Optimized Design and Simulation of Ultra-Low Power Embedded Systems for Energy-Constrained Applications

Abstract

The demand of the ultra-low power embedded systematic architecture has been increased as the battery-powered devices are rapidly growing in the Internet of Things (IoT), wearable health monitoring, and edge computing systems. The traditional architectures have a tendency to compromise and not achieve sufficient performance within the limited energy they have resulting to early power exhaustion and reduced working life. In this paper, we propose an optimized design and simulation framework that best suits energy-constrained setting, with particle emphasis being architectural and algorithmic power optimization. The idea is to equate the proposed system with a RISC-V-based ultra-low power processor core, low-leakage memory modules, event-driven sensor connection, and dynamic sleep-mode policy-based scheduling. An architecture model was made with the SystemC-TLM 2.0 co-simulation environment, whereas power and timing analysis were performed by Cadence Voltus and Synopsys PrimeTime PX. Periodic temperature sensing, BLE communication, motion tracking in the application-level workloads were used in the simulation to assess the performance. The conducted results reveal a 42 percent average power savings, and 30 percent better energy-delay product (EDP) than typical designs in a baseline. The results confirm the eligibility of the proposed framework in the field of allowing long-duration, sustainable operation of embedded systems in use conditions of power-critical and real-time, and the utilization of industrial implementation-standard tools of simulation and analysis.