Research on traffic information collection system based on wireless sensor network

0 Introduction
Urban traffic problems are universal problems in all countries of the world. Intelligent transportation systems are an effective way to solve the increasingly serious urban traffic problems. Traffic information collection systems are key subsystems of intelligent transportation systems, the foundation for the development of intelligent transportation systems and the premise of intelligent traffic management.
At present, common methods of traffic information collection include induction coils, infrared, and video. Among them, induction coils are intrusive detection, which requires a large area of ​​road surface to be chiseled for installation, affecting the service life of the highway. In addition, the road surface needs to be dug up again when repairing and changing applications, which causes great damage to the road surface. Infrared, video, etc. belong to non-intrusive detection. They do not need to dig up the ground for installation, but a large number of lines still need to be laid, and the detection results are greatly affected by strong winds, rain, snow and other weather conditions. Using wireless sensor networks (WSN) to collect traffic information can reduce the laying of lines, cause very little damage to the road surface, and the measurement results are basically not affected by the weather, which greatly facilitates installation and maintenance work.

1 System Design
The traffic information collection system based on wireless sensor network mainly consists of three parts: sensor network, communication control server and user end, as shown in Figure 1. The wireless sensor network is responsible for collecting traffic information on the road surface and transmitting the data to the communication control server through Ethernet; the communication control server and the user end adopt B/S mode. This structure combines the advantages of asp.net in Web applications, and can realize the functions of remote users using browsers to monitor, analyze and process sensor node data online, effectively organize and manage the information collected in the monitoring area of ​​the sensor network, and respond to user settings and queries.

1.1 Wireless sensor network structure
The wireless sensor network in the system is implemented using TI's CC2520 chip. CC2520 is a second-generation ZigBee/IEEE 802.15.4 wireless RF transceiver dedicated to the 2.4 GHz unlicensed ISM band, with excellent characteristics such as high stability and low operating voltage. The sensor network uses different network nodes to form a cluster-tree network. The network topology is shown in Figure 2. The network nodes include aggregation nodes, routers, and collection nodes. The collection nodes collect traffic information data and send it directly or via routers to the aggregation nodes.

[page]

The collection node detects traffic information by measuring the geomagnetism. The geomagnetic field value is small, about 0.5×10T. It can be considered that the strength of the earth's magnetic field is constant in a certain area (about several thousand meters). When a magnetic object passes by, it will cause the change of the magnetic field strength around the object. The car can be regarded as a model composed of multiple bipolar magnets, causing disturbances in the earth's magnetic field. The interference of the car on the geomagnetism is shown in Figure 3.

Therefore, geomagnetic sensors can be used to detect changes in the magnetic field when vehicles pass through to obtain information about traffic flow. This article uses Honeywell's magnetoresistive sensor HMC1052, which is based on the principle of magnetoresistance effect. Magnetoresistance effect refers to the phenomenon that the resistance value of certain metals or semiconductors changes with the change of the external magnetic field. That is, when the external magnetic field changes, the resistance inside the sensor will also change accordingly, converting the magnetic field strength into voltage output. Using this effect, HMC1052 can measure geomagnetism and the disturbance of geomagnetism when cars pass through. The
collection node is installed on the road section that needs to be measured, responsible for collecting traffic information on the road surface, and sending the data directly or through a router to the aggregation node.
The routing node is installed on the roadside as close as possible to the collection node, used to forward the data of the collection node to the aggregation node. Multi-hop routing can be used according to the requirements of the transmission distance. At the same time, it also enhances the robustness of the network.
The aggregation node is also installed on the roadside. It is responsible for starting and establishing the wireless network, and also establishes network security mechanisms, binding in the network, etc., and receives data from the collection node and the routing node. The aggregation node also has an Ethernet interface, which sends the received data to the communication control server through Ethernet.
1.2 Communication Control Server
The system separates the management of data in the wireless sensor network from the use of users. The communication control server is mainly responsible for data management such as analysis, calculation and storage of traffic flow data. The communication control server collects data from the aggregation node through the Ethernet interface. Each frame of the received traffic information data contains a packet header, ID number, collected data, check bit and packet tail. The data structure is shown in Table 1.

The communication control server first parses and repackages the original data and stores it in the database. By abstracting this information, the overall logical structure of the database is formed. The database entity relationship diagram (ER diagram) is shown in Figure 4. According to the entity relationship diagram, it is converted into the relational model (data table) in SQL Server 2005 as shown in Tables 2 to 4.

The SQL Server 2005 database is used. It uses a graphical user interface, intuitive database management, and supports Web technology. It is easy to publish the data in the database to the Web page, which is very suitable for B/S architecture. The database is accessed through ADO. NET (ActiveX Data Object) to implement user operations such as storage, query and deletion of the database. ADO. NET is the preferred data access interface in the . NET programming environment. It provides platform interoperability and scalable data access. It is easy to access the database using a few objects, such as the SqlConnection object to manage the connection with the data source. The SqlCommand object communicates with the data source and sends commands. Use SqlDataReader to quickly read data "forward", and use DataSet or SqlDataAdapter to write and read data sources. [page]

1.3 User-side design
The user-side provides a visual interface for users to interact with the database through the browser, responsible for data query, graphical display and monitoring network operation. The Web page running on the user-side browser is developed based on the .net framework, and the development environment is Visual Studio 2005. It combines the advantages of asp.net in Web applications and uses multi-threading and modularization ideas to complete the design of the entire interface. Users only need to care about the data they need, without worrying about the specific implementation details of the sensor network. Thus, the remote terminal browser can realize the function of online monitoring, analysis and processing of sensor node data. The main functional modules include real-time traffic parameter monitoring, historical data playback statistics, sensor network management, network user management, alarm notification and other message processing. The functional modules of the user side are shown in Figure 5.

Network user management module Assign different permissions to users of different levels. Ordinary users do not have sensor network management and user management permissions, but can query and view data in the database. Administrator users can add users, delete users, grant permissions, and allow users to change passwords.
Real-time traffic flow monitoring module Receives field data sent from the database and plots it in the form of real-time curves, so that users can intuitively observe the information of each collection node and thus grasp the traffic flow in the monitoring area.
Historical data playback statistics module The historical data module allows users to select a historical time range and query the data uploaded by each collection node. It can also analyze historical data and count the daily and weekly traffic flow at intersections.
Sensor network management module Allows users to rename, add, and delete device nodes, display the topological structure in the network in real time, and display the routing status and link information of the nodes to grasp the overall status of network operation. This module also provides configuration and management of wireless sensor network nodes. Through the management of these nodes, users can timely understand the distribution of nodes in the monitoring area.
Alarm notification and other message processing modules When the sensor network is abnormal, such as node damage, power failure, etc., some prompts, alarms, etc. will be generated. The system needs to manage these messages. On the one hand, these messages should be stored in the log for users to query in the future. On the other hand, these messages should be notified to users in a timely manner so that users can take appropriate measures to deal with them.
The user interface is shown in Figure 6.

2 Conclusion
This paper implements a traffic information collection system based on wireless sensor networks. The magnetoresistive sensor HMC1052 is used to detect vehicle information, and the wireless network transmits data, which is easy to install and maintain. The data management adopts the B/S mode to achieve remote monitoring, maintenance and upgrading. The sensor network is easy to manage and arrange, and the efficiency of wireless sensor network development and maintenance is improved. Multi-threading and modularization technology are used to realize data collection and processing, which improves the security, reliability and consistency of data. Long-term experiments have proved that the information collection system implemented is stable and reliable, and the design goals have been achieved.