An Evolutionary Approach for Video Application Energy Consumption Estimation in Mobile Devices

  • Irandir O. de Amorim UPE
  • Jose F. V. de Melo UPE
  • Andson M. Balieiro UFRPE
  • Bruno B. dos Santos UPE


In the last years, the multimedia traffic has increased significantly and the mobile devices (e.g. smart phones and tablets) have been widely used to consume this content type. Video applications demand high energy consumption of the device because they perform complex operations and deal with a large data amount. Although hardware improvements in the mobile devices have been achieved, the advances in battery technology have not kept the same pace. In this respect, the combination of video applications with the limited battery capacity of the mobile devices has challenged the academia and industry in the development of techniques for energy management and provision of quality of experience (QoE) to the user. Energy consumption estimation models may assist these techniques, as well as, the decision made process when the computational offloading from the mobile device to the cloud is considered. This paper presents an evolutionary approach based on Genetic Algorithms (GAs) and Swarm Particle Optimization (PSO) for energy consumption estimation in mobile devices running video applications. The proposal is directly applicable to different model types (linear and non-linear ones), without the linearization cost, and it is evaluated in terms of mean squared error (MSE), using energy consumption measurement data of videos with different configurations. Results show the superiority of our proposal in comparison to the literature that adopts the Ordinary Least Squares method.
Como Citar

Selecione um Formato
AMORIM, Irandir O. de; MELO, Jose F. V. de; BALIEIRO, Andson M.; SANTOS, Bruno B. dos. An Evolutionary Approach for Video Application Energy Consumption Estimation in Mobile Devices. In: SIMPÓSIO BRASILEIRO DE SISTEMAS MULTIMÍDIA E WEB (WEBMEDIA), 25. , 2019, Rio de Janeiro. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2019 . p. 169-175.