Energy Complexity of Software in Embedded Systems

TL;DR

This paper introduces the concept of energy complexity to estimate energy consumption of algorithms in embedded systems, demonstrating its effectiveness with matrix multiplication as a test case.

Contribution

It proposes a novel energy complexity measure for algorithms, linking it with computational complexity to predict energy use in embedded systems.

Findings

Energy complexity combined with computational complexity accurately estimates energy consumption.

Matrix multiplication algorithms serve as effective test cases.

Energy complexity provides a new perspective for energy-efficient software design.

Abstract

The importance of low power consumption is widely acknowledged due to the increasing use of portable devices, which require minimizing the consumption of energy. The energy in a computational system depends heavily on the software being executed, since it determines the activity in the underlying circuitry. In this paper we introduce the notion of energy complexity of an algorithm for estimating the required energy consumption. As test vehicle we employ matrix multiplication algorithms and from the results it can be observed that energy complexity in combination with computational complexity, provides an accurate estimation for the energy consumed in the system.