, implying wireless-powered underground sensor networks (WPUSNs), to prolong the network lifetime. In addition, the backscatter communication (BSC) technology seems promising for improving the utilization of resources and network throughput according to preliminary studies in terrestrial wireless-powered communication networks. However, this technique has not yet been investigated in WPUSNs, where channel impairments are incredibly severe. In this work, we aim to assess BSC’s performance in WPUSNs and evaluate its feasibility for sustainable smart agriculture. For this, we first conceptualize a multi-user backscatter-assisted WPUSN (BS-WPUSN), where a set of energy-constrained underground sensors (USs) backscatter and/or harvest the radio frequency energy emitted by an above-ground power source before the sensed data are transmitted to a nearby above-ground access point. Then, we formulate the optimal time allocation to maximize the network throughput while assuring real-world users’ quality of service (QoS). Our analysis considers the non-linearities of practical energy harvesting circuits and severe signal attenuation in underground channels. By simulating a realistic farming scenario, we show that our proposed solution outperforms two baseline schemes, i.e., underground harvest-then-transmit and underground BSC, by an average of 12% and 358% increase in network throughput (when USs are buried at 0.35 m), respectively. Additionally, several trade-offs between the network throughput, time allocation, network configurations, and underground parameters are identified to facilitate the practical implementation of BS-WPUSNs.