Traffic surveillance systems provide the data used by Intelligent Transportation Systems (ITS). The disadvantages of inductive loop detectors have led to the search for a reliable and cost-effective alternative system. This report summarizes a three-year research project in the prototype design, analysis and performance of wireless sensor networks for traffic surveillance, using both acoustic and magnetic sensors.

A robust real-time vehicle detection algorithm for both signals is developed. Magnetic sensors turned out to be superior, achieving detection rates above 97% in the field, and led to the abandonment of further research using acoustic sensors.

Vehicle classification and reidentification schemes for low-cost, low-power platforms with very limited computation resources are developed and tested. The vehicle classification algorithms require orders of magnitude fewer computation resources while achieving correct classification rates comparable to the best of all published vehicle classification schemes in tests with a large database, including 800 trucks. The algorithm for vehicle reidentification is tested on a limited left-turn reidentification experiment. The result is encouraging, but much more work is needed.

The flexibility, easy of installation, remote maintenance, low cost and high accuracy of wireless sensor networks will lead to their ubiquitous deployment and thereby provide the fine-grained vehicle detection required to implement effective traffic monitoring and control.

Wireless sensor networks are ‘future proof’. Additional modalities, such as temperature, moisture, and pollutant sensing, can be incorporated in the same node or in separate nodes to monitor other aspects of the traffic system. The wireless communication network can be used to communicate with vehicles to provide another path to ‘vehicle-infrastructure integration’.

This report presents the algorithm development and experimental work of the sensor node signal processing for vehicle detection. The signals used for vehicle detection are acoustic and magnetic signals. The acoustic signals are characterized by short time FFT analysis and two acoustic vehicle detection algorithms are proposed: the Adaptive Threshold algorithm (ATA) and the Min-max algorithm (MMA). The ATA detects vehicle by searching for a sequence of 1's after slicing the acoustic energy curve using an adaptive threshold. The MMA detects vehicles by searching the local maximum in the acoustic energy curve. Real time tests and offline simulations demonstrate the effectiveness of the two algorithms. For magnetic signals, a simple threshold slicing algorithm is utilized and real time tests give good performance. Finally, FPGA implementation of ATA is also presented for power efficiency requirement and the implementation justifies the use of dedicated hardware for low power implementation.

Wireless magnetic sensor networks offer a very attractive, low-cost alternative to inductive loops for traffic measurement in freeways and at intersections. In addition to vehicle count, occupancy and speed, the sensors yield traffic information (such as vehicle classification) that cannot be obtained from loop data. Because such networks can be deployed in a very short time, they can also be used (and reused) for temporary traffic measurement. This paper reports the detection capabilities of magnetic sensors, based on two field experiments. The first experiment collected a two-hour trace of measurements on Hearst Avenue in Berkeley. The vehicle detection rate is better than 99 percent (100 percent for vehicles other than motorcycles); and estimates of vehicle length and speed appear to be better than 90 percent. Moreover, the measurements also give inter-vehicle spacing or headways, which reveal such interesting phenomena as platoon formation downstream of a traffic signal. Results of the second experiment are preliminary. Sensor data from 37 passing vehicles at the same site are processed and classified into 6 types. Sixty percent of the vehicles are classified correctly, when length is not used as a feature. The classification algorithm can be implemented in real time by the sensor node itself, in contrast to other methods based on high scan-rate inductive loop signals, which require extensive offline computation. We believe that when length is used as a feature, 80-90 percent of vehicles will be correctly classified.