A new approach to estimate real-time traveling speed with accelerometer

The research of real-time traffic condition are mainly concentrated on GPS module, which is energy-intensive and become invalid when suffering from enclosed areas. In this paper, taking advantage of accelerometer embed in mobile devices, we propose a new approach to estimate the vehicle traveling speed. We analyze the characteristics of time domain and frequency domain, and have discovered excellent characteristics. We have trained the classified model to match the acceleration with traveling speed successfully and achieve a good performance. Therefore, we could evaluate the traffic congestion scale with the estimated value. With this approach, we could get the traffic information in closed areas. The experiment result has achieved a satisfied accuracy, and above all the accelerometer is more energy-efficient than the GPS module.