Duty-Cycling is Not Enough in Constrained IoT Networking: Revealing the Energy Savings of Dynamic Clock Scaling

TL;DR

This paper demonstrates that combining dynamic clock scaling with duty-cycling significantly enhances energy efficiency in constrained IoT devices, leading to substantial battery life extension.

Contribution

It introduces the integration of dynamic voltage and frequency scaling (DVFS) into IoT operating systems and systematically evaluates its energy-saving benefits in real-world networking scenarios.

Findings

Energy savings of 24-52% for MAC operations

Up to 37% energy reduction for encrypted CoAP communication

DVFS reconfiguration overhead is minimal compared to energy savings

Abstract

Minimizing energy consumption of low-power wireless nodes is a persistent challenge from the constrained Internet of Things (IoT). In this paper, we start from the observation that constrained IoT devices have largely different hardware (im-)balances than full-scale machines. We find that the performance gap between MCU and network throughput on constrained devices enables minimal energy delay product (EDP) for IoT networking at largely reduced clock frequencies. We analyze the potentials by integrating dynamic voltage and frequency scaling (DVFS) into the RIOT IoT operating system and show that the DVFS reconfiguration overhead stays below the energy saved for a single, downscaled MAC operation. Backed by these findings, we systematically investigate how DVFS further improves energy-efficiency for common networking tasks -- in addition to duty-cycling. We measure IoT communication scenarios between real-world systems and analyze two MAC operating modes -- CSMA/CA and time slotting -- in combination with different CoAP transactions, payload sizes, as well as DTLS transport encryption. Our experiments reveal energy savings between 24% and 52% for MAC operations and up to 37% for encrypted CoAP communication. These results shall encourage research and system design work to integrate DVFS in future IoT devices for performing tasks at their optimal frequencies and thereby significantly extending battery lifetimes.