A Measurement Framework for Explicit and Implicit Urban Traffic Sensing

Abstract

Urban traffic sensing has been investigated extensively by different real-time sensing approaches due to important applications such as navigation and emergency services. Basically, the existing traffic sensing approaches can be classified into two categories by sensing natures, i.e., explicit and implicit sensing. In this article, we design a measurement framework called EXIMIUS for a large-scale data-driven study to investigate the strengths and weaknesses of two sensing approaches by using two particular systems for traffic sensing as concrete examples. In our investigation, we utilize TB-level data from two systems: (i) GPS data from five thousand vehicles, (ii) signaling data from three million cellphone users, from the Chinese city Hefei. Our study adopts a widely used concept called crowdedness level to rigorously explore the impacts of contexts on traffic conditions including population density, region functions, road categories, rush hours, holidays, weather, and so on, based on various context data. We quantify the strengths and weaknesses of these two sensing approaches in different scenarios and then we explore the possibility of unifying two sensing approaches for better performance by using a truth discovery-based data fusion scheme. Our results provide a few valuable insights for urban sensing based on explicit and implicit data from transportation and telecommunication domains.

BuildSenSys: Reusing Building Sensing Data for Traffic Prediction With Cross-Domain Learning

“…However, with the inherent biases and random data deficiency, most trajectory data cannot cover the entire traffic dynamics. For instance, GPS data of a 6,000-taxi network can only cover 28% of the overall road segments in a large city with distinctive operating time [18]. With such data sparsity, trajectory sensing data cannot guarantee the sustainability and reliability in traffic prediction, especially on explicitly targeted road segments.…”

“…Compared with the existing methods [11], [14]- [18] of predicting nearby traffic volume, reusing building data has considerable merits of both low cost and high reliability, which are crucial to traffic sensing [13]. Most of the conventional traffic prediction methods heavily rely on the fixed road-based traffic sensing systems, such as loop detectors [14] and traffic surveillance cameras [15].…”

“…On the contrary, re-using off-the-shelf building sensing data will significantly reduce the cost, as with no extra deployment and maintenance [8], [11]. In terms of opportunistic sensing data of floating vehicles, GPS trajectories [11] and cellular records of passengers [18] are recently exploited to infer traffic volume to address the highcost issue. However, opportunistic sensing suffers from the uncontrollable property of users and data deficiency of particular roads, resulting in unreliable performance in traffic volume prediction [18].…”

“…In terms of opportunistic sensing data of floating vehicles, GPS trajectories [11] and cellular records of passengers [18] are recently exploited to infer traffic volume to address the highcost issue. However, opportunistic sensing suffers from the uncontrollable property of users and data deficiency of particular roads, resulting in unreliable performance in traffic volume prediction [18]. Buildings and their surrounding roads, as two basic constructions in urban infrastructure, are both stationary with long-term spatial relations.…”