presentation_A-Study-of-Bluetooth-Low-Energy-Performance-for-Human-Proximity-Detection-in-the-Workplace

Paper Summary

Paper Name A Study of Bluetooth Low Energy Performance for Human Proximity Detection in the Workplace
Published in 2017 IEEE International Conference on Pervasive Computing and Communications (PerCom)
Authors Alessandro Montanari ; Sarfraz Nawaz ; Cecilia Mascolo ; Kerstin Sailer
Paper Overview This paper proposes an innovative way for social interaction sensing by using BLE devices. This paper contains:

  • A detailed analysis of BLE parameters that play a central role in proximity detection
  • Analysis of BLE capabilities and limitations on commercial wearable devices
  • A machine learning method for human proximity detect
  • A experiment using custom device that can get 97% accuracy
BLE parameters analysis

(Fig. 1. Average number of received packets changing the Advertising Interval and the number of transmitting devices.)

(Fig. 2. Average number of received packets changing the Advertising Interval, the Scan Interval and the Scan Window.)

(Fig. 3. Parameters’ Impact on Power Consumption)
Experiment results