首页 | 本学科首页   官方微博 | 高级检索  
     

基于HMM的移动预测模型与仿真
引用本文:沈波,刘云. 基于HMM的移动预测模型与仿真[J]. 系统仿真学报, 2007, 19(18): 4118-4121
作者姓名:沈波  刘云
作者单位:1. 北京交通大学轨道交通控制与安全国家重点实验室,北京,100044;北京交通大学电子信息学院,北京,100044
2. 北京交通大学轨道交通控制与安全国家重点实验室,北京,100044
摘    要:提出了一种基于隐马尔柯夫模型的移动预测模型,并给出增强模型预测能力和提高预测精度的方法。该模型用于预测移动IP网络中移动节点的运动方向和将要连接的接入路由器,为避免或减小由于移动造成的通信中断和时延赢得准备时间。仿真结果说明,模型的预测准确率较高,在适当选择状态数的条件下,模型对移动的随机性具有较好的适应能力。

关 键 词:移动预测  隐马尔可夫模型  隐状态  观察序列
文章编号:1004-731X(2007)18-4118-04
收稿时间:2006-11-13
修稿时间:2007-01-05

Movement Prediction Model Based on HMM and Simulations
SHEN Bo,LIU Yun. Movement Prediction Model Based on HMM and Simulations[J]. Journal of System Simulation, 2007, 19(18): 4118-4121
Authors:SHEN Bo  LIU Yun
Affiliation:1 .State Key Laboratory of Rail Traffic Control and Safety, Beijing Jiaotong University, Beijing 100044, China; 2.College of Electronics and Information Engineering, Beijing Jiaotong University, Beijing 100044, China
Abstract:A movement prediction model based on HMM and methods that could enhance predicting ability and accuracy were proposed. In order to win the time for avoiding or decreasing the communication interrupt and latency caused by movement,the model aims to predict the movement direction of the mobile node in mobile IP network and the access router which the mobile node is going to attach to. The simulation results indicate that the prediction exactness rate of the model is higher and the model has better adapting ability to the randomicity of movement under the condition of proper number of states.
Keywords:movement prediction  hidden markov model  hidden state  observing sequence
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号