Mobile Phone System Passes Texas Test

Could "software-defined radio" transform digital communications?

4 min read

Photo: Getty Images /Image Manipulation: Michael R. Vella

Texas, according to the old saw, is just miles and miles of miles and miles. Today, if you happen to be traversing some of those miles, say in Comanche County, and you want to use your digital cellphone, there's a good chance you can't. That's a headache for Toney Prather, CEO of Mid-Tex Cellular Ltd. in DeLeon. His company provides basic analog and time-division multiple-access (TDMA) digital service but not the two other second-generation formats, code-division multiple access (CDMA) and GSM, the Global System for Mobile Communications. And that's not to speak of third-generation systems like cdma2000 and WCDMA, with enhanced data-handling abilities.

Those are major drawbacks for Prather. He would like visitors to Comanche County to be able to use any phone in his territory, not just for their convenience but also so that his company can reap a little of the roaming revenues. What's more, he'd like to reduce or eliminate upgrades for Mid-Tex Cellular's base station hardware--the equipment that receives cell calls and transmits them to other cell or landline phones.

That's why, last July, Prather agreed to start testing a technology known as software-defined radio (SDR). He is trying out equipment and software provided by a communications research and development firm, Vanu Inc., in Cambridge, Mass. Now five years old, the company is named for its founder, Vanu Bose, the only son of Amar Bose, who developed the famed Bose audio system. The firm's focus on SDR grew out of Vanu Bose's doctoral work at the Massachusetts Institute of Technology on radios that use software instead of hardware to handle signaling.

Basically, the idea of software-defined radio is for smart, or "cognitive," devices to be able to operate across a wide frequency spectrum, dynamically picking the vacant portions and using them, and then to perform multiple functions. This is in contrast to conventional radio, which requires a different device for each function [see "A Cellphone for All Standards," IEEE Spectrum, May 2002, pp. 34-39].

In a standard radio, a transceiver, tasked with getting a signal in or out, plugs into a signal processing subsystem (SPSS) that does all the hard work of getting useful information (words or data) into or out of a signal. The subsystem performs such functions as coding/decoding, compression/decompression, modulation/demodulation, and spreading/despreading (where spreading is sending a signal over a wider channel than strictly necessary, in order to handle interference better).

It's the SPSS, limited at present to specific communications standards or modes, that has been the focus of an approach to software-defined radio called "reconfigurable SDR." The SPSS is made up of digital signal processors (DSPs) and field-programmable gate arrays (FPGAs), both of which can be programmed. The only problem is that the software available for programming them is very device specific. While C/C++ compilers exist for DSPs and FPGAs, they produce inefficient output unless the input code is written specifically for the target chip. Therefore, writing the code tends to be expensive, and the software for reconfigurable SDR isn't portable from one chip to another.

By using portable, off-the-shelf, nonproprietary technology that tracks Moore's Law, "Vanu will always win," says Intel's Mike Chartier

What Vanu has done is to adopt what it sees as a cheaper and, in the long run, much more effective approach. The company has written all the signal-processing functions into portable high-level code running on general-purpose or hybrid general-purpose processors rather than the DSPs and FPGAs that are the crux of the reconfigurable approach to software-defined radio.

"In traditional SDR," says John Chapin, the company's chief technology officer, "the software is unique to the hardware it's built for. Vanu's software is portable. We can meet the customer's [SDR] requirements easily, whether [they are implemented] on a laptop or a high-reliability telecommunications server." Processors to which Vanu has ported its SDR code include the Intel Pentium and StrongARM, the SuperHHitachi SH4, the Compaq Alpha, and the Motorola PowerPC.

Mike Chartier, Intel Corp.'s director of regulatory policy in the corporate technology group in Phoenix, Ariz., describes this use of off-the-shelf technology as "disruptive," in the sense that it will shake up the communications business. Off-the-shelf technology "tracks Moore's Law," he says, becoming exponentially more capable over time. "By using such technology, Vanu will always win, because such a solution is cheaper than a proprietary one. It's a winning business solution."

It's the off-the-shelf solution that makes the trial in Comanche County notable. At this stage of the game, however, the Vanu SDR technology is located in Mid-Tex Cellular's base stations, not in the cellphones. Conventional phones limited to one digital mode will still be used, because it's simply too expensive at present to build consumer SDR handsets that can change modes. Those might not be available for a few years yet.

Prather wants Mid-Tex to be able to transmit and receive signals in the main U.S. digital modes to support all domestic cellphones. So with the equipment and software installed, Prather's base stations will transmit multiple control signals on different frequencies. For example, when both GSM and CDMA have been enabled, both of Prather's base stations will send and receive signals on a GSM control channel at 840.1 MHz and on a CDMA control channel at 845 MHz, operating GSM and CDMA software simultaneously. When someone's cellphone seeks a base station, that phone will find the control channel of the correct type and connect to it.

To perform the trial, Vanu installed SDR base stations incorporating Hewlett-Packard ProLiant servers, ADC transceivers, and Telos soft switches into two Mid-Tex Cellular facilities in DeLeon that were a few kilometers apart. Vanu's high-level software running on the H-P Proliant servers was the key element in the trial. For the approximately 400-square-kilometer area of the trial, the software in the base stations supported GSM, and GSM phones could function.

Two radio transmitters and two antennas were used in the trial. One transmitter-antenna pair was in Gorman, and the other was in DeLeon. The two towns are about 20 km apart. Prather says, "We tested cellphone 'handoffs' in traveling from DeLeon to Gorman [to make] sure the signal didn't get dropped."

By the end of last September, the trial was declared a success. However, the two companies kept the trial going until year's end, so that Vanu could use it for more demonstrations. In mid-December, Vanu, H-P, and Mid-Tex signed contracts to begin the phased overlay of GSM and GPRS (for General Packet Radio Service, the radio packet side of GSM) over 24 Mid-Tex sites. The entire rollout is expected to be completed by year's end.

Once the network is in place, Prather plans to start transitioning his TDMA customers over to GSM. CDMA will be overlaid "whenever Vanu is ready," Prather says. "It's all a software load." (Vanu's technology chief Chapin notes that Vanu hasn't committed itself officially to any specific time frame for providing CDMA software.)

The hardware changeover needed for the trial was the third in 11 years for Mid-Tex Cellular. "But it is actually the last time for major hardware," Prather says. Because of SDR, "in the future, it will all be software."

This article is for IEEE members only. Join IEEE to access our full archive.

Join the world’s largest professional organization devoted to engineering and applied sciences and get access to all of Spectrum’s articles, podcasts, and special reports. Learn more →

If you're already an IEEE member, please sign in to continue reading.

Membership includes:

  • Get unlimited access to IEEE Spectrum content
  • Follow your favorite topics to create a personalized feed of IEEE Spectrum content
  • Save Spectrum articles to read later
  • Network with other technology professionals
  • Establish a professional profile
  • Create a group to share and collaborate on projects
  • Discover IEEE events and activities
  • Join and participate in discussions