It is known that if traffic controls are implemented on the whole traffic network, transportation capacity could be maximized. The controls are greatly depended on the data from traffic surveillance systems, which have a high installation and maintenance costs. In view of this, we would like to offer a very attractive, low cost solution, which apply wireless sensor networks for traffic surveillance.

A traffic surveillance system requires four components: a sensor to catch the signals generated by vehicles, a processor to process the sensed data, a communication unit to transfer the processed data to the base station, and an energy source. Thanks to sensor technology, all of these components could now be integrated into a single tiny device.

Sensor nodes of this kind have been developed by the Department of EECS at Berkeley under a DARPA-sponsored project, named as Smart-Dust. The whole idea of this project is to apply these sensor nodes to form a wireless network for monitoring the traffic at a relative low cost.

Conventional inductive loop detectors are not suitable for large scale deployment as they are destructive, disruptive and having a high cost of installation and maintenance. Other above-ground systems are either having a high equipment cost or accuracy greatly depend on environment conditions. On the other hand, the purposed sensor network has the potential to revolutionize traffic surveillance technology because of its low cost and ease of deployment.

Besides detecting vehicles, it also has the potential to do vehicle classification and road conditions monitoring. Moreover, with its FM radio, RF-tag for ETC application and more advance cooperative controls with vehicle computers are also possible in the future. The sensor network would be very cost-efficient if we could turn it into a multi-functions traffic surveillance system.