Wind Powered Short-Range Communication Between Motorcyclist And Passenger With Alert System

Authors

  • Winarto Wijaya , Chandrasekharan Nataraj , Sathish Kumar Selva Perumal, Low Yee San , Raed Abdulla , K.C. Sriharipriya

Abstract

This paper presents Voice over Internet Protocol communication system for motorcyclists
interfaced via speech recognition module. It is also implemented the real-time collision detection alert system
for motorcyclists through IoT. A rider distraction is one of the main reasons for the motorcycle accidents. The
IoT based VOIP has developed to prevent the rider distraction and thereby the accident rate could be reduced.
The efficiency and accuracy of the system are evaluated by conducting five different experiments under the
following specific parameters. A latency of the voice communication, word error rate of speech recognition,
accuracy on both fall-detection system and GPS tracking system, and the power generation rate of wind power
generator against wind speed are used. The performance of the system is determined through the below results.
The measured value of latency delay from voice communication is 480–520 milli-seconds. The percentage of
word error rate on the accuracy of speech recognition is 42.3% in a quiet environment, whereas 61.65% in a
noisy environment. The accuracy of a helmet worn detection is 86.7%, the alert response rate is 90%, and the
accuracy of the GPS tracking system is 60%. The developed system has produced useful results and will
support to reduce the number of accidents reasonably.
Keywords--- Voice over IP Communication, Fall-detection, W

Published

2020-01-31

Issue

Section

Articles