Farpoint Group –
You've been hearing a lot about RFID recently, and with good reason. The promise of RFID is essentially the creation of a radio-based bar code, meaning that we can use radio, and not light from a laser in close proximity to a bar code, to obtain a small amount of information from an RFID tag. Maybe you saw the IBM TV commercial that ran last year depicting a shady looking guy stuffing grocery items into his coat, and then making for the door. As he exits, a security guard calls to him. Are we about to see an arrest for shoplifting? Nope - the guard merely informs the shopper that a receipt had been forgotten.
As our shady shopper exited, there was some obvious scanning going on, shown as a kind of laser-beam effect. But with RFID, again, no lasers are involved. As our hungry friend passed through the scanner (or interrogator, in RFID parlance), all of the RFID tags on his various items were scanned, and they responded, again, with a small amount of unique information sent via radio. A total amount was computed, and his credit card automatically charged. Sound cool? It is.
But don't get carried away with this vision just yet. RFID is in use today, and there are lots of companies building tags, readers, and software, and offering systems-integration services as well. But consumer applications? Not yet, and here's why:
RFID tags cost a lot more than bar codes, which are essentially free. There are, by the way, many different types of bar codes; what you see most often is the Universal Product Code, or UPC. All, however, work pretty much the same way: scan a laser over the code, and read the reflection. There's enough error checking in the codes so that mis-reads are very uncommon. And, again, bar codes are very, very cheap: just the cost of the ink to put them on a label. You can find a lot of information about bar codes at the BarCode 1 Web site.
RFID tags, on the other hand, cost anywhere from about 50 cents to a few dollars apiece, meaning that today they're only suitable for high-value items, or collections of items (e.g., shipping pallets). If we add location capability to the tags (allowing them to be tracked and located in a given venue), the cost can be much higher. I also want to stress here that simple tags designed to simply indicate the presence of the tag and not radiate a particular value (and thus identifying an object to which the tag is attached) also count as RFID, albeit in a very simple form. These tags can cost less than a penny apiece in large quantities, and are most often used in retail environments for "shrinkage reduction" (aka theft prevention).
As we noted above, bar codes are very, very reliable. The scanner usually can address only one code at a time, and audio feedback is the usual mechanism to inform the operator of a good read. We can even read relatively large amounts of data using so-called 2D bar codes specs like PDF 417.
It's not so easy in the case of RFID. Radio propagation being what it is, a tag can make it past a reader without any indication of its presence. Tags with batteries (the so-called active tags) might go dead and this be unreadable regardless. A large number of simultaneous tags in a given location can confuse a reader, again leading to errors. We can improve the quality of RF tags, but it's not easy or cheap to do so. And we'll always have to deal with RF issues.
The RFID Journal offers a good FAQ on RFID, and there are lots of other sources on the Web.
Many people are also concerned about privacy, and RFID doesn't help here. But can we track objects over long distances using cheap RFID tags? Not likely - the radio signals just don't propagate far enough. And who really wants to know that a certain gallon of milk is in your particular refrigerator? Still, we can use implantable, subcutaneous tags to track both animals and people under some circumstances. A few years ago my family helped train a guide dog for the blind. He came equipped with such a chip in his ear (which was something like this one), a worthy addition for such a valuable animal.
And there are lots of other applications for RFID - a brief search on the Web will produce way more information than anyone can ever read. But don't assume that RFID is going to take over the world and be integral to every can of corn you buy - I personally don't believe this will ever happen. However, for the right applications, RFID can be a valuable and appropriate wireless tool, and you'll be hearing a lot more about it regardless.