ISSN 2277 2277-2685 IJESR/Sept. 2015/ Vol-5/Issue 5/Issue-9/1248-1253 T. Lakshmi Deepak et.al.,/ International Journal of Engineering & Science Research WIRELESS SENSOR APPLICATION IN TRAFFIC CONGESTION CONTROL AND CLASSIFICATION OF VEHICLES T. Lakshmi Deepak* eepak*1, D. Hepzibha Rani2, D. Ravi Kumar3 1 2 M. Tech (Embedded Systems), Dept. of ECE, Gudlavalleru Engineering College, Seshadrirao Knowledge Village, Gudlavalleru (A.P), India. Asst. Prof, Dept. of ECE, Gudlavalleru Engineering College, Seshadrirao Knowledge Village, Gudlavalleru (A.P), India. 3 DGM, BEL, Bangalore, India. ABSTRACT In today’s y’s world the vehicular traffic is increasing constantly, especially in case of urban traffic system, existing traffic management solutions are not efficient enough to meet the requirement. It is being witnessed in our daily life through persistent traffic jam and high number of accidents. Wireless sensor networks (WSN) based intelligent transportation systems (ITS) had emerged as a cost effective solution that bear a pivotal potential to overcome these difficulties. Technology enables a new broad range of smart city applications around urban areas and these sensing technologies includes traffic safety, traffic congestion control, road state monitoring, vehicular warning services, and parking management. The main contribution of this paper is to identify the presence of the vehicles and to classify vehicles based on the collected data using WSNs based ITS. 1. INTRODUCTION In today’s world, number of traffic jams and accidents in urban and metropolitan areas are constantly increasing towards more stressful and nd leads to dramatic consequences on human health, and environment. Intelligent transportation System (ITS) detects vehicles in predefined positions and these are based on bulky and power hungry devices which are using wired technologies for communication and power supply. This increases their installation, maintenance, and reparation cost and subverts the scalability of ITS affecting thus their major objectives. Advances in embedded systems and wireless technology give birth to wireless sensor networks (WS (WSNs) which are composed of cheap, effective and tiny devices that communicate wirelessly and sense the surrounding environment. Each device node contains a sensor, a processor, a memory, a radio, and energy source as depicted. This technology has a great potential potential to overcome existing difficulties of ITS. Existing solution for vehicle detection are ultrasonic sensor, induction loop, video camera and infrared sensor. Ultrasonic sensors are much more expensive than the pulse models and therefore rarely used and and the main disadvantage of ultrasonic system is that its performance is affected by temperature change and air turbulence. An inductive loop provides a high accuracy, but the installation requires road cutting and maintenance is also high. Video camera can provide good visualization, but it is expensive and requires complicated video process. The infrared sensor is sensitive to environment noise. On the contrary, the magnetic sensor shows more immune towards environment noise, such as rain, fog and wind thann any other sensor and in addition the magnetic sensor has the advantage of easy installation and maintenance. Fig. 1: Block diagram of simple vehicle detection and monitoring system *Corresponding Author www.ijesr.org 1248 T. Lakshmi Deepak et.al.,/ International Journal of Engineering & Science Research In this paper, the focus is to build a traffic and road monitoring system for intelligent route planning, road usage and protection that fulfil the constraints. The system should work under diverse road condition, dense and shapeless traffic and a large diversity of vehicles. The vehicle detection based on WSN using An isometric Magnetic Resistive (AMR) sensor. The aim is to provide a price effect, easy to organize, robust, flexible and less maintenance wireless solution for vehicle detection and classification different classes of vehicles. The main contributions of this paper are as follows. a) A WSN based system architecture for identification of the vehicles using the magnetic sensors and collecting the data. b) Classification of the vehicles based on the data collected by the sensors. 2. OVER VIEW OF DETECTION SYSTEM AND CLASSIFICATION A Wireless Sensor Network (WSN) is a network of small sensor nodes (SN) communicating among themselves using wireless communication, to sense the physical world. The WSN composed of three kinds of nodes including detection nodes, routing nodes and a base station node and these nodes can be freely and dynamically organized into an ad-hoc wireless network. In the traffic application, magnetic sensors are placed at known locations either on the pavement or on the road and it obtain the magnetic signature of vehicles travelling over the sensors. The signals are processed through a vehicle detection algorithm by the sensor nodes, detection events are then generated and transmitted to the access point and having collected the event data from these synchronized sensor nodes, the access point can then calculate the count of different types of vehicles, occupancy of the road and speed of the monitored traffic. Finally, this real time traffic information is passed to the Traffic Management Center (TMC) or to the local control unit for applications like traffic monitoring and signal control. Fig. 2: WSN for vehicle detection and monitoring Almost all vehicles have significant amounts of ferrous metals in their body (iron, steel, nickel, cobalt, etc.), the magnetic field disturbance created by a vehicle is sufficient to be detected by a magnetic sensor, which makes it a good enough for detecting vehicles. Potential vehicle detection applications that could use these magnetic sensors include traffic surveillance, railroad crossing control, parking lot space monitoring and automatic gate opening. Copyright © 2015 Published by IJESR. All rights reserved 1249 T. Lakshmi Deepak et.al.,/ International Journal of Engineering & Science Research Fig. 3: The disturbance of Earth’s magnetic flux lines by a vehicle The above picture gives a pictorial representation of the disturbance of the magnetic flux lines when the Earth's magnetic field penetrates a vehicle. The earth’s magnetic field strength is roughly equal to a half-gauss in magnetic flux density, so a low field and high sensitivity magnetic sensor is needed to measure the disturbance. The flux lines prefer to pass through the ferrous material of the vehicle. As the vehicle moves along, it is always accompany by a concentration of the flux lines known as its “magnetic shadow” There is condensed flux to the sides of the vehicle and amplified flux above and below it. An AMR sensor installed within the pavement detects the amplified flux below the vehicle. One type of these low field magnetic sensors is the magnetoresistive sensor, which is very suitable for use in a sensor node because of its small size. Magneto-resistive sensor can be further classified into Anisotropic Magneto-Resistive (AMR) and Giant Magneto-Resistive (GMR) types. An AMR sensor is directional, which means it only provides an amplitude response to the magnetic field along its sensitive axis, whereas a GMR sensor has little directionality. 2.1 AMR Sensor Anisotropic Magneto-Resistive (AMR) sensors as an upgrade from older and simpler vehicle detection systems. With the small size and simplicity of these wheat stone bridge based sensors, many applications are now able to deploy many of these sensors cost-effectively, and gain more information on nearby vehicles. Appealing to the fact that almost all road vehicles have significant amounts of ferrous metals in their chassis (iron, steel, nickel, cobalt, etc.), magnetic sensors are a good candidate for detecting vehicles. Today, most magnetic sensor technologies are fairly miniature in size, and thanks to solid state technology, both the size and the electrical interfacing have improved to make integration easier. In traffic application, it is essential to isolate one vehicle’s magnetic signal from the other vehicle signals from vehicles in different lanes and travel direction. So an AMR sensor is a much better choice for use in the sensor node. Basically, the AMR sensor is a Wheatstone bridge device as shown in Fig. It is made out of a nickel-iron (Perm alloy) thin-film deposited on a silicon wafer and patterned as a resistive strip element. In the presence of a magnetic field, a change in the bridge resistive elements causes a corresponding change in voltage across the bridge outputs. These resistive elements are aligned together to have a common sensitive axis that will provide positive voltage change with magnetic fields increasing in the sensitive direction. Fig. 4: AMR Sensor Bridge and vehicle detection system Copyright © 2015 Published by IJESR. All rights reserved 1250 T. Lakshmi Deepak et.al.,/ International Journal of Engineering & Science Research For AMR sensors, the sensor resistive elements are oriented as a resistive “wheat stone bridge” that varies resistance slightly as the magnetic field changes upon each element. The resistive elements are made of perm alloy thin films and have around 1000 ohms of resistance, but each element is precision matched to within an ohm of each other when no magnetic fields are present. Each bridge has four resistive elements with opposite elements being identical. For example, if the bridge receives a positive magnetic field or lines of magnetic flux in the sensitive axis, the Vb to Out+ and Out- to GND elements will slightly decrease in resistance while the other two elements will increase in resistance. The result will be that the voltage at Out+ increases above Vb/2 and the voltage at Out- decreases from Vb/2. The vehicle detection system consists of the two types of modules and they are master module and slave module. The slave module contains the sensor and transceiver for communicating with the master, buried along the length of the road with a distance between them and the master module contains transceiver for receiving data from the slave module. This master module and slave module communicate with each other using the radio channel. 2.2 Communication Ability of wireless communication is one of fundamental properties of the sensor node. Communication capability of WNS nodes are in most practical cases limited by communication range and data rate. It is important to remember that increasing of communication range requires increasing of transmit power/receiver sensitivity, i.e. energy consumption is increased as well. The communication protocol can be summarized as follows: 1. All the sensor and repeater nodes are configured with a pre-assigned radio channel and a communication time slot before installation. 2. A synchronization message is send by the AP periodically with a random back off time, so all the reachable nodes can catch this message eventually. 3. The radios of all the synchronized sensor nodes will wake up during its assigned time slot for two-way communication, and quickly switch back to sleep mode afterward. 4. The repeater nodes are configured with the time slots of a multiple number of sensor nodes, so that they can wake up at the right time to collect data from these sensor nodes and transmit the data back to the AP at their own time slots. 5. After data processing is done at the AP, useful traffic data are transmitted to the Traffic Management Center (TMC) for further traffic controls. The exclusion of multi-hop communication between sensor nodes greatly simplifies the protocol. The drawback is that the network coverage is limited by the maximum communication range between the AP and repeater nodes (e.g. 1000). For the implementation of a traffic surveillance station, this coverage is more than enough in most cases. It is claimed that a system lifetime of 10 years can be achieved for a typical vehicle detection application. Fig. 5: Communication links structure 2.3 Classification Vehicle classification refers to the process and methodology to classify a vehicle based on the data collected in a specific format into a pre-defined vehicle class. It is an important source of information for transportation design and management that can be used for many purposes. The distribution of vehicle types also provides valuable Copyright © 2015 Published by IJESR. All rights reserved 1251 T. Lakshmi Deepak et.al.,/ International Journal of Engineering & Science Research data analysis input to the prediction of highways capacity, assessment of the effectiveness of traffic legislation, automatic toll collection, weight enforcement strategies and environmental impact studies. Vision-based classification can achieve a correct classification rate. The major limitation of vision-based classification is that the system’s performance is greatly affected by the environmental and lighting conditions. On the other hand, vehicle classification by wireless sensor networks provides a much more flexible deployment configuration, making the system portable and, once again, scalable for large scale deployment. The magnetic sensor measures the Earth’s magnetic field, which has different inclination angles in different geographical locations. Since the Earth’s magnetic field is uniform over the surface in the scale of kilometre, vehicle signatures of the same vehicle measured at adjacent locations in the same travelling direction are expected to be the same. Since the transfer of information is very energy intensive, we try to minimize the volume of transmitted data during the WSN design stage. It is therefore important to realize essential part of the data processing directly at the point of its origin - in a sensor node. Now the data is collected at the Traffic Management Center (TMC), based on the essential part of the data collected by the sensors the vehicles are classified into different classes and the count of the each vehicle class is done based on the data i.e. each vehicle class has a specific sequence of numbers for identification. The data will be collected from the database and a graph is plotted in the Mat lab taking the x-axis as the type of vehicle class and the y-axis as the number of vehicles of each class. Fig. 6: Graph plotted from data collected by the sensor 3. CONCLUSION AND FUTURE PLAN The experiments done suggest that two magnetic sensors achieve good vehicle estimates. Further, a tri-axis magnetometer can classify classes of vehicles. The low-cost, ease of deployment and maintenance, and more detail information provided by these sensor networks, can give us a base for an accurate, extensive, and dense traffic system. The interesting thing about the wireless network is that the same communication and node architecture can be used to process and communicate measurements from different sensors, hence these are known as modular architecture. Conducting the experiment with more no of sensors and estimating the traffic and classifying different classes of vehicles and getting vehicle data from different sensors, storing the data received from different sensors and using that data for real-time traffic management. Copyright © 2015 Published by IJESR. All rights reserved 1252 T. Lakshmi Deepak et.al.,/ International Journal of Engineering & Science Research REFERENCES [1] Zhou B, Cao J, Zeng X, Wu H. Adaptive Traffic Light Control in Wireless Sensor Network based Intelligent Transportation System. Vehicular Technology Conference Fall (VTC 2010-Fall), 2010. [2] Nadeem T, Dashtinezhad S, Liao C, Iftode L. Traffic View: A Scalable Traffic Monitoring System. In Proceedings of the IEEE International Conference on Mobile Data Management (MDM 2004), Berkeley, GA, January 2004. [3] Collins K, Muntean G-M. A Vehicle Route Management Solution Enabled by Wireless Vehicular Networks. INFOCOM Workshops 2008. [4] Sharma A, Chaki R, Bhattacharya U. Applications of Wireless Sensor Network in Intelligent Traffic System: A Review. Electronics Computer Technology (ICECT), 3rd International Conference, 2011. [5] Tubaishat M, Shang Y, Shi H. Adaptive Traffic Light Control with Wireless Sensor Networks. Consumer Communications and Networking Conference, CCNC 2007. [6] Yousef KM, Al-Karaki JN, Shatnawi AM. Intelligent Traffic Light Flow Control System Using Wireless Sensors Networks. Journal of Information Science & Engineering 2010; 26: 753-768. [7] Salama AS, Saleh BK, Eassa MM. Intelligent Cross Road Traffic Management System (ICRTMS). 2nd International Conference on Computer Technology and Development (ICCTD 2010). [8] Xia R, Ye C, Zhang D. Vehicle to Vehicle and Roadside Sensor communication for Intelligent Navigation. Wireless Communications Networking and Mobile Computing (WiCOM), 6th International Conference, 2010. [9] Amine Ka Md, Challal Y, Djenouri D, Doudou M, Bouabdallah A, Badache N. A study of Wireless Sensor Networks for Urban Traffic Monitoring: Applications and Architectures. 3rd International Conference on Sustainable Energy Information Technology (SEIT-2013). [10] Jankuloska B, Zahariev M, Mateska A, Atanasovski V, Gavrilovska L. SRM: Traffic Regulations Monitoring Using VSNs. 17th Telecommunications forum TELFOR, Serbia, Belgrade, November 24-26, 2009. [11] Wenjie C, Lifeng C, Zhanglong C, Shiliang TU. A Realtime Dynamic Traffic Control System Based on Wireless Sensor Network. Proceedings of the International Conference on Parallel Processing Workshops (ICPPW), 2005. [12] Srinivasan D, Choy MC, Cheu RL. Neural networks for real-time traffic signal control. IEEE Transactions on Intelligent Transportation Systems 2006; 7(3): 261–272. [13] Murty RN, Mainland G, Rose I, Chowdhury AR, Gosain A, Bers J, Welsh M. CitySense: An Urban Scale Wireless Sensor Network and Test bed. Proceedings of the IEEE International Conference on Technologies for Homeland Security, Waltham, MA, May 2008. [14] Tachwali Y, Refai HH. System prototype for vehicle collision avoidance using wireless sensors embedded at intersections. Journal of the Franklin Institute 2009. [15] Cheung S, Coleri S, Varaiya P. Traffic Surveillance with Wireless Magnetic Sensors. University of California, Berkley, USA. Copyright © 2015 Published by IJESR. All rights reserved 1253