Network Coding in TDMA-based Scheduling for Fluctuating Linear Wireless Sensor Networks

Wireless sensor networks have been applied in diverse applications, from facility monitoring to localized event surveillance, e.g., in power distribution grids of developing regions where traditional cables are not feasible. Linear wireless sensor networks are employed in such situations, where nodes link through lossy tandem links. Sensors generate and send data packets through intermediaries to reach network endpoints, where gateways forward aggregated data to a central server. Our previous research introduced a TDMA-based framework to ensure stable, cost-effective deployment, optimize delivery efficiency, and feature a proactive loss recovery mechanism in lossy links. However, as link loss increases, maintaining a high successful delivery probability that all packets from all nodes to the server becomes more challenging. Besides, the number of packets generated at sensors can fluctuate and sometimes exceed downstream link capacity, resulting in unrecovered packet losses. This paper investigates the benefit of using network coding schemes to enhance data transmission reliability in linear multi-hop wireless sensor networks, especially in a high link loss rate and packet fluctuation situations. The simulation results show a significant improvement in overall successful delivery probability and fairness (small deviations of success delivery probability among nodes) compared to the previous method in scenarios with a high link loss rate and moderate packet fluctuation.