Proceedings of 4th RIT Post Graduates Conference (RIT PG Con-18) NOVATEUR PUBLICATIONS JournalNX- A Multidisciplinary Peer Reviewed Journal (ISSN No:2581-4230) April, 13th, 2018 270 | Page TRAFFIC CONGESTION DETECTION WITH COMPLEX EVENT PROCESSING IN VANET Priyanka Shinde Department of Electronics and Telecommunication, Ramrao Adik Institute of Technology (RAIT), Navi Mumbai 400706 University of Mumbai Manoj M. Dongre Department of Electronics and Telecommunication, Ramrao Adik Institute of Technology (RAIT), Navi Mumbai 400706 University of Mumbai Abstract—Magnitude of urban population and unplanned development of cities have led to road traffic congestion in major cities and added increasing pressure on overall road transport and road related infrastructure. Nowadays, intelligent transportation systems such as Vehicular Ad Hoc Networks (VANET) are used for distributed road traffic. VANET is a wireless network that gathers complex and randomly generated data related to distributed traffic along with other information such as weather con ditions on real time basis. In this paper, the Author has attempted to implement a system on Event Driven Architecture (EDA) to detect complex but similar levels of traffic congestion which resembles real life traffic situation. The system is designed with the help of a technique called Complex Event Processing (CEP) that treats all messages in the network from corresponding vehicles as an event. The effectiveness of this system can be observed by using different performance metrics. Keywords—Complex Event Processing (CEP), Event Driven Architecture (EDA), Traffic Congestion, Vehicular Ad Hoc Networks (VANET) I. INTRODUCTION The rapid growth in urban population and unplanned development has led to traffic congestion in major cities and towns. The traffic congestion not only makes driving experience stressful for vehicle-operator but also detrimental to the environment and causes economic losses to the society at large. Thus, detection of traffic congestion is a major problem on urban roads and highways. The intelligent transportation systems are increasingly used in identifying the traffic congestion. One of the intelligent transportation systems is vehicular ad hoc network (VANET) through which inter- vehicular communication is usually achieved. A VANET is a mobile network whose nodes are vehicles that drive on a road and communicate each other through wireless technology. In addition, each vehicle of the network broadcasts data messages at regular time intervals called beacon messages, which comprise, among other data, vehicle’s current speed and position. This way, each vehicle of the VANET manages to detect other vehicles in its surrounding; however, handling of beacon messages from various vehicles on a continuous basis would imply a high volume of events across the VANET that each vehicle of the network must deal with. To process such a large amount of events, Event-Driven Architecture (EDA) is used. EDA is a software architecture pattern promoting the production, detection, consumption of, and reaction to events. Thus, a high volume of events could easily be processed with minimum delay. Currently, EDAs play a pivotal role in many business areas such as financial industries or supply chain management. Major component of EDA is a complex event processing (CEP) which deals with events of the network. For example, given a road area where an Information System (IS) is used to detect traffic congestion, relevant real-world vehicles’ movements are reflected as events in the lowest layers of the IS. Overall, these events form complex relationships that establish particular patterns. CEP enacts such a way that it is possible to become aware of a particular pattern of real-world vehicles’ movements by monitoring the IS and looking for the predefined patterns i.e. vehicles’ speed to identify traffic jam The key task of CEP is the continuous identification of complex event patterns in the streams of events that flow through the IS. A CEP system can be seen as a sophisticated form of EDA that deals with a large number of heterogeneous events from different streams and perform event aggregation and correlation in a very short time. Such systems are suitable for distributed environments specially road traffic, whose dispersed elements act as data sources, and as a result, a great deal of global heterogeneous data is generated; hence, CEP could be used to get insight into these streams and detect useful information. Thus, VANETs are a clear example of that kind of environments. Thus, here we have designed a programmed CEP based EDA. Further, through simulation, random events are generated which resemble a real life traffic congestion situation. The vehicles are assumed to be equipped with VANET. Thus, Vehicles are broadcasting beacon messages to other vehicles. With the help of CEP, the complex events are broken in sub-activities that match with predefined patterns. As a result, traffic congestion points are identified on a motor road. Further, the system performance is measured in terms of the packet delivery ratio, throughput of the network and amount of delay in receiving and sending messages between vehicles. II. RELATED WORK The cooperative approach to traffic congestion detection with complex event processing and VANET are explained in ref [1]. In ref [2] proposes a congestion detection and notification scheme using VANETs for urban roads. Additionally, the scheme develops a spatial–temporal effectiveness model based on the potential energy theory