A Research of RFID Supply Chain Data Anomaly Detection in EPC Network
|School||Shanghai Jiaotong University|
|Keywords||RFID Anomaly Detection Supply Chain EPC Network|
With RFID technology are more and more concerned about the deployment of RFID systems gradually to large-scale development of networked , distributed trend , the reliability analysis of RFID data becomes particularly important . According to the survey showed that the competitiveness of enterprises is closely related to their ability to control information . Data analysis in order to obtain useful information on the analysis of the premise that the \Anomaly Detection technology for RFID supply chain in-depth analysis and research , RFID supply chain anomaly detection system based on distance and the rules proposed for RFID supply chain transport delayed , theft and other abnormal problem , the system is divided into a total of three levels: data preprocessing layer , the exception analysis layer and graphical display layer . The data pretreatment layer is divided into time preprocessing and path pretreatment , mainly for cleaning, for the upper layer to provide clean, useful data acquired RFID event data from the EPC network . The anomaly analysis layer is the core of the system is divided into distance-based anomaly detection and rule-based anomaly detection . Comparing the path sequence of the discrete Fourier transform structural similarity distance - based method of detecting abnormality discovered abnormalities, rules-based anomaly detection method to use the node or between nodes analysis rule to detect the efficiency of the flow of the supply chain and abnormal conditions , the two were to detect the abnormal condition of the supply chain from both global and local . Among them, the rule-based approach is divided into traffic statistics , speed, consistency , consistency of the residence time of transportation of receiving matching rules . Graphical display layer abnormal data stream by way of a map-based display provides a visual interface to facilitate managers view exception information . This paper first analyzes the challenges faced by anomaly detection technology and the characteristics of RFID data ; RFID supply chain based on distance and rule-based data anomaly detection solutions and supply chain exception type and existing anomaly detection method ; Finally its achieve.