Research Information System
IEEE 42nd Conference on Local Computer Networks (IEEE LCN), Singapore, Singapore, 9 - 12 October 2017, pp.61-65
In this paper we propose a vehicular speed learning framework that recommends best traffic load based on a particularly required latency and throughput conditions to be achieved. The framework is composed of two main layers, the base layer and two enhancement layers. The base layer aims at providing an in-vehicle wireless receiver to inform the driver about the speed limit within the area he/she is in. The first enhancement layer aims at providing the driver with a high level intelligence that can be utilized for various applications. The second enhancement layer aims at utilizing smartphones as an intelligent sensor. By having the vehicle speed and the speed limits available, the smartphone can be used to detect particular conditions such as congestions, accidents or defects in the roads. Observing drivers' monitored behavior while given the current road conditions, the system extrapolates from these new and other previously available observations to predict an optimal speed limit. In this fashion, the framework is able to continuously adapt and learn autonomously to manage efficient and safe transportation.