不确定RFID数据流上基于熵的数据推导方法

针对海量RFID数据中存在的不准确性以及语义信息鸿沟,提出了一种基于概率分布熵的数据推导方法.该方法采用时变图模型并充分利用历史RFID识读,从不确定RFID数据流上有效捕获贴标对象所处的状态,并采用基于概率分布熵的方法分别推导对象最可能的位置和包含.该方法可以同时处理RFID识读中的漏读和多读.最后利用模拟RFID数据进行参数调优和算法评价,实验结果显示：该方法在获得准确推导结果的同时,能确保其高效性和高伸缩性.

Entropy-based data interpretation method over uncertain streaming RFID data

Owing to the uncertainties and the information gap in raw RFID(radio frequency identification) data,an entropy-based data interpretation method over uncertain RFID streams was proposed.Firstly a time-varying graph model supplemented with history RFID readings was employed to efficiently capture the states of tagged objects.Then an entropy-based method was used to estimate the most likely location and containment for each object.Both the false negatives and false positives could be handled with the proposed method.Finally,the parameters and evaluated the interpretation algorithm were turned with synthetic RFID data.Experimental results show that the method is efficient and scalable.