01697762

Component

Two traffic features (unique cellphone counts, UCC, and pseudo speed, PS) could be extracted from cellphone activity (CA) data which contains all wireless communication records of a standby or active cellphone. PS has been validated to have a close relation with traffic speed in the previous works. This study furtherly investigated into the representativeness of UCC to the traffic volume. Firstly, this paper proposed a dynamic time warping (DTW)-based method to analyze the characteristics of UCC, which included a DTW-based hierarchical clustering step and a DTW-based classification step. The case study indicated that this proposed analysis method could help identify a link-based characteristic of UCC, and efficiently classify the links into different groups based on the different representativeness of UCC to volume. And then, the analysis results were applied to the formalization of the relationship model between UCC and traffic volume. The case study showed that a general linear model could be calibrated for the relationship of data from the links classified in the same group and this model performed well under the uncongested traffic condition.

This paper was sponsored by TRB committee ABJ35 Standing Committee on Highway Traffic Monitoring.

19-03312

English

7p

2019

Transportation Research Board 98th Annual Meeting

Digital/other

Figures; References

Cellular telephones; Data analysis; Traffic data; Traffic flow

Data and Information Technology; Highways; Operations and Traffic Management

Transportation Research Board Annual Meeting 2019 Paper #19-03312

TRIS, TRB, ATRI

Dec 7 2018 9:36AM