Fabrication of Spatial-Temporal Patterns by Refining High Density RFID Data Sets with Spatial Granularity

摘要

RFID technology is used for various applications ranging from identifying, locating of physical objects to tracking and monitoring them with and without line of sight. It has various advantages over Bar Code system such as cost effective technology, do not require line of sight for the objects to be tracked etc. RFID is automatic and fast and will replace the barcode system in the near future. This technology finds application in a broad range of real time applications such as inventory management, vendor management, and Patient Management Systems. Before using RFID technology, a lot of processing has to be done on RFID data. Data generated by using RFID technology has several drawbacks. These include duplicate records, massive amount of data occupying huge storage. Hence a lot processing has to be done such as filtration, aggregation and transformation into semantic application data. In order to overcome the above mentioned drawbacks present in RFID data sets, many RFID systems employ "smoothing and filtering" techniques. This paper presents efficient and robust RFID data techniques that employ spatial and temporal granularity techniques to reduce the duplication of the data and to increase the granularity to the user defined level. Our experimental results are based on the real time dataset obtained from the website. As RFID data mainly depends on space and time and is gathered at a primitive level, data analysis must takes place at different granularity levels. Hence, in this paper, we propose an approach that will help in extracting and identifying data patterns at real time for different granularity levels.