Every hour, nearly 40 people under the age of 25 die in road accidents around the world. According to the World Health Organization, this is the second most important cause of death for 5- to 29-year-olds. In India, drunk driving and inefficient law enforcements are major contributing factors. The current system of visual identification of traffic violation, conducted by the traffic authorities, cannot work everywhere and every time. There is a great demand for simple and cost-effective solutions to traffic safety problem. In this paper, we propose a traffic violation detection technique for vehicular ad hoc networks to detect crossing speed limits and analyzing the behavior of driver. In this work, we used a sensor device, a digital map and GPS-based system for area of 1000 m × 1000 m. We analyzed the behavior of each vehicle in the network. Here, we have divided a network into a number of clusters, and each cluster has an infrastructure node (base station); the infrastructure node will be the point of contact for all the vehicles in that area. All infrastructure nodes communicate with a control center (master control room). If the driver violates traffic rule(s), then the infrastructure node will send an alert message to the control center. We have simulated our proposed model on a graphics package, and the simulation result suggests that drunken drivers can no longer escape from the law enforcers, which is the foundation for traffic safety. Copyright © 2012 John Wiley & Sons, Ltd.