Research On RFID Data Cleaning Method In Agricultural Products Logistics Management

As the core technology of the Internet of things,Radio Frequency Identification(RFID)technology has been widely used in every areas of all walks of life.In the meantime,the application of RFID technology unavoidably takes up lots of storage space and generates mass of data.Based on this,how to clean RFID data in a effective manner has aroused people's attention.This paper,by combing the correlated characteristics of RFID data in logistics management of agricultural products,studied the RFID data cleaning process and realized the corresponding multistage data cleaning algorithm.The main work involved was as follows:(1)Data pre-processing.Because of lots of time redundancy and a large number of RFID data generated from the process of RFID data in agricultural products logistics management,the paper carried out redundancy pre-processing on data generated at different time in the same place.(2)Establishment of data model.The paper made improvement on previous studies and established the pertinent data cleaning model based on the complete reading process of RFID object in agricultural products logistics management including a string of relevant information,such as number of object,travelling speed,reading cycle and reading rate.On this basis,the paper made detailed study and analysis against the transition of label.(3)Realization of label transition adaptive and dynamic sliding window.The paper realized effective the control against reading cycle of RFID reader according to transition of label and reading rate,so as to enhance and gradually restore to the initial reading static label level.As a result,it enabled RFID reader to not only reduce the data generation,but also guarantee the reading rate and prevent omission of data reading.After that,the paper made clustering analysis on RFID data and cleaned RFID object data.(4)Experimental analysis.With the help of established RFID object model from agricultural products logistics management,testing data were generated automatically through simulation,and it can be found that the cleaning of object data mitigated data generation and realized the purpose of data cleaning.Experimental results showed that the data cleaning model proposed by the paper had an ideal effect on RFID data involved in agricultural products logistics management,and such data cleaning model could lower data generation and increase RFID original data recognition accuracy.On top of that,the cleaning algorithm applied herein was practical to some extent.