Adaptive and Intelligent Route Learning for Mobile Assets Using Geo-tracking and Context Profiles

Currently available asset tracking systems are not cost effective in mobile and wireless computing infrastructures, and they do not learn the asset's route to act accordingly. Hence our main goal is to not only to track mobile assets by using an efficient geographical tracking approach, but also adapt their routes by means of intelligent route learning techniques. We thus designed and implemented an adaptive learning based scheme that makes an optimized judgment of data transmission to reduce the transmission rate and increase the tracking and routing precision. The experimental learning mechanism customizes tracking, routing, and reporting and generates traces which are more fitting with the actual routes to develop a realistic approach. Context profiles, which indicate the characteristics of a route based on environmental conditions, are utilized to dynamically represent the values of the asset's properties. This paper is complemented with the implementation infrastructure and protocol evaluations that prove that significant costs can be saved and operational efficiency can be achieved.