摘要
分析了移动路径预测的已有方案,尤其针对k阶Markov预测器中存在的状态空间膨胀以及知识更新较慢问题,提出了一种新的WLAN位置预测器模型,并在1200个实际WLAN用户的移动跟踪数据集上对Markov预测器和新预测器的预测精度进行了比较分析。分析结果表明,新的预测器模型比k阶Markov预测器中复杂度最低的一阶Markov预测器更节省空间和搜索时间,并且比预测效果最好的二阶Markov预测器有更好的预测精度和普适性。该模型以很小的代价获得更好的性能,具有较高的实用价值。
Some problems existing in the available themes, especially the state space expansion and knowledge aging problem in Order-k Markov predictors, were analyzed. A new predictor model was proposed. Prediction accuracy of the proposed predictor was compared with that of Order-k Markov predictors by using a trace of the mobility patterns of 1,200 users on real Wi-Fi wireless network. The new model＇s good practicability is shown by the results. It saves more space and time than Order-1 Markov predictor that is the simplest in Order-k Markov predictors, and outperforms Order-2 Markov predictor that has the best prediction performance.
出处
《计算机应用》
CSCD
北大核心
2006年第12期2813-2816,共4页
journal of Computer Applications
基金
国家自然科学基金资助项目(60573128)
公安部应用创新计划资助项目(2005YYCXJLST045)
