Template-Driven DIY Routing for Energy-Constrained IoT: A Multiobjective Optimization Study

This study proposes a template-driven “DIY routing” method for battery-powered IoT networks. Three routing templates—Energy-Minimal, Balanced, and Reliability-Max—are defined using weighted combinations of ETX, residual energy, and hop count. Parameters are tuned using NSGA-II over 20,000 evolutionary iterations. The templates are evaluated on a 600-node Cooja simulation and a 120-node environmental sensing testbed deployed over 18 days. The Energy-Minimal template extends network lifetime by 34.1%, while Balanced improves PDR from 92.8% to 98.3%. Reliability-Max reduces 99th-percentile latency by 41.7% during congestion events. Compared to baseline RPL-EA, the templates consistently reduce per-node energy consumption by 0.11–0.19 J per hour with p < 0.01 (paired t-test). Results show that template-based routing provides a quantifiable and tunable compromise for developers lacking access to protocol internals.