A data-driven study on the sample size of cellular handoff probe system

Abstract

The popularity of big data from cell phone communication records makes cellular handoff probe system become an optimal choice for the collection of traffic speeds, especially for the large-scale freeway network. As a probe-based technique, the issues related to sample size are vital for the accuracy of cellular handoff probe system. Based on the current findings from simulation and field test, this study made an extensive discussion about the sample size problems using the field data. The subsequent investigation began with the correlation analysis between the sample size and the potentially influencing factors (the length of handoff link and traffic volume). Furthermore, a Weibull distribution model was set up to fit sample size for different handoff links. And then, the impact of sample size on the accuracy of cellular handoff probe system was following. Finally, based on the above explorations, several concluding remarks were presented as a reference for determining appropriate sample size to achieve high precision.

A Data-Driven Feature Based Learning Application to Detect Freeway Segment Traffic Status Using Mobile Phone Data

“…The whole mobile phone dataset includes both network signals, such as location updates (LU) and handover (HO), and records for billing, such as phone call started, phone call ended, text messaging, and data service. Previous studies have usually focused on estimating traffic status using HO data [13,14,18,20]. In the cellular data of this study, raw HO records were only 12.7% of the total dataset.…”

“…Besides the problem of noise, the sample size of HO data is another essential attribute that affects the performance of a phone-based system, much the same as other data sources [20]. Regarding the mechanism of HO data generation, when a phone is on a call and simultaneously passes the boundary of two cells, the HO data is only a small portion of the whole mobile phone dataset.…”

Intersection Congestion Analysis Based on Cellular Activity Data

“…Meanwhile, researchers realized that cellular probe is another promising technique to collect real-time data for traffic status detection with a higher penetration rate and less cost of equipment installation and maintenance, especially on arterials in urban areas [16]. Cellular networks provide mobility-related events that can be obtained during conventional operations, such as Location Area (LA) update, Route Area (RA) update, and cell update (handoff) [17].…”

“…Cellular networks provide mobility-related events that can be obtained during conventional operations, such as Location Area (LA) update, Route Area (RA) update, and cell update (handoff) [17]. The results of the CPTIAL-ITS Test and Demonstration Program showed positive potentials for using cellular data to estimate traffic states [16]. After that, many research results reported that cellular probe was a feasible technology to collect traffic data [18]- [21].…”

A Progressive Extended Kalman Filter Method for Freeway Traffic State Estimation Integrating Multisource Data

“…The wireless communication records are also known as the cellular communication records, including the records of handoff, normal location update, and data transition. The traffic features extracted from these wireless communication records are different from the traffic information from the handoff-based collection method [21]. Our recent work shows that the application of these traffic features in the traffic estimation is feasible [1].…”