Microwave Journal
www.microwavejournal.com/articles/3165-a-band-for-all-reasons

A Band For All Reasons

The use of software-defined radios to improve military communications

March 1, 2001

Software radios promise to radically redefine battlefield communications.

This Special Report is reprinted from the January 2001 issue of our sister publication, the Journal of Electronic Defense. Although it describes military applications, much of the material is common to emerging commercial radios. The rapid increase of activity in the field of software-defined radio has implications far beyond the software aspects. Microwave and RF components and subsystems must be configured to accept the diversity of functions that these new radios offer. We hope that this report will be of interest to you.

Michael Puttré
Journal of Electronic Defense


1j14A
In the Gulf War, CENTCOM corps and wing levels conducted a significant portion of the war over commercial telephone lines, partly because of the incompatibility of military communications. Software-defined radios (SDRs) are seen as a tool to improve communications between different services as well as between allies in coalition warfare. (Raytheon photo)

There was a situation reported during the 1999 NATO air campaign over Kosovo where German and American aircraft on the same mission with radios operating on the same band in the same mode received different in-flight instructions. The Americans had a layer of encryption that the Germans did not, so when the American mission commanders issued a recall order, the Germans didn't get it and continued on. No harm, no foul - this time.

Even if somebody had thought to provide German forces with the proper encryption, there was no easy way to modify them in the field. This is because a military radio has evolved into an incredibly complex, specialized device where functions require unique application-specific integrated circuits (ASICs) or other hardwired modules to operate. Adding or changing functions or operating modes requires physical alterations to the radio, with subsequent systems integration work handled by a specialist or subcontractor. Such modifications are not likely to happen in coalition warfare, where partners are apt to offer, withdraw and exchange units with little advanced notice.

Coalition warfare is becoming the rule rather than the exception, with operations in Sierra Leone and East Timor providing a glimpse of the future. The requirement to reconcile incompatible communications will only increase in importance with increasingly complex orders of battle. The software-defined radio (SDR)-or software radio-is emerging as a promising technology to enable different nations and different services of the same nation to communicate with each other without hindrance. What's more, the SDR seems poised to redefine the role of the military radio, turning it into a multimedia, "information superiority" device for exchanging voice, text, images and other data on the battlefield.

Some have likened the software radio to a personal computer, where fairly standard processors do most of the heavy lifting as defined by software programs. Thus, the PC can function as an Internet portal, word processor, data manager and game platform (think "simulation") depending largely on what software is loaded. True, certain functions require additional boards, cards and adapters but the user can add a lot of new functionality simply by

loading new software. Moreover, broadband Internet connections provide access to software sources in remote locations.

A software radio has many of the same features as a PC in that it is a vessel into which myriad applications are poured. Radios that are software reprogramable have been in development since the 1970s, and there are a number of digital, software reprogramable radios in service today. However, in these cases, the software and the hardware are tied together. The concept behind next generation software radios is that they conform to an open architecture, where many vendors are making radios that will run third-party waveforms.

POINT, CLICK, TALK

The bad radio has become something of a cliché in military operations as conducted in the movies. Like many clichés, it has a strong basis in fact. There wasn't a military campaign in the 20th century where faulty communications haven't resulted in screw-ups, typically fatal ones. Innovative or heroic fixes for communications failures figure prominently in lore. Just recall the famous Grenada "dial-a-gunship" phone card episode of 1983, where a member of the invasion force placed a long-distance call to Ft. Bragg to obtain air support.

A decade later, things hadn't improved much. A 1992 report by the House Armed Services Committee of the US Congress indicated that tactical communications during Operation Desert Storm remained plagued by incompatibilities and technical limitations. Central Command (CENTCOM) corps and wing levels conducted a significant portion of the war over commercial telephone lines because of the volume and incompatibility of military communications. Compatibility problems among the services were a constant headache in integrating the air campaign. Multiservice strike packages were difficult or impossible to assemble because various aircraft communicated in different ways over secure voice channels. This caused not only planning problems, but also operational problems. For example, consider this excerpt from the report: "When the Iraqi Air Force was fleeing to Iran, an AWACS controller wanted to signal Navy F-14s and Air Force F-l5s to turn off their radars to fool the Iraqis into taking off so they could be shot down. Because the AWACS is an Air Force aircraft, it was able to communicate this information to the Air Force F-l5s over secure voice but not to the Navy F-14s. The F-14s continued to operate their radars, which kept the Iraqis on the ground."

1j14B
The Joint Tactical Radio System (JTRS) Joint Program Office (JPO) says production versions of this architecture validation model will replace legacy systems. (Raytheon photo)

The armed forces can perhaps be excused for making so little progress in resolving problems of incompatible communications. Services are not organized or even necessarily disposed to reconcile their operational modes or encryption schemes with each other, let alone with the services of other nations they may or may not be fighting alongside in the future. The radio itself is the handicap. Acquisition processes are long, expensive and unpredictable. In the example cited at the opening of this article, the American and German aircraft were equipped with the most advanced-and theoretically compatible-radios available, yet the encryption barrier remained.

All this is changing, largely because of technological advancements initiated in the private sector (see sidebar: "Toward an Open Radio Architecture"). Advances in standard processors, digital signal processing (DSP) and software interoperability have made it possible to develop a radio that is not as dependent on its hardware architecture. Dr. Joseph Mitola, a consulting scientist at MITRE Corp. (Bedford, MA), pointed out that efforts are under way to solve pressing military communications problems with technology that is available today. "There have been many situations in real battlefields where one military unit used an analog, citizens band [CB]-like radio while the other, for example, tactical air, used a digital, HAVE QUICK radio, and the ground force could not talk directly to the supporting aircraft, resulting in less than perfect weapons employment," Mitola said. "One of the first success stories was the bridging of an unclassified CB radio to a HAVE QUICK network, which is a secure frequency-hopped network, using SPEAKeasy I."

1j14C
The JTRS JPO called for the production of 220 ruggedized prototype radios for field testing on a number of vehicles, including several HMMWV variants. (Raytheon photo)

SPEAKeasy is the Rosetta Stone of software radios. The project was undertaken by the US Department of Defense with the goal of developing a multiband, multimode software programmable radio that could allow different branches of the military to communicate on and off the battlefield. The US Air Force Research Laboratory in Rome, NY; the US Army's Communications-Electronics Command (CECOM) at Ft. Monmouth, NJ; and DARPA were the key sponsors.

Mitola describes an early SPEAKeasy interoperability experiment that was conducted at Air Force Electronic Systems Command in Bedford, in 1993. SPEAKeasy I was housed in a six foot rack, used IF analog-to-digital conversion, TI TMS320C30 DSP chips and VME form factor commercial off-the-shelf (COTS) boards (space was not a design goal). That link showed that a single radio could, under software control, operate on two different frequency bands and using two different air interfaces ("modes" or "waveforms" in military jargon) link two disparate radios. Leveraging the SPEAKeasy I experience, Motorola developed SPEAKeasy II and performed a similar demonstration at the National Training Center at Fort Irwin, CA. In this case, an analog air interface waveform was created in about a week, downloaded by modem from Motorola Scottsdale to the Ft. Irwin desert, and uploaded to the SPEAKeasy II units in the field. This happened in 1996-97, approximately a year after the contract was awarded. In this case, the tactical air control party (TACP) was able to operate on the ground using COTS handheld radios using this new SPEAKeasy II waveform, allowing them to talk directly to a HAVE QUICK radio in a tactical aircraft.

The successful demonstrations of SPEAKeasy-type radios led a group of more than 50 companies worldwide to form the Modular Multifunction Information Transfer System (MMITS) Forum, later renamed the Software Defined Radio Forum. The goals of the SDR Forum are to accelerate the development, deployment and use of software defined radios, and to work toward the adoption of an open architecture for the equipment. Working groups within the Forum address specific technology issues related to software radio development and deployment, such as base stations and antennas, handheld units, mobile units and software downloads.

Ideally, this open architecture would allow different manufacturers to make software for radio equipment. "The future of software radios in regards to SDR Forum is similar to the personal computer approach to software," said Kevin Kane, director of Army and Marine Corps business development at Harris Corp.'s RF Communications Division (Rochester, NY), a Forum member. "Instead of a conglomeration of software loaded into radios, pieces of software can be loaded along with file configuration information that assigns significance to each software piece. Today's radio model has very little sharing of technology or code. SDR Forum's model will allow a standard waveform or set of waveforms to be applied to multiple platforms defining the infrastructure by a common architecture."

In practice, manufacturers are working on software radio programs that suit their own technological strengths. Plus, there are prominent SDR manufacturers that are not part of the Forum, such as Rohde & Schwarz (Munich, Germany). Nevertheless, the activities of the Forum, as well as other international SDR development projects, is moving industry fitfully toward a workable consensus on what constitutes interoperability.

A major program in the US incorporating SDR technology is the Joint Tactical Radio System (JTRS), which addresses a family of tactical radios intended to provide line-of-sight (LOS) and beyond-line-of-sight (BLOS) transmission of voice, video and data from the 2 MHz through 2 GHz bands. JTRS can be seen as a direct descendent of the SPEAKeasy program. The JTRS Joint Program Office, led by U.S. Army CECOM but representing all services, has awarded a series of contracts for the definition, development of supporting software, validation and fabrication of prototype systems. A team of manufacturers led by Raytheon Command, Control, Communications and Information Systems (Ft. Wayne, IN), and including ITT Industries (Ft. Wayne, IN), Rockwell Collins (Cedar Rapids, IA) and Marconi CNI Division (Wayne, NJ), defined the JTRS architecture. This, say proponents, is a concrete step toward eliminating fundamental problems in interservice and coalition warfighting. "If a ground-based warfighter, performing a reconnaissance mission, were to see approaching enemy forces, he would be unable to radio that information directly to fighter aircraft in the area," says Tom Schuerman, Raytheon's JTRS program manager. "Currently fielded systems cannot support a direct communications link between them. JTRS will."

In December 2000, the Raytheon-led team delivered the latest version of the Software Communications Architecture (SCA) that will be the foundation of JTRS applications development, which comprise the so-called Step 1 and Step 2A phases of the program. Other manufacturers, including Harris, Boeing (Anaheim, CA), Motorola (Scottsdale, AZ), and THALES, formerly Thomson-CSF Racal Communications (Rockville, MD), are performing validation work on the evolving SCA standard as part of Step 2B activities. The idea here was to get manufacturers who were not part of the original consortium to see if they could integrate hardware and write software to the architecture as developed. Onto Step 2C, BAE SYSTEMS (Wayne, NJ) has a $14-million "other transaction" contract to develop working prototypes in order to validate the networking portion of SCA. To be delivered are 40 engineering development models and 220 ruggedized, 2-channel prototypes for testing under field conditions. Requirements for the latter include the ability to mount the radios in vehicles, such as HMMWV variants, the M1068 tracked command post and the M4 C2V command vehicle.

Harris is using its AN/PRC-117F(C) Falcon II multiband radio as a basis for implementing its portion of the SCA contract. This radio, which is currently being supplied to the US Navy, incorporates several features that will be important to JTRS, notably it is software re-programmable, operates across multiple bands and incorporates multiple waveforms. Some of the work will involve extending the radio's capabilities into the HF spectrum. "In addition to creating the software architecture, one of the ideas driving the JTRS program is to create new waveforms that are more spectrally efficient and have more data carrying capacity," said Mark Turner, director of multiband radio engineering at Harris' RF Communications Division. "The continued challenge of the JTRS program is to support new and emerging requirements, and not just duplicate the old waveforms."

Motorola's contract for Step 2B work required it to verify the SCA by porting it over to an advanced prototype of the Digital Modular Radio (DMR) is building under a separate (and considerably more substantial) contract for the US Navy. The DMR is a software radio that predates the inception of JTRS, and will support the DAMA, HAVE QUICK and SINCGARS waveforms, as well as providing data link coverage for Link 4A and Link 11. According to Tom Nicholson, Motorola's JTRS program manager, Col Anthony Badolato, USAF, program manager for JTRS JPO, wanted something he could show people in a hurry. Motorola's history with software-radio systems (dating back to SPEAKeasy) enabled it to miss many blind alleys to turn the work around in six months.

What they did was adapt its DMR for SCA and demonstrated a complete system with two waveforms, AM and FM.

1j14D
The M3TR is a multiband multimode radio that supports NATO's HAVE QUICK I and II, SATURN and SECOS waveforms. Other national or proprietary standards, such as SINCGARS, PR4G and SEM can be integrated with software. (Rohde & Schwarz photo)

Nicholson's team also documented an applications programming interface (API) for SCA that will enable third-party firms to create software and waveforms for JTRS. "I think what we showed in our in-process review in December [2000] is that the concept of an open, vendor-neutral platform for secure communications is viable," he said. "If the government is serious about going forward, we're serious about trying to get the business."

One concern expressed by MITRE analyst Brian White in a JTRS study sponsored by the Air Force Electronic Systems Center is that Link 16 be supported in the program's development in a timely way. Link 16 is what NATO calls its antijam, secure data and voice system that incorporates waveforms and messages supported by Joint Tactical Information Distribution System (JTIDS) and Multifunctional Distribution System (MIDS) terminals. It could be 2010 before there is widespread deployment of JTRS-derived radios, and in the mean time, there are thousands of datalink systems incorporating Link 16 scheduled for deployment.

"It is our understanding that the USAF is very interested in JTRS-based Link 16 capability and the JTRS joint program office is funding a study for a JTRS-compliant implementation," said Raytheon's Scheurman. "Link 16 has been a requirement since the program's inception and has been accommodated in the definition of the SCA."

Ed Calhoun, principal marketing manager for communications systems at Rockwell Collins, said his firm's activities related to the Step 2B validation portion of the JTRS contract is aimed at providing Link 16 implementation. He says they are on track to delivering demonstration software in February 2001. Although Rockwell Collins is a member of the Raytheon-led team, the contract is independent of that work.

BROADENING THE BAND

Meanwhile, Europe has seen a considerable upsurge in software radio interest, just when the shape of the proposed Eurocorps is emerging. Raytheon and its partner THALES, formerly Thomson-CSF Racal Electronics (Bracknell, UK), are under contract with the UK Ministry of Defense for its Programmable Digital Radio (PDR) program. The award covers the first of a two-phase plan to develop programmable digital radio technology leveraging COTS hardware and standard interfaces. The team is working to define and then develop the language, interpreter and object library, which are the three key components of a waveform compiler that can generate a variety of communication waveforms for PDR. Phase 2 will involve running selected waveforms on a variety of digital radios.

Another THALES business unit, the former Thomson-CSF Comsys (Paris, France), and partner EADS, formerly DaimlerChrysler Aerospace (Munich, Germany) are pursuing an ambitious software radio project that is funded jointly by the French and German governments. The Multimode Mutiband Radio-Advanced Demonstrator Model (MMR-ADM) will act as "network nodes with radio capabilities," replacing a number of individual radio sets today associated with what is commonly known as a communication server." The developers say those nodes will be totally interoperable with legacy systems, although NATO Link 16 terminals are not on the scope for early versions. France and Germany each has a special interest for its national waveform and security concepts, principally the PR4G and SEM 93 waveforms, and share the same interest in reconciling with present and future standardization agreements (STANAGs). THALES and EADS will use the results of the common program work to define its own product line. MMR-ADM prototypes are expected in 2002 for systems tests to be performed during 2003 on seven proposed waveforms.

"Coalition warfighting is a key driver for MMR-based products," said Estelle Griton-Saulner, manager of battlefield business development at THALES. "The ability to communicate either with national security procedures, or NATO or non-NATO coalition protocols with the same equipment will be the basic of interoperability requirement of the coming years. The US JTRS program, in which THALES is participating, shares the same objective."

Closer to possible deployment is the M3TR, which maker Rohde & Schwarz bills as the first practical software radio. One of the key features of the M3TR is its ability to easily accept a variety of waveforms as software downloads. The M3TR supports the NATO HAVE QUICK I and II, SATURN and SECOS waveforms for secure, anti-jam ground-to-air communication. Other national or proprietary standards, such as SINCGARS, PR4G and SEM can be integrated by software, if the operator desires. "The number one problem is that up to now you always had to buy new hardware or to modify hardware whenever you wanted a change in waveforms," said Peter Iselt, head of technical marketing for radios at Rohde & Schwarz. "Now you can do it in software."

For integration into a TCP/IP-network M3TR has an IP-interface. With a data rate of up to 64 kbit/s and simultaneous speech and data transmission on a single channel, the unit can also be used for real-time data and video transmission. M3TR is not deployed as of yet, however Iselt says the company has several contracts with three European customers that are interesting in testing the radios in "diverse applications."

One such evaluation contract is with the Swedish Materiel Command (FMV). According to Rohde & Schwarz, this test program will allow the FMV to build up and evaluate a tactical radio communication environment incorporating voice and data communications, tactical Internet and other subsystem networking functions. The main aim of the tests is to form the basis for a main procurement phase during the first half of 2002 to provide some Swedish armed forces with digital communications at battalion level.

The M3TR covers HF, VHF and UHF, but is not a wideband device. "The RF front end is still primarily an analog hardware application, not a digital software application," said Gunther Wicker, project manager for M3TR. "A particular limitation is the availability of wideband antennas."

THE 'S' WORD

Go a step (or two) further down the development track for software radios, and information superiority (or supremacy) will crop up. This is because visionaries would just as soon drop the radio moniker and look at the SDR as a software-driven processor that, oh yeah, also functions as a radio, I guess.

"We are in a transition from a comfort zone on tactical radios that have been mostly push-to-talk voice radios toward data radios and maybe towards more sophisticated concepts," said MITRE's Mitola. He cited his experience at DARPA in 1998-99 as project manager of a program called Spectrum Supremacy. This involved taking nine HMMWV vehicles out into the desert in Ft. Irwin to find out what information could be gathered on the tactical band. The program proved that four tank radios configured like a SPEAKeasy II could be used to determine the location of enemy radios in the vicinity, say within 5 km of the tank platoon. Although the team performed the analysis after the fact, not in real time, it showed that a tactical radio could function as an intelligence-gathering device, given better signal processing and appropriate applications software.

Mitola said that if this spectrum supremacy concept goes forward, then the future tactical radio becomes a way of 'owning the radio spectrum like we now "own the night' [with advanced night-vision devices]." He reports seeing some aspects of what his modest program proved now appearing in the DARPA Future Combat System (FCS) and other related programs. "But if you visit the JTRS Web site and try to find such advanced concepts in the current JTRS CONOPS [concept of operations] made up by the JFCOM [Joint Forces Command], you will not find them there because not all of these concepts are ready for prime time," he said.

Lest anybody question the determination in some government circles to field fully functional software radios, the JTRS JPO does say the following on its Web site: "JTRS is the DOD radio of the future. Other radios are legacy systems that will be phased out...While this policy is not meant to halt equipment [currently] scheduled for platform installation, it is intended to force the Services to migrate quickly to JTRS...We are firmly committed to making the JTRS vision a reality and request you work with the JPO to join in that effort."

1j14E
Ultimately, the goal is for the soldier in the field to be able to commuincate with ground and air units of all coalition partners. Major technical challenges to any SDR-type program include dealing with miniaturization, power usage and heat dissipation. (Raytheon photo)

Those are bold words, and time will tell whether the money is put up to back them. When the day comes to select a single band radio in its fourth or fifth generation coming off of a well-established production line instead of a brand new multiband multimode military SDR, decision makers will be in the seat that has sunk a thousand promising programs. "The issue is not so much whether software radios are feasible, the question is whether the additional benefits to the warfighter (e.g., bridging of disparate waveforms) are affordable," said Mitola. So when you ask this question, you have to go back to the warfighter's basic need and plans for using radios. Are these to be next-generation information appliances that will bring the vision of information superiority closer? Or are the CONOPS rooted in the past where a radio is just a radio?"

On the other end of the spectrum, Harris' Kane indicated a potential downside in struggling to build a system that is too far removed from the radios currently in service. "You have to give some consideration to the evolutionary approach in fielding these systems," he said. Gingerly raising the oft-mentioned parallel with PCs, Kane pointed to the parade of processors (and operating system versions) that have crossed the average desktop in the last 10 years. "Developing open systems for software radios is an order of magnitude more difficult than PCs," he said. "Maybe several orders of magnitude. At some point you have to buy in, and you don't want that hardware to have too many features and therefor be too expensive to replace when a more capable version comes along."

The sheer number of players in the JTRS saga alone indicates the tremendous potential of next-generation military software radios. Add to that the number of different efforts under way in Europe, which may help push a common architecture forward or perhaps pull it apart. Certainly, no one manufacturer can claim any special prominence in the financial compensation they have received to date from any government. The JTRS JPO, for instance, has been doling out numerous contract awards a few million dollars at a time. And still the companies are queuing up, almost certainly in expectation of a significant payoff in hardware sales-if not software-down the road. *

TOWARD AN OPEN RADIO ARCHITECTURE

True military software radios encompass numerous hardware and software technologies that are either state-of-the-art or still under development. Here's an overview of the key technologies required for software radios:

* Wideband antennas that are able to address the low frequency (LF) through ultra-high frequency (UHF) portion of the spectrum is the main technical challenge that must be overcome. In this area, numerous critical and cost-driven factors are in play: cosite interference mitigation; aperture function; multiband, multimode antenna interactions; insertion loss; optical feed capability; and intelligent antennas that are readily reconfigurable all must be addressed by appropriate research and development (R&D) efforts.

* Radio-frequency (RF) conversion and intermediate-frequency (IF) processing are other domains where major technological breakthroughs are needed. Wideband power amplifiers and low noise amplifiers; signal purity processors; single wideband RF/IF up and down converters; superconducting devices; and tunable preselectors must be addressed by specific R&D actions not driven by the civilian market.

* Analog-to-digital conversion (ADC) and digital-to-analog conversion (DAC) are key technologies that are also in demand for civilian applications. Driven by the cellular base station market, ADC and DAC are evolving quickly, with high-performance products available or announced by various vendors. However, for the military radio market, high value spurious free dynamic range capabilities are required in order to support flexible operational configurations in an intense jamming environment.

* Digital signal processing (DSP) is a fundamental enabling technology for software radio products. Based on field programmable gate array and application-specific integrated circuit (ASIC) hardwired implementations or software algorithms, numerous high-performance DSP components are available, fueled in part by the booming base station and hand-held cellular market. For military applications, the defense community must take care to obtain adequate guarantees of technology accessibility, in particular for submicron ASIC technologies required for high-performance radio applications.

* The radio and processing interface for the definition of the SDR baseband platform is necessary to implement standards boundaries that ensure lifetime continuous upgrade capability. Through this standard interface, moreover, customers obtain the invaluable capability to implement their own confidential waveforms on the standardized plaftorm.

* The Global Radio Interface standard is necessary to allow dynamic implementation of new services via over-air or over-wire download procedures. Military software radios have to support as many radio interface standards as the multimedia networks in which they operate.

The move from legacy, monochannel, monoband radios to standardized, open-architecture, multichannel, multiband radios is a formidable challenge for the defense world. Faced with this aim, the industry and the customer have to focus on the technology needed to develop viable military software radios and, perhaps more importantly, on an innovative approach to acquire, field and support them once developed. *

--Estelle Griton-Saulnier

Estelle Griton-Saulner is responsible for battlefield business development at THALES Communications (formerly Thomson-CSF Comsys).