NYT on RFID

The New York Times has a piece on RFID tags. It's basic, but worth reading as a milestone - the technology is starting to enter the public eye. These RFID tags will have unique serial numbers - every RFID-tagged item you purchase will be uniquely different from every other nearly-identical item, enabling it to be identified and associated with you long after the purchase. And no, microwaving will generally not destroy the tags, and no, most items won't be microwaveable anyway. Try to microwave your couch.

Comments

Blocking

[phuqwit]

You have to start trust in the ingenuity of people ... RSA Security has already found a way to render RFID tags useless.

Privacy issues have surfaced because any reader can read the numbers on any tag. This means a reader in a department store, for example, could not only see what items a shopper has in her cart but could also see what other items she has purchased at competing stores, as well as how much money is in her wallet and what credit cards she's carrying.
The technology that RSA Labs is proposing would make it simple for corporations and consumers to decide which tags could be read by which readers and when. The solution uses what's known as a blocker tag to simulate all possible tag serial numbers. In doing so, it prevents the reader from discovering whether a specific tag is present.

Equipped with blocker tags it would seem that RFID tags become pointless once outside a controlled environment.

OK, here's how they work

[ajs318]

The RFID chip works in conjunction with a tuned circuit {capacitor and coil; the coil also behaves as an antenna} that extracts energy from an applied RF field. The resonant frequency of this tuned circuit is the operating frequency for the system. The size of the coil determines the operating range. An RFID device with integral tuned circuit measures about 20mm. by 10mm. by 2mm. and has a range of a few cm. A smaller device would require an external coil, but the bigger coil would extend the working range.

The transmitter feeds an RF power amp with a sensitive ammeter in one of its power supply leads.

Now, when the tuned circuit is brought within range of the transmitter, it will pick up the signal. But that is all. A voltage will be induced across the system, and a current will flow, but they will be out of phase. When the voltage is at a peak, the current is nil, and vice versa. Recall that power = voltage * current, so there is no power. Bringing the tuned circuit into range of the transmitter will not affect the ammeter reading.

However, if you connect a resistance across the two ends of the tuned circuit, then the current across this resistance will be in phase with the voltage. Energy is now being changed from electromagnetic waves to heat. And, strictly in accordance with the first law of thermodynamics, the reading on the ammeter will go up. Reduce the resistance and it will go up more. Of course, the imperfect coupling from transmitter to receiver itself behaves like a big resistance, which effectively limits the power available for the receiver {and therefore the ammeter swing}.

Anyway, if we switch this resistance in and out of circuit, we can watch the ammeter moving in sympathy with the switching.

The RFID tag gets its power by rectifying the AC induced in the tuned circuit, and using this to charge a capacitor. This capacitor stores enough energy to allow the tag to miss a few cycles, because it unavoidably will as a consequence of how it works. The tag then switches on and off a transistor which sits across the bridge rectifier {a transistor only conducts in one direction} in accordance with a predefined pattern. When the transistor turns on, more power is drawn from the transmitter. {As a side effect, the voltage is pulled down and the RFID tag has to rely on the capacitor contents to keep in this state, remember how far through the sequence it is, and so forth; so this state lasts only a few cycles}. The transmitter can see, by measuring the supply current to the RF power amp, whether the transistor in the RFID tag is on or off.

The external RF field also provides a stable timing reference to the tag, because it can count cycles accurately and dead-reckon a few cycles when it has to.

So, we have a one-way communication from the RFID tag to the transmitter, even though the RFID tag has no power supply of its own. If the RFID tag is absent or high resistance, this is a zero. When the RFID tag goes low-resistance, the transmitter can see this as a one. This allows us to send a binary number from the RFID tag.

All the RFID tag does, once it comes into range of the transmitter, is continuously send out a series of zeros and ones by going low and high resistance. It is up to the transmitter to spot the resistance of the remote end.

It is also possible to send data to the RFID tag, by switching the RF field on and off. While this could be used for programming of tags with serial numbers {instead of laser etching as is currently done}, it would require the tag to have some sort of EEPROM or Flash memory. These devices currently have a high power demand making them unsuitable for operation on RF power alone, but recall Clarke's first law: When a scientist says something is possible they are usually right; when a scientist says something is impossible they are usually wrong. So it is almost certain that future RFID tags could be reprogrammable.


The canonical method for deactivating