Energy‐Aware Learning Automata‐Based Backoff Algorithm in Cyber‐Physical Systems

ABSTRACT Cyber‐physical systems have emerged as a burgeoning field of research, encompassing a diverse array of network types. The complexity inherent in these heterogeneous networks necessitates diligent attention to challenges. One pivotal facet lies within the realm of the medium access control (MAC) layer, a pivotal determinant of system performance. This paper main objective is to reduce the number of collisions alongside minimized energy utilization. To tackle this goal, a novel energy‐aware learning automata‐based backoff algorithm, denoted as ELABA_CPS is introduced. This algorithm operates on nested case: inner loop, the Intranet, which represents a node within a specific network, and the outer loop, the Internet, a network entity within the broader cyber‐physical framework. Through the utilization of learning automata, the optimal number of attempts required to access channels at both Intranet and Internet levels is ascertained. ELABA_CPS enhances energy efficiency and reduces collisions in cyber‐physical systems (CPS). Compared with the energy hashed virtual backoff algorithm (EHVBA), ELABA_CPS presents a dual‐layer decision‐making process that optimizes performance at both the Intranet and Internet levels. The algorithm exhibit enhancements in key metrics such as packet delivery ratio, energy consumption, throughput, and collision rate. The proposed work transcends theoretical considerations, undergoing comprehensive testing, evaluation, and comparative analysis vis‐à‐vis the energy hashed virtual backoff algorithm (EHVBA). ELABA_CPS emerges as the frontrunner, boasting a 16.4% enhancement in packet delivery ratio, a remarkable 31.4% surge in MAC efficiency, a substantial 25.8% upswing in throughput, a remarkable 54.7% reduction in energy consumption, and a notable 31.6% diminution in collision occurrences. Collectively, these findings unequivocally underscore the superior performance of ELABA_CPS when compared to EHVBA.