电信科学 ›› 2016, Vol. 32 ›› Issue (2): 55-59.doi: 10.3969/j.issn.1000-0801.2016.02.008

• 研究与开发 • 上一篇    下一篇

基于自回归模型与神经网络模型的车流量预测对比

张婷婷1,2,张武雄1,3,裴冬1,2,赵铖1,2,俞涵1,2   

  1. 1 中国科学院上海微系统与信息技术研究所，上海 200050
    2 上海科技大学，上海201210
    3 上海无线通信研究中心，上海 201210
  • 发布日期:2017-02-03
  • 基金资助:
    国家自然科学基金资助项目;上海市自然科学基金资助项目

Comparison of traffic flow prediction based on AR model and BP model

Tingting ZHANG1,2,Wuxiong ZHANG1,3,Dong PEI1,2,Cheng ZHAO1,2,Han YU1,2   

  1. 1 Shanghai Institute of Microsystem and Information Technology，Chinese Academy of Science，Shanghai 200050，China
    2 Shanghai Tech University，Shanghai 201210，China
    3 Shanghai Research Center for Wireless Communications，Shanghai 201210，China
  • Published:2017-02-03
  • Supported by:
    The National Natural Science Foundation of China;Shanghai Natural Science Foundation

在线阅读

15

PDF下载

821

被引次数

6

摘要：

车流量建模是车联网（vehicular Ad Hoc network，VANET）路由、多媒体接入协议、无线算法设计的基础。准确的车流量模型将对智能交通系统（intelligent transportation system，ITS）实时调度和车联网的信息安全起到十分重要的作用。基于上海市的交通流量数据，利用自回归（auto regressive，AR）模型与神经（back-propagation，BP）网络模型对车流量实测数据进行了仿真对比，给出了相应的预测结果。研究发现，两个模型均能有效地对数据进行跟踪与预测，但对不同时段数据预测的准确性有所不同。研究结果将为未来智能交通应用、车联网的理论研究等提供有力依据。

关键词: 车联网, 智能交通, 自回归模型, 神经网络模型, 交通流量预测

Abstract:

Traffic flow modeling plays an important role in routing，MAC algorithm and protocol designs in vehicular Ad Hoc networks （VANET）. An accurate traffic flow model is crucial to traffic management of an intelligent transportation system （ITS）and information safety in a VANET. Based on Shanghai’s traffic flow data，the performance of the two different models was compared using auto regressive（AR）model and back-propagation（BP） network model，and the corresponding prediction result was given. Research finds that both of the two models can efficiently predict the traffic data，but they have different prediction accuracy for the data of different periods. The research result will provide support for future research on ITS and VANET.

Key words: VANET, ITS, AR model, BP network model, traffic flow predicting

No Suggested Reading articles found!