Browse Prior Art Database

RFID Tag Multi State Reader Function

IP.com Disclosure Number: IPCOM000029551D
Original Publication Date: 2004-Jul-06
Included in the Prior Art Database: 2004-Jul-06
Document File: 2 page(s) / 44K

Publishing Venue

IBM

Abstract

With the proliferation and use of RFID devices and the ever increasing amount of data being read, the ability to process and make local decisions at the RFID reader becomes very attractive. This disclosure introduces the idea of new multi-state coded RFID tag devices used in conjunction with new RFID device readers that can be programmed to do local data analysis and decision making.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 49% of the total text.

Page 1 of 2

RFID Tag Multi State Reader Function

RFID device identification tags are becoming more frequently used and popular with many diverse applications such as inventory tracking, security checking, pricing and many others. One of the main reasons for their popularity is that the tags can be read without physical or sight contact between the device and reader. Some tags are short range readable whilst others can be read at distances of several meters. Such is the read space/volume that many tags passing per second though a volumetric size of several cubic metres can be read essentially at once. In the case of a store exit check for stolen products each code read would have to be looked up in a central database to identify if the tag leaving the store was one that belonged to the store and for which there was no associated purchase transaction. With tags being embedded in the cloths we wear, the car key, and so on, this results in many hundreds of thousand searches being required. More importantly many of the tags passing the reader would be associated with products from many other stores and manufacturers.

  This disclosure introduces the concept of Care / Don't Care / n-State coding that essentially results in the reader establishing a local state on which it can decide what to do next, e.g. if a database look-up is required or not. This function would speed up the tag identification/screening time and reduce both the network and database/software loading in the case of the simple care/don't care case. In the general n-State case the data could be filtered out before transmission, or alarm conditions raised. There is also the potential privacy plus point (if suitably marketed) that only certain tags that have been read by the reader would have further identification and profiling done against them.

  RFID tags range from simple 'fuse' types to tags with bytes and even kilo bytes of storage. This storage can be read only, read only and write (mixed) and all read/writeable. There is currently no universal coding standard for the data in these bytes of memory however tags for specific use, e.g. those for livestock identification do adhere to a standard, but would be totally different to that used for a Mars chocolate bar. With the huge proliferation of tag usage, tag identification becomes ever more difficult. With this idea the reader would be programmed locally with one or more masks or patterns that would allow the reader to create a care / don't care / n-State conditions and thus be then able to locally decide on the action to be taken with the tag data.

  In the ideal case this idea would create a new RFID standard where a number of bytes of data were reserved for care / don't / n-State condition triggering. This data could be set during custom manufacturing of the chips or be programmable after manufacture, e.g. at a store. The number of bytes, codes, etc. is not for discussion here. Referring to them as 'state-specific' data, which can h...