Performance Evaluation of Wearable Computing Frameworks

  • Michelle Cacais
  • Danielo G. Gomes
  • Paulo Armando C. Aguilar
  • Rossana M. C. Andrade

Resumo


Internet of Things (IoT) aims to connect multiple devices and enable communication on a global scale. In this context, wearable computing turns into one of the most imperative technologies for data capture and transfer. Even though many solutions exist for fast development of wearable applications, it is difficult to know which one is the best for each situation. In this paper, we evaluated the performance of three wearable computing frameworks: SPINE, BNSM and IBM Bluemix. We used performance metrics such as data accuracy, memory footprint and energy consumption in a movement recognition scenario. Results have shown that, among the analyzed frameworks, (i) BNSM is the best one concerning energy-efficiency; (ii) IBM Bluemix is the best one for data accuracy. Both frameworks presented interesting results in terms of memory footprint.

Publicado
06/07/2017
Como Citar

Selecione um Formato
CACAIS, Michelle; GOMES, Danielo G.; AGUILAR, Paulo Armando C.; ANDRADE, Rossana M. C.. Performance Evaluation of Wearable Computing Frameworks. In: WORKSHOP EM DESEMPENHO DE SISTEMAS COMPUTACIONAIS E DE COMUNICAÇÃO (WPERFORMANCE), 16. , 2017, São Paulo. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2017 . ISSN 2595-6167. DOI: https://doi.org/10.5753/wperformance.2017.3354.