面向仓储的RFID数据清洗技术研究

摘要

Aiming at unreliability of original RFID data in warehouse, the redundancy and out-of-order event flows of RFID data in warehouse were studied to propose a warehouse-oriented RFID data cleaning model. Firstly, problems of RFID data in warehouse were described and an abstract scenario generated by redundant data and out-of-order event flows was set up. Secondly, combined with the RFID warehouse data cleaning model, the corresponding algorithms were proposed for cleaning redundant data and modifying out-of-order event flows and the im-provements of algorithm and detailed implementation procedures were introduced. Finally, by means of the reader inspection model, original RFID data flows were constructed to test the number of records and indicators of accuracy on the basis of different experiment parameters. The results indicate that the proposed data cleaning method can effectively remove redundant data.%针对仓储中无线射频识别(RFID)原始数据的不可靠性问题,对仓储中RFID数据冗余和事件流乱序问题进行了研究,提出了一种面向仓储的RFID数据清洗模型.首先对仓储中RFID数据存在的问题进行了描述,建立了冗余数据和事件流乱序问题产生的抽象场景;然后结合RFID仓储数据清洗模型提出了相应的冗余数据清洗和事件流乱序修正算法,并介绍了相应算法的改进之处及具体实现步骤;最后利用读写器检测模型构造了RFID原始数据流,通过不同实验参数对记录数量和正确率指标进行了实验测试.研究结果表明:提出的数据清洗方法可以有效去除冗余数据并提高RFID事件输出的正确率.