One of the exciting emerging uses of DSRC/WAVE technology is the ability to monitor real-time road traffic conditions with high resolution, using beacons transmitted by individual vehicles, and make informed traffic control decisions such as traffic light timing or route advice. However, previous studies have shown that achieving a high level of accuracy in traffic density estimation requires very frequent beacon transmissions as well as a high adoption rate of the technology, which raises a scalability problem in dense urban settings and effectively requires a dedicated radio transceiver, precluding the wireless channel from being used for any other purpose at the same time. In this paper, we propose an approach that allows the wireless channel load due to beacon transmissions to be significantly reduced while retaining very low traffic estimation error levels, by using tools from traditional traffic theory (such as the Cell Transmission Model, CTM) to analyze position and speed signals from infrequent wireless beacons and predict the dynamics of the traffic behavior in between. Our approach is evaluated in a typical urban scenario consisting of a signalized intersection of multiple-lane roads, leading to new insights about how the value of the information in vehicles' beacons depends strongly on their location with respect to the intersection.