Intelligent traffic management with wireless sensor networks

Vehicular travel is gaining importance everywhere, particularly in large urban areas. The current technologies that support vehicular travel like wired sensors, inductive loops, surveillance camera etc., are expensive and also require high maintenance cost. Further the accuracy of these devices also depends on environment conditions. The typical traditional approaches attempt to optimize traffic lights control for a particular density and configuration of traffic. However, the major disadvantage of using these techniques is that the dynamic behavior of traffic densities and configurations change is difficult to model constantly. Traffic seems to be an adaptation problem rather than an optimization problem. This paper therefore tries to address the above issue, and hence we propose algorithms which perform adaptive traffic light control using a wireless sensor network setup. The paper aims at analyzing methods to build an intelligent system that can blend and support some of the existing technologies of traffic control and therefore reduce the average waiting time of vehicles on a junction. The proposed algorithms are adaptive to traffic flow at any intersection point of roads. Simulations of the real-life traffic scenarios are conducted in a simulated platform called Green Light District Simulator (GLD) to generate graph average waiting time versus cycles. The results generated show that the proposed method is effective for the traffic control in a real road intersection.