An energy-efficient adaptive beaconing rate management for pedestrian safety: A fuzzy logic-based approach

Pedestrian road-safety applications require geolocation data at high refresh rate. This is achieved through high rate beaconing of the data from the safety devices owned by road users to servers. The servers perform threat analysis and send alerts to road users in case of risk of collisions. However, when low-battery devices like smartphones are used for such applications, they drain their energy rapidly as high beaconing rate entails quick energy consumption. In this paper, we propose an energy-efficient fuzzy logic-based adaptive beaconing rate management system while keeping traffic crashes on pedestrians as low as possible. The fuzzy inference system predicts road accident risk levels of pedestrians which are finally used to decide beaconing rates. The prediction relies on factors related to the pedestrians, drivers, vehicles, roads, environments, etc. The risk prediction accuracy of the fuzzy logic-based system is evaluated by comparing its output with the risk level prediction algorithm that relies on minimum distance for information exchange to predict the likelihood of accidents. Results show that collision risk level prediction of the two methods is the same for more than 90% of accident records investigated. Energy-efficiency evaluation of the adaptive beaconing rate management depicts that, the battery life of a mobile device can be doubled in comparison with a method that uses a mismanaged high rate beaconing frequency. Additional evaluations portray that energy consumption of adaptive beaconing rate is affected by vehicle arrival rates and factors that affect traffic safety. Comparative analysis with state-of-the-art related works indicates that despite considering several collision risk-pulling factors ignored by other researchers, the proposed solution has negligible energy overhead. (C) 2020 Elsevier B.V. All rights reserved.