Analysis of the computational costs of an evolutionary fuzzy rule-based internet-of-things energy management approach

Abstract

This study presents an in-depth analysis of the computational costs associated with the application of an Evolutionary Fuzzy Rule-based (EFR) energy management system for Internet of Things (IoT) devices. In energy-harvesting IoT nodes, energy management is critical for sustaining long-term operation. The proposed EFR approach integrates fuzzy logic and genetic programming to autonomously control energy consumption based on available resources. The study evaluates the system's computational performance, particularly focusing on processing time, RAM and flash memory usage across various hardware configurations. Different compiler optimization levels and floating-point unit (FPU) settings were also explored, comparing standard and pre-compiled algorithms. The results reveal computational times ranging from 2.43 to 5.23 ms, RAM usage peaking at 6.23 kB, and flash memory consumption between 19 kB and 32 kB. A significant reduction in computational overhead is achieved with optimized compiler settings and hardware FPU, highlighting the feasibility of deploying EFR-based energy management systems in low-power, resource-constrained IoT environments. The findings demonstrate the trade-offs between computational efficiency and energy management, with particular benefits observed in scenarios requiring real-time control in remote and energy-limited environments.