The FCC filing for the device listed it as having two passive tags, one UHF, and one HF. Seperately it references the 2.4GHz radio. See the FCC filings linked from wdwmagic's article here -
http://www.wdwmagic.com/other/fastp...isney-'magic-band'-rfid-bracelet-revealed.htm
Its true that RFID range can be more than just tap distance - it's all based on the design of course. But on an aside, most of those loss prevention systems at retail are not actually RFID - but other radio based technology.. under the umbrella term EAS - See
http://en.wikipedia.org/wiki/Electronic_article_surveillance for a listing of common methods out there. RFID is a successor to these types of systems as RFID is not just 'detection that the tag exists' - but the tag actually carries data that can be read too.
I'll admit to being at the limits of my direct knowledge of the specific details, and it's somewhat likely that I have many details wrong. Most of what I know comes from conferences, vendor demos, & articles aimed at trying to get independent software vendors to implement applications of the technology. They have to tell us the broad strokes of how it works, but you don't find out the fine details until you actually write applications.
The reason for the different tags is likely different uses. HF is generally looked at as a range of 1m, while UHF tags are looked at as up to 10m. But because they operate on such different frequencies, they have different characteristics when it comes to material interaction (penetration/reflection/etc).
Powered tags are used in part because they improve the response time and the range of tags. But the FCC info describes the RFID tags in use as passive tags - the battery appears to be for the 2.4GHz radio and whatever IC logic is in the band.
After having spouted off on RFID, I did go look at Wikipedia to see how far off I was -- there's some terminology that I may have backwards. But, one of the things that I found there was that there's a third category that's referred to as 'battery assisted passive': essentially a passive tag with a small battery to assist in interaction, but that the tag is powered down until it is activated by interaction with a reader (assuming that its a special type of reader). That sounds like just the thing for a tag that's not expected to be in the presence of readers for long periods of time. I wonder if that counts as passive for the purposes of FCC regulation.
My theory is the active radio is the 'crowd' or proximity feature. The band could be activated by picking up a signal in the area over the air.. and the active radio just sends out it's ID as a 'hey, I'm here' kind of beacon. Then receivers pick up those announcements and can tell how many bands are in the receiver's area. This allows monitoring of larger areas without having irradiate them with higher power RF. Also, the use of a 'trigger' over the air allows the band to automatically shut it's radio off or put itself in a power save state if it hasn't heard a trigger in a period of time.
This sounds pretty feasible. I tend to doubt that there's a security application being developed (no profit in it, mainly), but if there is, it's probably based on wi-fi as well.
Then, the passive RFID tags would be used for more specific 'explicit' transactions that require a direct, one-to-one, or explicit consent transaction. Such as point of sale.. player involvement in a game.. etc. These could be close range or explict tap systems.
Agreed. I'd feel a lot more secure if I knew that the financial information is only exchanged on a tap, and that the tap is required because the range is on the order of 10cm or so. Apparently the State Dept screwed up the RFID on passports, allowing it to be read at a few meters when it was supposed to be a tap-type system. They had to modify the existing passports when some researchers demonstrated scraping information from a few feet away.
Or.. it could be as simple as 'the systems we purchased were incompatible..'.. the door locks they wanted to use use HF.. and the POS systems used UHF
Could be as simple as that... but not really possible to say from here at this point in time. Since this is a NEW roll-out.. it probably has more to do with different application needs.
I particularly like this theory -- for as many different vendors have to be involved for a project this size (and given the reputations of the companies involved, both the vendors and the customer) a little accidental redundancy seems fairly likely.