Panel 1: The ABCs of RFID

The first panel of the day was essentially RFID for Dummies. This introduction was led by three panelists: Sue Hutchinson of EPCglobal, Manuel Albers of Philips Semiconductors and Dr. Daniel Engels of Auto-ID Labs of MIT.

First off, where are these speakers coming from?

		EPCglobal EPCglobal is a company to watch since it is trying to establish the worldwide industry standard called the Electronic Product Code (EPC) for RFID. This would be similar to the Uniform Code Council (UCC) that created the barcode standard thirty years ago. Created in September, 2003 and based in Brussels, Belgium, EPCglobal is an alliance of companies around the world, many of whom are working furiously to meet RFID mandates and deadlines set by large retailers like Wal-mart.
		Auto-ID Labs of MIT The Auto-ID Center, the predecessor to the Auto-ID Labs, was an academic research project headquartered at the Massachusetts Institute of Technology (MIT) with labs at five leading research universities around the globe. This center developed the EPC standards (see above description). In October 2003, the Auto-ID Center was closed, leaving administrative functions of EPC to EPCglobal and folding all technical research into the new Auto-ID Labs. EPCglobal and Auto-ID Labs continue to work together through technical committees and action groups.
		Philips Philips Semiconductors is one of the top ten semiconductor suppliers in the world. It boasts to have over one billion RFID chips (wow!) already in use around the world. The leading uses for Philip chips include public transportation cards, payment systems (credit and loyalty cards), access control and ID cards/passports.

And, what did they have to say?

A lot of this panel was devoted to describing the basics of RFID technology. We don't have to rehash that here because there are many resources online that explain this techonology and its uses. Here are some of our favorites:

• What is RFID? This web page has it all spelled out clearly with a great animation for more visual people.

• RFID in a Nutshell: A primer This page is also very clear and tackles the future and implications as well.

• RFID: A basic primer This page is for the more technically inclined. It covers electronics, range and power issues.

Industry rationale for RFID adaptation (based on these three speakers' presentations):

• Increase efficiency: Yes, the "E" word. Industry sees RFID as promising speedy delivery of products from manufacturer to you (the consumer). The products you want will always be on the shelf when you need them! You (the consumer) "win" because there is no waiting. RFID efficiency creates savings that are passed on to you (the consumer). Bottom line: consumer wins. [One speaker said the costs of implementing RFID hardware and accompanying database software would reach 1.3 billion by the year 2008. We wonder how these large costs trickle down to spell savings too?]

• Track and trace all tagged objects: RFID creates full disclosure. Retailers know where its capital is at all times. Knowledge means control and control means (again...) efficiency which ultimately leads to (you got it...) consumer savings.

• Greater depth of serialization: RFID chips allow for more storage capacity than a barcode and allow for unique identification of every single item tagged (that's a lot of numbers).

• Increase product visibility: Retailers know exactly where every product in the supply chain is at every moment. This means you (the consumer) will see more of what you want when you want it (is this starting to sound repetitive?).

• Safety: Retailers and you (the consumer) will be able to find out the manufacture source for every product. This is considered most important for food, airline and pharmaceutical industries.

• Authenticate: RFID tags can be used to prevent counterfeit. This is especially important to the drug industry and high value branding (yes, it really is a Gucci!)
  

History of RFID:

These RFID tags are nothing new. The evolution of these tags can be traced back to 1846 and the first passive RFID tags were around as early as 1973. See this diagram for more information.

What are some RFID technical failures?

OK, so how well does RFID really work? In general RFID tags and transponders can communicate up to 10 meters, but this is in lab conditions. In the real world this distance is reduced to more like 3-5 meters. Tags can communicate through most objects but they don't like metal or water.

TIP: The best place to stick a tag if you want to block its signal is under your armpit! It's probably a little bit damp and remember you are made of 90% water.

Why the RFID now?

Several major retailers are driving implementation through mandates. They are Wal-mart, Wal-Mart (wait did I already say Wal-Mart?), Target, Tesco and Albertsons. Lyle Ginsburg, a moderator in the second panel, stated that if it were not for the Wal-Mart RFID mandate we would not be in a dark FTC conference room on a beautiful summer's day. The Department of Defense (DOD) has also issued a mandate stating that all its suppliers must put passive RFID tags on the lowest possible piece part, case or pallet packaging by January 2005. Likewise the Federal Drug Association (FDA) is encouraging all drug suppliers to use RFID and believes that by 2007 all individual pill bottles will be tagged.

Overview of EPCglobal, its standards and system:

• All tags can be recognized by all readers.

Such tags were referred throughout the day as being promiscuous. Businesses want to ensure that all tags and readers can communicate (or "get it on") because afterall access to this information is the whole reason to adapt RFID in the first place. [Later in the day one speaker said that security would become an issue because businesses would be concerned about competition. You don't want your competitor to get their hands on your sensitive supplier information. There was obvious tension between making RFID data open and useful to all (businesses implied here) or closed and encrypted so that competition would not be an issue. Privacy advocates of course raised concern that if all readers can speak to any tag this creates the potential for unlimited and surreptitious tracking of objects and the people that carry them by anyone interested.]

• 96 bits of information

  EPC tags would hold 96 bits of information, enough to give every tagged item in the world a unique identification number. This was referred to earlier as serialization depth.

• Object Naming Service (ONS)

  EPCglobal is proposing creating, maintaining and controlling access to the universal database that translates RFID unique identification numbers into more useful information. An RFID tag number would be cross-referenced with the ONS database to discover that the object is in fact a 12 oz can of Coke. This number would then point to more detailed information on local ONS servers maintained by the manufacturers. EPCglobal says its system would be flexible like the Internet, accommodating growth (meaning amassing more and more data!) as more retailers incorporate RFID tags into their supply chain and find uses for them.

  The tricky part is who has access to this information, what kind of information is it and how much? An RFID identification tag means nothing, EPCglobal contends, without matching it to a database. EPCglobal said access would only be granted to its member companies and security would be tight. [Subsequent speakers mentioned that member companies would total hundreds and eventually thousands of retailers, so getting access to the list would not be much of a problem due to its wide circulation. Once the database is accessible then anyone with a reader can quickly determine what RFID tags reference which objects. Another speaker said with just a little bit of technical knowledge and spare time, it would not be very difficult for a person to reverse build this database by reading tagged items and recording the patterns. In this case getting access to ONS would not even be necessary to understand what a RFID tag data signifies. This means any person with a reader could read and collect tag data without much hassle.]

  Whether Wal-Mart will ultimately use the EPCglobal ONS system or create one of its own is still unclear.

• Kill Switch

  Kill switches are being built into the EPC system. Once the kill switch is triggered a tag is dead (cannot be resuscitated). Authority to kill the tag lies with the owner of the item; therefore at point of sale the authority to kill is transferred from retailer to customer. EPCglobal has designed this feature to appease privacy advocates. [Privacy Advocates are not so comforted. They say there are problems with the Kill Switch: the burden of kill rests on the customer, industry will undoubtedly come up with incentives for customers to keep tags alive, in store tracking is still a possibility and there are many RFID tag usages that are only applicable over time (e.g. tags in library books and ID cards) so kills switches don't make sense or address all privacy issues.]

What about the Costs?

This is the biggest block for RFID implementation for businesses: it still costs too much. If you buy in bulk (bulk meaning 1 million tags or more!) the lowest cost per tag today is 19 cents. The industry golden mantra is "5-cent tags." Once they are available at this low price then business RFID models begin to compute (in industry speak).