A Study Of RFID Data Warehouse Model Based On Composite Coding Approach

With improving in tag manufacturing process, the application level of RFID technology raised up. Currently, the RFID technology is applied to item-level in supply chain of some large companies, so that they can easily get the dynamic information of the goods movement. Followed, there are vast amounts of RFID data in Supply Chain Management System of the companies and it's difficult to effectively use this data.Based on analyzing the characteristics and query needs of RFID data , research comparing the existing RFID data management weaknesses; according to the need for RFID data query and Information integrity requirements, for RFID data generated in supply chain management, a RFID data warehouse model was proposed based on composite coding approach. The warehouse model consists of two parts: storage model for the enterprise to store RFID data on their central server and query model for providing query dataset for specific business units. Storage model is mainly to resolve the conflicts between data integrity and storage space. According to the structure of RFID data, the paper takes use of the composite coding approach to store the path information and time information in RFID data, to achieve the purpose of effective data compression in the context for ensuring data integrity Query model on the query server, is to provide the data set in the different level of abstraction for querying user to meet the different departments need of query RFID data. Query model use the division and key points of the transport map of of goods to produce movement tree of goods, and uses the coding method of most brothers to encode the movement tree and put all movement nodes into the path table. The code of the last location of the goods movement path in a path-tree as numerical code, together with the codes of all the locations that compose the path as the character encoding is stored in the EPC_Path table. It can improve query performance to make use of the composite coding approach that makes query with time-condition by using the remainder operator to join EPC_Path and Path table and can get the result of query without time condition by the character encoding of the path. Query data sets can be satisfied objects tracking query and object-oriented path query and the needs of traditional data analysis. Finally, since the data is stored in a relational database, the paper describes the item tracking query and object-oriented path query into SQL statements as the template form.