New Jersey startup has come out with the fastest cognitive radio yet. It works on the widest possible range of spectrum, and is part of a crop of improved technologies that are crucial to bringing the technology to market and avert network overload.
The gadget in question, called Cogradio, and made by Radio Technology Systems of Ocean City, New Jersey, can switch at fast-enough rates to be imperceptible for, say, a video viewer; as well as in sufficient quantities that any research done on it, or software written for it, will be applicable in future real-world commercial devices.
The device is the first that can operate from 100 megahertz to 7.5 gigahertz, meaning all the way from AM and FM bands though television and Wi-Fi and cellular frequencies. It can also sense available spectrum and switch between frequencies at around at 50 microseconds, and in some cases as little as one microsecond. This is a record speed, according to Peter Woliansky, a Bell Labs alumnus who made the gadget and founded the startup behind it.
Finally, it can handle 400 megabits per second of data—about eight times what the best home Wi-Fi can do. With this kind of rate—and since it can send on multiple frequencies at once—it could conceivably dispatch 20 HD movies at the same time.
Ultimately, intelligent commercial wireless technologies using such features could allow for more services. And as new and disruptive wireless technologies enter a field now dominated by a few major carriers, it could create competition that lowers costs for average consumers.
The gadget costs close to $6,000, but that’s cheaper than existing models. As prices keep coming down, such gadgets become available to wider swaths of code writers, not just well-heeled labs. “For people studying wireless technology, building the radio and getting it to work is ridiculously hard,” says Woliansky.
Cognitive radio technology could enable a range of new services. For example, it could route cellular calls to Wi-Fi signals—something that is done today in small wireless base stations called small cells—but also avoid having to use fiber to send the signal out over the Internet, and instead use available television spectrum in the 400 megahertz range.
With all such projects, the big challenge is rapid switching speed and high bandwidth—all things that pound on radio hardware. “You want to jump around in radio spectrum as fast as possible and as far as possible, and when you land somewhere, you want to grab as much spectrum as you can, and pump it in and out of the radio, and these are actually very challenging to do,” says Chip Elliot, project director for the NSF’s cognitive radio project at BBN in Cambridge, Massachusetts. “This radio is perfect for things like that.”
Someday, future smart phones and other gadgets will incorporate portions of such technologies. “While this is an important milestone for realizing high-performance and usable cognitive radios, much more work needs to be done by industry on chip design, interfaces, and much else,” Raychaudhuri says. Other companies and research groups are working on developing better and cheaper cognitive radios.
The field is heating up as demands on mobile networks rise. Bell Labs estimates that mobile data traffic will grow by a factor of 25 by 2016, and Cisco says it will grow 18-fold by that year. With the FCC making clear that existing spectrum will run out by next year, new technologies will be needed.