Dreaming of a Single-Chip Mobile Phone
By Mark Frauenfelder, Mon Oct 18 08:45:00 GMT 2004

Two mobile phone components -- antennas and frequency oscillators -- have stubbornly refused the join the rest of the circuitry that has moved onto wireless transceiver chips. But research at the University of Michigan could lead the way to a single-chip solution.

Are mobile phones small enough yet? Most users certainly seem to think so. Keypads can't get much smaller than they already are without becoming hard to use, and as devices continue to offer more features, a large display is a definite plus. So why is miniaturization still such a big factor in the handset industry?

One reason is that Moore's Law makes it possible to make handsets that are smaller than last year's models, and that type of thinking has pushed the industry forward for such a long time that it's hard to stop. Another reason is that having a cool, super-tiny phone in a product lineup serves as a kind of trophy for manufacturers to show off their technological prowess, even though the miniature marvel's appeal is limited to a niche group of techno-fetishists who want the least money can buy.

Nevertheless, today's mobile phones -- not including ones that have non-traditional form factors -- have become about as tiny as human anatomy allows. Does that mean miniaturization is coming to an end? For the phones themselves, the answer is probably yes. For the components inside the phone, the answer is definitely no.

The advantages to shrinking and integrating the internal components of mobile phones are obvious. Component miniaturization means that manufacturers can pack more functions into phones, and integration ultimately leads to lower manufacturing costs.

At the Wireless Integrated Microsystems Engineering Research Center (WIMS ERC) at the University of Michigan in Ann Arbor, two different research projects -- one involving antennas and the other involving frequency resonators -- could help achieve a long sought-after goal -- a true single-chip wireless transceiver.

Traditionally, a mobile phone's antenna is an off-chip component. Either poking out of the phone's case or hidden along an inside edge, the antenna must be large enough to match the frequency at which the phone sends and receives radio signals. But slot antennas, commonly used in TV broadcasting and in exotic applications such as missile guidance, can be much smaller than a comparable wire antenna because they do a better job of transmitting electromagnetic signals. Kamal Sarabandi, the director of the radiation laboratory department in electrical engineering and computer science at UM, and his research group have developed a unique slot antenna in the form of a spiral etched into a square centimeter of metallic plating. Sarabandi designed the shape of the slot antenna in such a way as to forgo the need for additional circuitry normally required to match the antenna's impedance with the transceiver network.

The slot antenna was developed for 2.4 GHz Zigbee wireless sensors, but Sarabadi says it can be converted for use in mobile phones and other wireless devices. Intel is talking with the University about incorporating the antenna technology into notebook computers.

The second UM research project is a MEMS-based effector that replaces the off-chip quartz frequency-resonator chips used in mobile phones to lock onto a frequency. The MEMS device, called a wineglass resonator, incorporates a microscopic mechanical "finger" to flick the material, much as a person would tap a crystal goblet to make it ring. At just 64 microns in diameter, the disk is small enough to fit right onto the transceiver chip (along with the slot antenna).

Associate professor Clark Nguyen, an expert in RF MEMS (microelectromechanical systems for radio frequencies) developed the wineglass resonator for the Zigbee ISO-802.15.4 specification, but the frequency can be changed to work with other wireless standards by adjusting the size of the disk, which is made from silicon.

The next step, says Michael Flynn, who heads the wireless research group at UM, is to put both the slot antenna and the RF MEMS oscillator onto a single wireless transceiver chip. Besides making traditional mobile phones more useful, a single-chip transceiver could foster the development of smaller smart dust motes and tinier non-traditional mobile devices as well. Obviously, the idea of a phone small enough to fit inside your ear canal is repugnant to most people, but a single-chip transceiver will certainly lead to innovations no one has thought of yet.