In experiments and even limited human clinical trials, electrode arrays implanted on the brain's surface have given monkeys and humans the ability to move objects with their thoughts. The experiments are proof that brain-computer interfaces could improve the lives of severely paralyzed people. But these systems rely on wires snaking out from the skull, which would affect a person's mobility and leave an opening in the scalp prone to infection.

Wireless brain-machine interfaces would be much more practical and could be implanted in several different areas of the brain to tap into more neurons. A typical scheme would have electrodes penetrating brain tissue, picking up neuronal electrical impulses, called spikes. A chip would amplify and process the signals and transmit them over a broadband RF connection through the skull to a receiver. Then, just as in wired systems, algorithms would decode these signals into commands for operating a computer or a robot.

The key requirement for such a system is that it consume very little power to keep the heat down. "Most of the guidelines for implantable devices say that you should not raise the surrounding tissue temperature by more than 1 C; otherwise, you'll kill the cells you're trying to record from," says Reid Harrison, an electrical and computer engineering professor at the University of Utah, in Salt Lake City.

Sending the complex analog impulses as they are would take up too much bandwidth. So it will be necessary to convert them into a simpler, robust form as close as possible to that of the neurons, says Brown University neuroengineer Arto Nurmikko. He and some of his colleagues were associated with the now-defunct Foxborough, Mass., start-up Cyberkinetics Neurotechnology Systems, which did the first human clinical trials of an implanted brain-computer interface. Now his team has a promising wireless interface scheme, which they presented last month at the IEEE Engineering in Medicine and Biology Conference (EMBC).

The researchers have implanted a 16-electrode microchip into the arm-control region of a monkey's brain. The chip amplifies neural signals and sends them via a flexible wire ribbon through the skull to another chip beneath the scalp. Here, the signals are digitized, and a diode laser transmits the data as infrared light pulses through the skin to a receiver. The system uses 12 milliwatts of power in its current configuration; expanding it to a 100-electrode scheme, which would allow for controlling a prosthetic, would require 30 mW, the safety limit for cranial use.

Working along similar lines, Utah's Harrison, together with engineers and neuroscientists at Stanford, has designed an 8-mW radio chip with 100 amplifiers that could be connected to a 100-electrode array. A radio transmitter sends data at 350 kilobits per second but only over a distance of 5 centimeters. Increasing the range to 4 meters shoots the power needs past 63 mW. To stay within the transmitter's low RF bandwidth, Harrison designed the amplifiers to send signals only when a neuronal spike exceeds a certain voltage threshold. He is now working on extracting other neuronal spike features, such as width or height, to be able to distinguish between different neurons.