Design of a resource-constrained IoT sensor node architecture with fault-tolerant telemetry for environmental monitoring systems

Abstract

Environmental sensor networks deployed in remote, power- and connectivity-constrained locations require careful co-design of the sensing, communication, and power subsystems. This paper addresses the design and optimization of a low-power IoT sensor node architecture with fault-tolerant telemetry, using continuous water-resource monitoring in an arid region as a representative deployment case. A four-layer system architecture is proposed, and the connectivity layer is optimized using store-and-forward buffering and dynamic channel allocation adapted from SDN-enabled LEO satellite networks, reducing the design cost of achieving reliable data delivery over intermittent GSM/LoRa links. Analytical models of link availability, buffer occupancy, and solar energy budget are derived to guide design trade-offs. Simulation results show that the optimized connectivity design raises packet delivery success rate from 61.2% to 92.8% at 60% link availability, demonstrating an effective low-cost design strategy for resource-constrained edge sensing systems.