And is it difficult to set up a stored-value system?
With a debit card, you would typically identify an individual and withdraw funds for a purchase directly from his or her bank account, rather than giving that person short-term credit to make the purchase. So RFID could be used to identify that shopper. All you would need to do is to create unique identifiers and link them in a back-end database to a specific cardholder. The key would be to ensure that a thief could not spoof a particular identifier and thus have a purchase be billed to someone else.
A stored-value system involves loading funds onto a card and then deducting the appropriate amount every time a person makes a purchase. This can be accomplished using a secure read-write RFID transponder embedded in a card. Encryption is important in preventing criminals from creating and using bogus cards. Most transit systems, such as the Hong Kong Mass Transit Railway, use a card with a transponder based on the ISO 14443 air-interface protocol, which supports encryption (see When Will All Transit Systems Adopt RFID?).
There are several ways to set up a stored-value system. One would be to read the amount on the transponder and then write a new value. So, if someone were to purchase a card with a face value of, say, $50, that information would be written to the card. If he or she made a purchase for $5, for example, the RFID interrogator would read the amount on the card, calculate the value remaining after the purchase and write that new figure to the tag.
You could also simply store a random serial number on the tag. When a purchase was made, the system would look up how much cash was associated with that serial number and instruct the back-end system to deduct the purchase amount. The next time the card was read, the solution would then look up whether there were sufficient funds remaining to cover the next purchase. The latency of the Internet makes the first approach preferable, in my view.
—Mark Roberti, Founder and Editor, RFID Journal
Previous Post What Is the Main Problem of RFID in Health Care? »