Cellular Wireless Mesh Sensor Network for Comprehensive Spatial Traffic Movement Detection and Data Fusion: Phase 2 - FY08 NATSRL

Principal Investigator:

Taek Kwon, Professor, UMD-Electrical Engineering

Project Summary:

This project continued research begun in fiscal year 2006-07 under the same project title, with the overall research goal of developing a practical method for detailed monitoring of traffic movements using a mesh network of wireless sensor nodes. The motivation behind this approach was that low-power wireless nodes naturally form cell coverage areas that can be easily configured as a mesh network and used to detect the motion of vehicles in the coverage area. In the project?s first year, anisotropic magnetoresistive (AMR) sensor nodes were designed and integrated with commercially available 802.15.4 chip sets to form the basic wireless node. Basic node-to-node protocols were then developed and implemented. The objective of the study?s second year was to complete development of the remaining required protocols and software for the mesh network so that the network functions as a working prototype. The performance of the prototype was then evaluated for tracking vehicle movements in an intersection. The main protocols required included a congestion control protocol and a set of network and sensor management protocols. Upon completion of the protocol implementation, a vehicle-tracking algorithm based on the aggregated sensor data was developed. The completed system was installed in a live intersection to evaluate the performance of the network and to learn about the field installation and operational issues. The final report describes details on the system component design, implementation, experimental results, and analysis.


Project Details:

  • Start date: 06/2007
  • Project Status: Completed
  • Research Area: Transportation Safety and Traffic Flow
  • Topics: Traffic Modeling and Data