Traffic information acquisition is often implemented by video cameras or inductive loops, which is expensive or inconvenient from installation and maintenance perspectives. We designed and implemented a pervasive traffic information acquisition system based on wireless sensor networks called EasiTia. Unlike existing solutions, the implementation of the system does not require extra devices in the road infrastructure or vehicle, nor the excavation of the road surfaces. EasiTia can easily be deployed at roadsides. It is of low cost and resource efficient. Our contributions are given as follows: 1) To deal with low signal-to-noise ratios (SNRs) and stochastic disturbances in traffic information acquisition, we proposed and implemented a cross-correlation-based vehicle-detection algorithm. 2) To resolve the problems of data association, vehicle velocity calculation, and vehicle identification, we proposed a collaborative traffic information processing mechanism in the EasiTia system. Based on real road environment experimental analysis, we demonstrate that EasiTia is an applicable and cost-effective candidate for a pervasive traffic information acquisition system.