Wearable Computing Going Green

One of the six Best Papers awarded at CHINACOM 2015, on August 2015 in Shanghai, was presented by a team coming from the School of Computer Engineering of the Nanyang Technological University in Singapore.

‘Wearable Computing Going Green: Energy-Optimal Data Transmission for Multi-Sensor Wearable Devices’ is the title of the research by Weizheng Hu, Weiwen Zhang, Han Hu and Yonggang Wen, whose scope is to investigate how to reduce the data-transmission energy for multi-sensor wearable devices over stochastic wireless channels.

State-of-the-art Nowadays, wearables devices are widely applied in healthcare to track people’s health status, by collecting an amount of data thanks to multiple sensors. Due to the small physical size and low battery capacity, wearable devices are inherently resource-constrained in energy, and a daily use implies a large consume of energy, resulting in issues which can affect users’ experience. Existing researches on how extending battery life focus on energy harvesting with small PV panels. Alternatively, a reduction of data sampling rate and novel communication protocols have been considered too.

A new approach In the study exposed at CHINACOM, the objective is to minimize the data transmission energy cost of multiple sensors by strategically setting the data transmission rate. Each sensor has been modeled as a queue, and each queue was limited by a total onboard buffer size. Then, the data transmission scheduling has been formulated as a constrained optimization problem subject to limited queueing buffer size. Finally, the researchers adopted the Lyapunov optimization framework to derive an online algorithm to control the data transmission for energy saving. The following results came out:

  • A fundamental tradeoff between energy cost and data-transmission delay.
  • A threshold effect of the total buffer size on the energy consumption, below which the buffer size limit will be active and the energy cost rises as the buffer size decreases.
  • Finally, compared to a random transmission algorithm, the new approach can save up to 85.08% of energy.

Future works Assuming the successful demonstration that the experimented method can significantly reduce the energy consumed by data transmission, next steps of the research will be to extend it from wearables to smartphones. Furthermore, real-world conditions, such as the possibility that different sensors may use different data frequencies, will be also taken in consideration.

The full research is available online on the European Union Digital Library (EUDL).