The rapid growth of the Internet of Things (IoT) and wearable healthcare devices has intensified the need for ultra-low-power Very Large Scale Integration (VLSI) designs that can operate reliably under changing and often unpredictable workloads. Established low-power techniques, such as clock gating, voltage–frequency scaling, adiabatic logic, and operand isolation, offer meaningful energy savings but are often limited by their dependence on stable workload patterns. This review examines a broad range of bio-inspired dynamic power optimization strategies that build on conventional circuit-level approaches by introducing adaptive, feedback-driven control. Through a systematic analysis of Particle Swarm Optimization (PSO), Genetic Algorithms (GA), swarm intelligence models, and hybrid metaheuristics, the study highlights their suitability for real-time power management in IoT and wearable systems. Reported implementations show power reductions from 18.5% to 82% across platforms such as wireless sensor networks, FPGA-based biomedical devices, and neuromorphic circuits. Additional improvements emerge when these algorithms are combined with advanced circuit techniques such as multi-modal power gating, retention SRAM, and CMOS–NEMS hybrids, contributing to longer device lifetimes and the possibility of energy-neutral operation. The review also integrates insights from developments in process technologies, subthreshold and steep-slope devices, and neuromorphic computing to frame the broader relevance of bio-inspired optimization in future low-power architectures. Overall, by linking deterministic hardware design methods with adaptive, bio-inspired intelligence, this work outlines a practical direction for achieving scalable, energy-efficient, and reliable solutions for IoT and wearable applications. The findings indicate that blending traditional low-power approaches with bio-inspired strategies can more effectively address workload variability while supporting sustainable and autonomous electronic systems.
Pasaya et al. (Mon,) studied this question.