The many uses for RFID

There’s been a lot of talk about radio frequency identification (RFID) in the IT press recently. For a technology that has been around in various forms since the second world war, its taken a long time to come to market (OK, that’s not strictly true it’s employed within the ID cards that many of us use to access our office buildings, and for Londoners with the strangely named Oyster Card, which is the largest smartcard payment system in the UK, excluding credit and debit cards) but now that RFID transmitters are tiny enough to embed in just about anything, some large organisations are starting to wake up to the potential uses of this technology.

Some of the uses I’ve seen for using RFID in the press over the last couple of weeks include:

RFID is a technology which has the potential to enable enterprises to know every move of every product and service. To privacy campaigners that sounds scary (yeah right, so you have a mobile phone? If so, then your location can already be tracked by the authorities) and the European Union is conducting a public consultation looking at concerns over data protection and how the technology is being used. To me, it sounds scary for another reason – the sheer volume of data that needs to be managed!

The success of RFID deployments is likely to be linked to a network’s ability to handle the data intelligently and securely, according to an IDC report (not surprisingly commissioned by Cisco), predicting that RFID will have a significant impact on enterprise networks not just because of the number of tags involved, but because of the amount of data each tag could hold and the number of times it is scanned during transit or processing.

I recently read an excellent article in Enterprise Server Magazine (now renamed Server Management), contributed by Mark Palmer and entitled “Making Meanings”. I could not find it online, but the nice people at ObjectStore were happy to send me a copy, which I can’t publish here (for copyright reasons), but which I’m sure they would send to anyone else who is interested. In the article, Palmer sets out seven principles for the effective management of RFID data:

  1. Digest RFID event data close to the source of the RFID activity (i.e. convert from many raw events to a collection of meaningful events) to ensure greater reliability and protect the IT infrastructure.
  2. Whether or not a complex event processing (CEP) tool is used or one is built specially, the principle is the same – to turn simple events into meaningful ones in order to derive knowledge on which actions may be taken.
  3. Data concentrators can be used to achieve reliable speed, by buffering event stream flows, combining RFID middleware, event processing and in-memory data cache.
  4. RFID event data can be processed in context by caching reference data.
  5. Federate data distribution so the RFID system can scale and yet still provide information in near real time.
  6. Age RFID data to keep the working set manageable, enrich raw data with context and reduce the load on downstream systems.
  7. Automate exception handling to improved overall business efficiency.

Another area which needs to be addressed for RFID to take off is that of standards – many of the existing standards are US-based and some experts would like to see the RFID electronic product code (EPC) standards body work with the International Organization for Standardization (ISO), so that EPC can focus on product codes and ISO on frequency.

In the meantime, the Computer Technology Industry Association (CompTIA) which runs the A+ and Network+ certifications is said to be developing a certification scheme for RFID skills.

Microsoft plans to launch its RFID services platform in 2006.

Leave a Reply