«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

27

期数:

2017年01期

页码:

139-142

栏目:

应用开发研究

出版日期:

2017-01-10

文章信息/Info

Title:

 Detection of Traffic Congestion Based on GPS Data

文章编号:

1673-629X（2017)01-0139-04

作者:

 张亚茹; 赵海涛; 刘南杰; 黄波; 李大鹏

 南京邮电大学 通信与信息工程学院

Author(s):

 ZHANG Ya-ru; ZHAO Hai-tao; LIU Nan-jie; HUANG Bo; LI Da-peng

关键词:

 ; GPS; 交通拥堵; 预处理; k -means; 抽样

Keywords:

 ; GPS; traffic congestion; preprocessing; k-means; sampling

分类号:

TP301

文献标志码:

A

摘要:

 随着交通拥堵问题的日益突出，如何有效检测区域内实时的交通状态是城市交通发展中的研究热点。为了实现对区域内交通状态进行有效的检测，提出了一种基于多源GPS数据交通拥堵的检测方法。首先，剔除原始GPS数据中的异常数据并且对缺失的数据进行修复，以保证GPS数据的合理性，对采集的原始GPS数据进行了预处理；其次，在保证采集的交通信息完整性的基础上，利用GPS数据的时空特性，使用k -means算法对GPS数据进行聚类分析，实现了不同簇下的研究区域的划分；最后，为获取每个簇的实时交通状态，采用了一种基于簇内样本数量进行随机抽样的交通状态检测方法。仿真结果表明，所提算法可以实现对区域内交通状态的有效检测且算法时效性得到了有效提升。

Abstract:

 With traffic congestion becoming increasingly serious,how to effectively detect the real-time traffic status has become a re-search hotspot in the field of urban transportation. For effective detection of traffic area,a detection method based on multi-source GPS data is proposed. Firstly,in order to eliminate abnormal data and ensure the rationality of data,the method makes a preprocessing of GPS data to eliminate abnormal data and repair missing data. Secondly,on the basis of ensuring the integrity of data,the method uses k-means algorithm to cluster GPS data in accordance with the temporal-spatial characteristics of GPS data,and then divides the study area into dif-ferent clusters. Finally,for obtaining the real-time traffic conditions of each cluster,a sampling algorithm based on the number of samples within each cluster is used to detect the traffic conditions. Simulation results show that the proposed algorithm with simple process can ef-fectively detect the state of traffic within the region.

常用功能

更新日期/Last Update: 2017-04-01