Distributed Acoustic Sensing for Vehicle Speed and Traffic Flow Estimation

For traffic monitoring systems it is desirable to have a low-cost, low-maintenance, and reliable system to facilitate continuous and safe traffic flow. In line with this, we introduce distributed acoustic sensing (DAS) for traffic monitoring applications. In DAS systems, series of light pulses are transmitted along a fiber-optic cable and the back-scattered light, which is affected by the mechanical strain of the fiber-optic cables due to ground vibrations, is measured and analyzed. With fiber-optic cables installed parallel to a highway where ground vibrations are induced by passing vehicles, traffic information such as vehicle flow and average speeds can be estimated from the DAS signals. In this paper, we present an algorithm based on image processing methods that estimates traffic information from image representations generated from 1-minute segments of DAS signals. We tested the results of the algorithm on DAS data recording on a real highway against reference traffic data measured by road-side sensors.