In the midst of any crisis, whether it be a peacekeeping operation in Afghanistan, an ice storm in Quebec or a terrorist attack in Madrid or Mumbai, communication is key to protecting civilians and keeping rescue and military personnel coordinated and safe. Yet, when firefighters from outside the city poured into Madrid to help pull victims from the smoldering trains, they couldn't radio a local ambulance to report a critically injured child. And a Canadian peacekeeper in Afghanistan can't radio the pilot of a British F-16 flying directly overhead to request much-needed back-up from his NATO colleague. Even hydro crews from New Hampshire coming north to repair power lines in the aftermath of a storm can't coordinate their efforts with teams from Ontario and Quebec. How can this happen with so much at stake?

The answer, says Steve Bernier, project leader for the Communications Research Centre's Advanced Radio Systems Laboratory, lies in the limitations of traditional radio technology. Radios - whether AM/FM, cell phones or walkie-talkies - send information by changing or modulating radio waves. For two radios to communicate, both must send and receive information using the same waveform (wave shape) at the same frequency, and the waveform and frequency are fixed in the radio's hardware. Once built, a traditional radio cannot transmit or receive a new waveform without replacing major components. To further complicate matters, each manufacturer designs their radios to send and receive their own waveform using proprietary hardware. Sharing, in the world of radio technology, has not been a part of the business model.

"It's the reason you can't switch from one cell phone provider to another and keep the same phone," explains Bernier. "Networks use different waveforms and protocols (e.g. GSM, CDMA, TDMA) so you'd have to actually change the signal processing hardware inside the phone to switch networks."

The solution, says Bernier, is software defined radio, where signal processing is carried out by software rather than dedicated hardware. By using software to process signals, a single radio handset can transmit and receive new waveforms by simply downloading the software needed to process that particular waveform. The radio becomes, in essence, a mini-computer with an antenna.

It is a technology, Bernier says, that is on the cusp of transforming wireless communications, and CRC has been at the forefront since its inception.

Military standards

The history of software defined radio dates back to the late 1980s when the US Department of Defense was striving to develop better field coordination between the various arms of the US military. One of the biggest hurdles to tactical unification, it turned out, was radio communications. With each group - Air Force, Army, Navy and Marines - in charge of its own radio procurement, a Marine standing on a beach couldn't necessarily radio an army platoon stationed a mile inland. It was an untenable situation, and one that cost lives.

The solution, the Defense Department decided, was to move to software defined radio, but to ensure interoperability between radios - and avoid the pitfalls of the past - they knew they would need a stringent set of rules that had to be adhered to for all software-defined military radios. This "standard" would ensure that the waveform used by one radio could be readily downloaded by another, regardless of the manufacturer.

At the Communications Research Centre (CRC) in Ottawa, both Steve Bernier and Claude Bélisle, Vice-President of Satellite Communications and Radio Propagation Research, were watching the military developments with interest. Both they and their team of engineers and researchers had recognized the enormous potential for software defined radio, a potential, they felt, that extended well beyond military applications. But for that potential to be realized, an international standard was required, and such a standard would have to be founded on component-based software design. The combination of an open-source standard and a component-based design would not only ensure wide-ranging interoperability between radios, it would also allow small companies to compete in a field traditionally dominated by multinational manufacturers. Smaller companies would, for the first time, be able to develop and sell specialized applications to multiple radio manufacturers or directly to the consumer, in the same way independent software companies develop and sell programs directly to owners of Windows PCs. In short, Bélisle and Bernier saw that SDR technology, governed by an international standard, would transform wireless communications, and they wanted Canada, and Canadian companies, in on the ground floor.

Helping Canadian companies get in the game

Their first step was to obtain an early version of the military standard then under development. Once they had that in hand, says Bernier, his team spent two months pouring over the specifications. Their goal was to do what a manufacturer would ultimately have to do: design a functioning radio that would follow all the standards and specifications.

"In the end," says Bernier, "we couldn't do it. There were too many missing pieces in the specifications."

With a detailed knowledge of the shortcomings of the existing version of the standards, they arrived at the next SDR Forum meeting with a proposal to help solve the problems. To convince the Forum of the viability of their approach they had even developed a working prototype of a software defined radio which they showcased at one of the meetings. Their proposal was accepted, and CRC found itself exactly where it wanted to be: in a position to influence the direction of the standards. In 2000, the Software Communications Architecture (SCA), which CRC helped develop, was accepted as the US military standard. Essentially a set of rules, the SCA governs the development of the software and hardware, and how they must interact, on any software defined radio used by the US military.

But, for Bélisle and Bernier, a standard that applied only to the US military wasn't good enough. For the technology to reach its full commercial potential, the standard had to be recognized on an international level, and for that to happen industry had to buy in. The SDR Forum agreed. With the SCA now accepted as the military standard, the challenge was to promote it on a wider scale. Given CRC's proven track record in contributing to the SCA and building a functioning software defined radio, the SDR Forum turned to CRC for help.

For industry to buy in, Bernier and his team reasoned, they needed to see a software defined radio in action. Better still, they needed to be able to build one themselves. With CRC radio designers and software developers working in concert, the team designed a prototype software-defined radio that could be easily built out of readily available standard parts. At the same time, the team developed a "reference implementation" of a core-framework - similar to an operating system that runs a computer - based on the SCA specification, and posted the source code for the core framework on the CRC website. Companies could then download it and use it as a reference guide when developing their own SCA-compliant systems. Along with the hardware schematic of a prototype SDR radio, which they also posted on the CRC website, companies now had all they needed to build an SCA-compliant FM radio for less than US$100.

"That got things moving," says Bernier. "Companies began to call asking us how they could get into SDR, everyone from little software firms to big radio manufacturers."

Tools for developers

But all the interest in SCA-compliant software-defined radio also had the potential to create a major problem for the lab. As the foremost authority on the implementation of what was a very complex standard, CRC risked being swamped with questions from companies building SCA-compliant radios and developing software. What the industry needed, Bernier and his team realized, was a precision tool: an SCA software development kit that would take care of all the SCA-required code. By freeing up the developer from learning and writing all the intricate code required by the SCA, they could concentrate on what was important and profitable: creating their own applications and bringing components quickly together into a functioning system. The result was the SCARI++ software suite, now used by manufacturers and developers worldwide.

"Our toolbox," says Bernier, "helps people program these devices quickly, easily and in such a way that they must comply with the SCA standards. They can go from a little bit of source code with the mathematical algorithms they have, to an application that can be launched and run on an SCA-compliant radio."

Between CRC's core-framework and the SCARI++ development kit, CRC has provided companies with the tools they need to get a running start in SDR, but has it benefited any of those Canadian companies that both Bélisle and Bernier had hoped to help? Has their strategy been a success?

The answer, according to Carl Villeneuve, Software Engineering Manager of Montreal's Ultra Electronics, is a resounding yes. Ultra Electronics is a major supplier of tactical high-capacity radios for military applications. Their radios must not only move large volumes of data in difficult operating conditions, they must accommodate specific client needs for high security and jam-free operation, as well as be fully SCA compliant. Their radios, says Villeneuve, run CRC's core framework.

"The core framework is the nuts and bolts of the software communications architecture," explains Villeneuve. "For example, a customer might want their radios to synthesize a certain frequency, so we develop an application that allows them to do that. However, all the work of starting, integrating and stopping that application is done by the core framework."

Using the core framework developed by CRC, says Villeneuve, has been critical to their business. "We couldn't have done it on our own. To be able to take advantage of an investment that the Canadian government made, as well as having a group like CRC, with its resources, available to us has really springboarded us ahead of our competitors."

Maxime Dumas, Business Development Manager of Quebec City's Lyrtech, agrees. Lyrtech provides university labs, research centres and radio manufacturers with software-defined radio development platforms that allow customers to quickly develop SCA-compliant applications. Their platforms, says Dumas, run the CRC core framework, and customers are provided with the SCARI++ software suite.

"We had the expertise for the hardware and the embedded software, but in terms of the global scope of the SCA, we didn't have the complete expertise. CRC not only had the expertise, but also the tools, so we could offer customers a completely integrated solution, which had tremendous added value. We couldn't have SCA-related customers without that piece of the puzzle."

CRC's core framework along with the SCARI++ development tools are also integral to Spectrum Signal Processing by Vecima. Located in Burnaby, B.C., Spectrum Signal Processing is a leading supplier of software defined platforms for defense electronics applications. CRC's core framework and development tools, says Mark Briggs, Vice President of Marketing and Operations, has added significant value to their existing products, especially for military-communications customers.

"Access to these technologies has made our firm more competitive in global markets," says Briggs, "and we're firmly committed to continuing this highly productive industry/government partnership."

While both Bélisle and Bernier are delighted to see that their strategy worked - that the software defined radio technology they laboured to develop is giving Canadian companies a competitive edge in international markets - Bélisle, at least, is not surprised. "Technology transfer and support to industry have always been, and will continue to be, a key element in all of CRC's research programs."