Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Approxify: Automating Energy-Accuracy Trade-offs in Batteryless IoT Devices

Muhammad Abdullah Soomro, Naveed Anwar Bhatti, Muhammad Hamad Alizai

TL;DR

Batteryless IoT devices face intermittent power, making checkpointing costly. Approxify automates approximation techniques (loop perforation and function memoization) guided by energy traces to reduce checkpointing while bounding error, achieving significant energy savings. The approach is validated on SUSAN and LQI, showing up to ~40–50% checkpoint reductions with manageable fidelity loss, and corroborated by a real hardware testbed and emulators. This work advances practical energy-aware computing for intermittently powered IoT by balancing energy efficiency with computation accuracy, enabling more scalable green networking in constrained environments.

Abstract

Batteryless IoT devices, powered by energy harvesting, face significant challenges in maintaining operational efficiency and reliability due to intermittent power availability. Traditional checkpointing mechanisms, while essential for preserving computational state, introduce considerable energy and time overheads. This paper introduces Approxify, an automated framework that significantly enhances the sustainability and performance of batteryless IoT networks by reducing energy consumption by approximately 40% through intelligent approximation techniques. \tool balances energy efficiency with computational accuracy, ensuring reliable operation without compromising essential functionalities. Our evaluation of applications, SUSAN and Link Quality Indicator (LQI), demonstrates significant reductions in checkpoint frequency and energy usage while maintaining acceptable error bounds.

Approxify: Automating Energy-Accuracy Trade-offs in Batteryless IoT Devices

TL;DR

Batteryless IoT devices face intermittent power, making checkpointing costly. Approxify automates approximation techniques (loop perforation and function memoization) guided by energy traces to reduce checkpointing while bounding error, achieving significant energy savings. The approach is validated on SUSAN and LQI, showing up to ~40–50% checkpoint reductions with manageable fidelity loss, and corroborated by a real hardware testbed and emulators. This work advances practical energy-aware computing for intermittently powered IoT by balancing energy efficiency with computation accuracy, enabling more scalable green networking in constrained environments.

Abstract

Batteryless IoT devices, powered by energy harvesting, face significant challenges in maintaining operational efficiency and reliability due to intermittent power availability. Traditional checkpointing mechanisms, while essential for preserving computational state, introduce considerable energy and time overheads. This paper introduces Approxify, an automated framework that significantly enhances the sustainability and performance of batteryless IoT networks by reducing energy consumption by approximately 40% through intelligent approximation techniques. \tool balances energy efficiency with computational accuracy, ensuring reliable operation without compromising essential functionalities. Our evaluation of applications, SUSAN and Link Quality Indicator (LQI), demonstrates significant reductions in checkpoint frequency and energy usage while maintaining acceptable error bounds.

Paper Structure

This paper contains 23 sections, 4 equations, 5 figures, 2 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Approxify
  3. Design Goals
  4. Approximation Techniques
  5. Loop Perforation
  6. Function Approximations
  7. System Architecture
  8. Code Parser
  9. Output Validator
  10. ApproxSafe Mapper
  11. Approximator
  12. Implementation
  13. Parsing and Modifying the Source Code
  14. Performance and Accuracy Evaluation
  15. Evaluation
  16. ...and 8 more sections

Figures (5)

  • Figure 1: Approxify architecture.
  • Figure 2: Performance index vs. approximation percentage for two applications with smallest capacitors.
  • Figure 3: Testbed and schematic for checkpointing validation
  • Figure 4: Energy traces
  • Figure 5: Performance of the SUSAN under dynamic energy harvesting with 220uF and 330uF capacitors. The graphs show RF energy harvested voltage and checkpointing instances with and without approximation, validating Approxify's efficieny.