You might know Ramtron International for its ferroelectric random-access memory (F-RAM) devices that rely on ferromagnetic polarization in a thin film of crystalline material to store binary information. But the company has started to expand the reach of the F-RAM technology by including it in radio-frequency identification (RFID) devices, or tags. RFID tags are not new, but the characteristics of F-RAM combined with wireless communications will make them a top contender in applications that go beyond basic product identification and tracking.

A thin film of the ferroelectric compound lead zirconate titanate (PZT) changes its polarity in an electric field and serves as a one-bit memory element. Unlike static and dynamic RAMs, an F-RAM retains its data without power because the PZT crystal maintains its polarity. Courtesy of Ramtron.

In particular, F-RAMs operate at high speeds, consume little power during write operations, offer high memory densities, and can maintain data almost indefinitely. Because RFID tags obtain operating power from the RF field that queries them, you can understand the need to operate on small amounts of power. The speed advantage comes to the fore in high-speed applications such as those that exchange information with vehicles on express toll roads. You can read more about the F-RAM technology and operation at: www.ramtron.com/about-us/what-is-f-ram.aspx.

Today, more and more RFID devices use the standard "Gen-2" protocol, or more formally, the "Class 1 Generation 2 UHF Air Interface Protocol Standard" created by EPCglobal, the standards organization for Electronic Product Code (EPC) and Radio Frequency Identification (RFID) technologies. RFID-tag manufacturers have used the Gen-2 protocol for several years and more and more companies have adopted it. You can learn more about this standard and obtain other information about EPCglobal activities at: www.epcglobalinc.org/standards.

That protocol requires some processing overhead, so the F-RAM tags comes in with about a two- to six-fold access-speed advantage over EEPROM-based RFID devices. And the high voltage needed to program an EEPROM device requires an on-tag charge-pump circuit, which increases power demands. Keep in mind that because the tag obtains its energy from a tag-reader's RF field, either the field must be stronger or closer to an EEPROM-based tag to deliver the needed energy. Conversely, because an F-RAM-based tag can operate at lower power, it can offer longer ranges than EEPROM-based devices.

The initial Ramtron RFID ICs in the company's MaxArias family include storage capacities of four, eight, or 16 kbits of information in F-RAM. So you can put an IC on a circuit board to provide version, firmware revision, manufacturing date, test results, and other information. Unlike a bar code that points to a remote database of such information, the F-RAM stores key information locally. Ramtron should have F-RAM RFID development kits available soon, perhaps as early as September 2009. Also, the company plans to announce an association with a company that will produce an appropriate reader for the F-RAM-based tags.

But what about security of tag information? The Gen-2 protocol includes simple password functions. According to Dan Secrest, Ramtron's senior marketing manager for wireless products, the latest EPCglobal revisions acknowledge the need for higher levels of security.

Dan Secrest:
By the end of 2009 we expect to release a chip that will provide 128-bit AES [advanced encryption standard] encryption. Also the F-RAM RFID chips will have an instant-erase function, so when someone tries to 'attack' a chip, it instantly clears the memory. The Gen-2 specs also include a 'kill' function that will destroy the RFID chip.

I bet many engineers will have their own ideas about how to apply the F-RAM-based RFID technology in unusual and unexpected ways. I hope to take a hands-on look at the Ramtron development kit when one becomes available.

--Jon Titus