The above video gives an idea of what places like Walmart would like to do with RFID, and this type of item-level tagging is what has been driving the market. RFID has been in use long enough that this sort of thing ought to be available, but it’s not.
Let’s start with RFID: it works like a barcode. You have a scanner that reads the barcode using lasers. In order to read the barcode, you need a line of sight: the scanner has be able to “see” the light hitting the barcode.
RFID is similar, except it’s using a frequency of electromagnetic wave different than the range of visible light. This frequency doesn’t require “line of sight” because objects that are opaque in the visible frequencies will be transparent at the operational frequencies that RFID systems use. This is because RFID tags use antennas to capture and modulate electromagnetic waves.
As an example, I can show you one we whipped up at work:
This picture comes from the simulation software used to design the tag. (Unfortunately, I don’t have any actual pictures of the tag.) The antenna, which captures the electromagnetic radiation, is the gold letters. The little grey block between the words is where the IC goes. When the wave hits the antenna, it induces a current through the IC. The IC is powered by the current and then modulates it, causing changes in the reflected wave. A reader (like a barcode scanner) will pick up the reflected wave and be able to ID the tag from that. This method of communication between a reader and an RFID tag is called backscatter.
A few years ago, a group at Georgia Tech did a study of RFID tags on a variety of different materials. They looked at how the signal dropped from a reference level, found by placing the RFID tag in free space, versus on materials like wood, deionized water, propylene glycol, hamburger, and metal. What they found was that the signal strength often drops when the tag is next to those materials. Certain materials also would reshape the antenna’s beam: in other words, the antenna would have ‘null spots’ where the reader wouldn’t pick up a signal anymore.
The worst culprit was metal, but water, hamburger and propylene glycol also adversely affected the backscatter signal from the RFID tag. Prior to this, Dobkin and Weigand performed a study showing that the result of the reduced ability to detect RFID was because the electromagnetic field close to metal surfaces must become zero or very small close to dielectrics. Sometimes the antenna can also become detuned: that is, its resonance point becomes shifted so it doesn’t operate very well at the frequency it was designed for. However, this is less an issue than the reduced field strength.
Going back to the commercial, we can now understand why we don’t see RFID on individual items quite yet. Certain materials, especially those high in water, won’t allow RFID to work very well. You have to consider that your body is also made mostly of water, so it won’t be terribly friendly to any RFID tags sitting near or next to it. Put it on hamburger and next to your body, and it may as well be invisible. Many grocery carts are metallic, and those will really have a negative effect on the signal.
At this point, RFID isn’t sophisticated enough to do item-level tagging, but many major retailers still want it to be. As work on improving the technology continues, we may at some point be able to breeze through checkout like in the commercial above.
Griffin, J., Durgin, G., Haldi, A., & Kippelen, B. (2006). RF Tag Antenna Performance on Various Materials Using Radio Link Budgets Antennas and Wireless Propagation Letters, 5 (1), 247-250 DOI: 10.1109/LAWP.2006.874072
Dobkin, D.M., & Weigand, S.M. (2005). Environmental Effects on RFID Tag Antennas Microwave Symposium Digest, 2005 IEEE MTT-S International : 10.1109/MWSYM.2005.1516541