Adaptive Multi-VDD VLSI Architecture for Ultra-Low-Power Edge Computing Applications

Authors

  • K P Uvarajan Department of Electronics and Communication Engineering, KSR College of Engineering, Tiruchengode

Keywords:

Adaptive Multi-VDD, Ultra-Low-Power VLSI, Edge Computing, Dynamic Voltage Scaling, Power Gating, Energy-Efficient Architectures

Abstract

Applications of edge computing have been growing rapidly, and for the applications that are emerging there is a need for VLSI architectures that are ultra-low power and energy efficient to be able to support real-time processing with tight specifications. In conventional single-VDD architectures, the high dynamic power consumption and leakage power are not suitable for battery-powered and resource-limited edge devices. The present paper presents an adaptive Multi-VDD VLSI platform for ultra-low consumption edge computing applications. The proposed architecture uses dynamically controlled voltage islands and adaptive voltage scaling, along with power-gating mechanisms, to achieve optimum energy usage based on workload variation. To further minimize unnecessary power waste, there is also a workload-aware control strategy employed which allows for switching between multiple voltage domains in order to maintain computing performance. The effectiveness of the proposed system is analyzed using mathematical power models of the dynamic power and leakage power. The architecture is emulated in Cadence Virtuoso / HSPICE environment in various workload conditions using 45nm CMOS technology. The experimental results show that the dynamic power and leakage power are greatly reduced compared to the conventional single-VDD and static Multi-VDD systems without compromising the delay performance of the next-generation edge computing systems.

Downloads

Published

2026-01-03

How to Cite

K P Uvarajan. (2026). Adaptive Multi-VDD VLSI Architecture for Ultra-Low-Power Edge Computing Applications. Annals of Energy-Efficient VLSI Architectures, 1(1), 111–119. Retrieved from https://iaeces.com/index.php/AEEVA/article/view/15

Issue

Section

Articles