City Cellular Traffic Map (C2TM)
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
LICENSE Initial commit Oct 9, 2014
README.md Update README.md Oct 5, 2016
heatmap_animation.gif Update README Jan 27, 2015
traceset.zip Add traceset May 18, 2015

README.md

City Cellular Traffic Map (C2TM)

This repository gives detailed description about city-cellular-traffic-map dataset. You can access the CRAWDAD repository from Dartmouth to download the data directly. Please cite our paper below if this data contributes to you research conclusions.

Introduction

Traffic characteristics over space and time constitute an important aspect of cellular networks in consideration of resource provision, traffic engineering and system optimization. Despite recent progress in revealing temporal dynamics and spatial inhomogeneity of cellular traffic, limited knowledge about traffic dependence is gained. One of challenges comes from the absence of sustained observations at a network-wide scale. To complement this gap, we make an analysis on week-long traffic generated by a large population of people in a median-size city of China. Here we contribute the basic dimensions of this data to public communities to stimulate more research enthusiasm on this topic.

Data Collection and Preprocessing

Our analyses make use of request-response records extracted from HTTP traffic at the city scale, consisting of individuals’ activities during a continuous week (actually eight days from Aug. 19 to Aug. 26, 2012), with accurate timestamp and location information indicated by connected cellular base stations (BS). Each individual is detected by the hashed International Mobile Subscriber Identity (IMSI).

To remain principle traffic characteristics and preserve user privacy from our analysis simultaneously, we derive hourly statistics at base-station granularity, which implies a maximum of N x M records (N the number of base stations, M the number of recording hours) in this public version. Meanwhile, under requests of our data provider, the real location information of network infrastructure is also meshed, whereas the relative topology of underlying network is remained for study of network optimization or more.

Basic Dimensions

This public cellular traffic map contains two files, i.e., traffic and topology. The former provides hourly traffic statistics for each base station, while the latter stores the relative topology of underlying cellular network. The time dimension is represented by Gregorian calendar, with the starting point of each hour to denote following 60 minutes (Note: GMT+8 time zone is assumed in the data). The relative location of base station is in longitude/latitude form to facilitate some analysis with standard geographic processing about great circle distance.

  • Traffic trace file (1625680 rows, 5 columns)
  • BS: identity of each cellular base station in this public data.
  • Time_hour: hourly timestamp in UNIX epoch time (time zone GMT+8).
  • Users: the number of active users associated with specific base station and hour.
  • Packets: the number of transferd packets associated with specific base station and hour.
  • Bytes: the number of transferd bytes associated with specific base station and hour.
  • Topology file (13296 rows, 3 columns)
  • BS: identity of each cellular base station in this public data.
  • Lon: relative longitude of given base station.
  • Lat: relative latitude of given base station.

Visualization

To show basic dynamics of this data, we attach a heatmap animation (plotted by R) below for the urban area of measured region. The packets column in traffic file is used to estimate the traffic density over space, while a normalization over space and time dimensions is involved to show up the dynamics more obviously.

traffic_dynamic_heatmap

Download

You can access the CRAWDAD repository from Dartmouth, or download from this repo. directly. Please cite our paper below if this data contributes to you research conclusions.

Citation

@INPROCEEDINGS{Modeling15-Chen,
    author={Xiaming Chen, Yaohui Jin, Siwei Qiang, Weisheng Hu, Kaida Jiang},
    booktitle={Communications (ICC), 2015 IEEE International Conference on},
    title={Analyzing and Modeling Spatio-Temporal Dependence of Cellular Traffic at City Scale},
    year={2015}
}

Copyright

The original copyright of this dataset is reserved by © OMNILab, Shanghai Jiao Tong University, and the cooperative organizations. This data is made public under GPLv3 license and encouraged to be used in researches and academic applications.