Learning automata and firefly algorithm based RPL for dynamic Internet of Things networks

Abstract The increasing number of Internet of Things (IoT) devices demands a rigid and adaptive routing methodologies to ensure the reliable communication especially for the low-power and lossy networks. The Routing Protocol for Low-Power and Lossy Networks (RPL) stands as an IoT communication standard. Nevertheless, its performance deteriorates when it is happened to be in the network environments facilitated with the dynamic nodes. This article introduces a combination of Learning Automata (LA) and the Firefly Algorithm (FA), optimized RPL (LA–FA–RPL) to focuses on enhancing the overall efficiency of mobile nodes present IoT network. The LA component facilitates parent selection while FA refines routing metrics through metaheuristic optimization. This combined approach is evaluated in the Contiki Cooja simulator by altering node densities and mobility rates. The results obtained indicate that LA–FA–RPL attains 15% increase in packet delivery ratio 18% boost in throughput and 12% decrease in energy usage compared to the conventional RPL, respectively. Based on the research outcomes the proposed LA–FA–RPL achieves improved routing, for upcoming IoT deployments. The first-ever hybrid integration of Learning Automata and the Firefly Algorithm into RPL, which enables joint adaptive learning and metaheuristic optimization for routing stability makes the study novel. The suggested methodology increases PDR, throughput, latency, and energy efficiency holistically across both static and mobile IoT contexts, in contrast to previous RPL advancements that optimize a single parameter.