A Wireless Sensor Network for Smart Roadbeds and Intelligent Transportation Systems

We have developed a wireless sensor package to instrument roadways for Intelligent Transportation Systems. The sensor package counts passing vehicles, measures the average roadway speed, and detects ice and water on the road. Clusters of sensors can transmit this information in near real-time to wired base stations for use controlling and predicting traffic, and in clearing road hazards. The sensor package draws a maximum time-averaged current of 17 tA from an internal lithium battery, allowing it to operate in the roadbed for at least 10 years without maintenance. The nodes cost well under $30 to manufacture, and can be installed without running wires under the road, facilitating wide deployment. Unlike many other types of traffic sensors, these sensors count vehicles in bumper-to-bumper traffic just as well as in widely separated traffic. The devices detect vehicles by detecting the perturbations in the Earth's magnetic field caused by the vehicles. They measure this perturbation using an anisotropic magnetoresistive magnetic field sensor. The radio transmitters in the sensor are frequency-agile, and the sensors use a randomized sparse TDMA protocol, which allows several transmit-only devices to share a channel. The sensor package includes a custom-designed, compact, broadband, inexpensive printed circuit microstrip antenna for the 915 MHz U.S. ISM band. We built a prototype sensor package, and installed it in a pothole in a city street. We used the sensor to monitor the traffic flow rate during free-flowing traffic and a traffic jam. Thesis Supervisor: Joseph Paradiso Title: Principal Research Scientist, MIT Media Lab