For its first 100 years or so, the improvement of radio was driven by the development of better hardware, including improved components, more sophisticated circuits and more precise manufacturing.

During this time, each type of radio was designed with a specific purpose in mind, and the waveband it operated on and the wave form it used was fixed by the hardware. To change a device’s behaviour was impossible unless you ripped out and replaced substantial parts of the circuitry.

The arrival of software-defined radio (SDR) changed all that.

SDR takes advantage of the processing power of modern computer technology to emulate the behaviour of a radio circuit. The software defines how the radio performs, making it feasible for a single radio to emulate and communicate with many different types. Unlike its hardware-based counterparts, these devices can be improved and updated and given new capabilities by changing the software.

Theoretically, it is possible to reduce SDR to a computer that decodes everything arriving at an antenna that’s connected to it. In practice, however, this is not feasible: antenna voltages are tiny, way below anything a computer’s analogue-to-digital converters (ADC) can handle, so at the very least an SDR has to have a low-noise amplifier between the antenna and the ADC.

“SDR takes advantage of the processing power of modern computer technology to emulate the behaviour of a radio circuit.”

Amplification can create problems, though, because spurious signals from other equipment and even background radiation are also amplified and can distort the desired signals, or even block them completely. One solution is to put band-pass filters between the antenna and the amplifier, but these reduce the radio’s flexibility. Some SDRs have several switch-able channels, each with its own filters and amplifiers, increasing discrimination but reducing flexibility and adding complexity to the circuit.

How it started

The SDR concept originated through work carried out in the defence sector in Europe and the US, although the term wasn’t coined until 1991 when Joseph Mitola published the first paper on the topic at the IEEE National Telesystems Conference 1992.

The US SPEAKeasy projects in the 1990s were the first SDR implementations using programmable processing to emulate existing military radios, while General Dynamics‘ Digital Modular Radio (DMR) system, incorporating SDR, was adopted around ten years ago by the US Navy. Previously, ships often carried several racks of radio equipment to communicate with aircraft, shore, small boats and assorted allies; with an SDR facility on board, this was reduced to a single rack, saving space and weight, and reducing complexity. General Dynamics claims its DMR system typically replaced 14 different radio racks and took up 50% less space.

Early SDR systems employed proprietary software making it difficult to ‘port’ components from one radio built by one company to another. As a result, the Department of Defense wanted the next generation, Joint Tactical Radio System (JTRS or ‘jitters’), to be standardised. JTRS is based on an open software communications architecture (SCA), which is increasingly recognised as an international standard and used by equipment manufacturers around the world, while the European Secure Software Radio Programme (ESSOR) standard is also based on SCA.

“Manufacturers view software as just another component that can be bought, like batteries and switches.”

JTRS radios have become much smaller and more energy-efficient since DMR was first used on board ships, and the system is now widely used in hand-held radios too.

Field-programmable gate arrays (FPGAs), with their inherent flexibility and reprogrammability, have also dramatically increased the capabilities of modern SDR systems, and enables them to support emerging and changing SDR waveforms. FPGAs meet the radio’s performance needs while keeping size, weight and power consumption to a minimum, extending battery life.

Software design

With SCA standardising the way software interacts with radio hardware, it is no longer necessary or desirable for each hardware manufacturer to create unique software. Hence, manufacturers now view software as just another component that can be bought, like semiconductors, batteries and switches.

Specialist software companies such as Prism Technologies and Wind River exist to supply this market.

Prism’s Spectra SDR development suite allows SDR manufacturers to create their own compliant waveforms, either new or to match legacy equipment. Wind River produces operating systems for SDR, and the two companies have joined forces to provide a complete, high-performance, commercial off-the-shelf SDR solution for a wide array of hardware.

Next Step – cognitive radio

Having done much to start SDR in the 1990s, Joseph Mitola turned his attention to cognitive radio, which he defined in his 2000 doctoral thesis:

"The term cognitive radio identifies the point at which wireless personal digital assistants (PDAs) and the related networks are sufficiently computationally intelligent about radio resources and related computer-to-computer communications to: (a) detect user communications needs as a function of use context, and (b) provide radio resources and wireless services most appropriate to those needs."

What applies to a PDA can apply equally well to military SDR systems. The idea is that radios should be smart enough to find space in a crowded spectrum and use it to join whatever network the operator requests, breaking the link between network, geography and a particular location on the electromagnetic spectrum.

The radio spectrum is already crowded, and during conflict or in disaster relief operations it becomes even more so. The objective of cognitive radio is to improve interoperability during joint and combined operations, and coexistence with commercial and civil systems. The US DARPA XG (neXt Generation) programme exists to develop this technology.

DARPA XG

DARPA awarded Shared Spectrum Company (SSC) contracts to develop XG, and in 2006 SSC successfully demonstrated XG radios at Fort A.P. Hill, Virginia. The demonstration used six mobile XG radios that operated in the same spectrum as a suite of fixed, instrumented military and commercial legacy radios. A wide-area instrumentation system was used to record the XG radio connectivity and the performance of the legacy radios.

The field exercises demonstrated the operational utility of XG: that XG causes no harm to existing military radios in compliance with emission/regulatory rules; XG will allow additional radio networks or communication capacity then currently possible using existing procedures; and that XG can operate in the presence of electromagnetic interference (i.e. jamming).

Cognitive radio is some way from deployment, and technical issues remain as well as regulatory ones. Conventionally, the radio spectrum is divided up by national governments and allocated to various uses, so a system that arbitrarily makes use of any available bandwidth would be illegal almost everywhere.