A convergence of technology trends will change the operational and design characteristics of modern radios, increasing certification challenges. First is software defined radio (SDR), which allows much of what was done on predefined hardware, such as signal processing, to be programmed in software. Second are devices that can change how and where they operate within the radio spectrum, moving among a set of frequency bands in response to interference or other constraints. Third is cognitive radio technology, where the device can autonomously make decisions about its operation, in response to environmental changes, such as interference. To vastly oversimplify, such capabilities enable SDRs to vary waveforms and frequencies within their hardware constraints, as well as to reconfigure the operation of their network protocols.
Although many features today are more a hope than a reality, progress has been made. The Pentagon is pouring billions into its Joint Tactical Radio System program. The Defense Advanced Research Projects Agency (DARPA), as part of the military's commitment to SDRs, is funding a project called neXt Generation (XG) communications that aims to combine SDR, cognition and frequency agility in a reasonably priced radio. In the private sector the U.S. Federal Communications Commission (FCC) has certified Vanu Inc.'s SDR-based GSM (mobile phone) base station product. And the Institute of Electrical and Electronics Engineers has formed a standards committee (IEEE P1900.X) dedicated to next-generation radio and spectrum management.
But what guarantees the good behavior of these chameleons? How will governments regulate radios that can be reconfigured over the airwaves? Until spectrum management catches up with technology, how can regulators make sure that SDRs stay within their assigned spectrum; don't interfere with safety of life aviation, public safety or other allocated areas; and are not capable of being hijacked by hostile parties?
In its most recent SDR Order, FCC defines an SDR this way: "a radio that includes a transmitter in which the operating parameters of frequency, range, modulation type or maximum output power (either transmitted or conducted), or the circumstances under which the transmitter operates in accordance with Commission rules, can be altered by making a change in software without making any changes to hardware components that affect the radio frequency emissions." The italicized phrase is critical to dealing with the potential dynamic network protocol operation of such radios. Of course, FCC also requires that manufacturers implement security techniques to prevent users (or hackers) from modifying parameters, such as operating frequencies, output power and modulation types.
Radio system design to date has made it easy to contain misbehavior through certification, fault isolation and legal action. The devices are managed by certification to strict operational constraints (to prevent interference) that are backed by the force of law. Furthermore, operators and government authorities have some ability to isolate misbehaving radios.
The difficult challenge in certifying SDR-based devices stems from their potential to reconfigure operational characteristics and the subsequent use of the spectrum. The distinction that makes this different from telephones and mobiles is that those devices have severely limited transmission and operational capabilities, whereas SDRs are being designed for extreme flexibility. This flexibility strains the ability of FCC's traditional certification models, as any deviation is liable to cause local interference and potential network-wide cascade issues.
Many issues can be handled by limiting the physical components to specific frequencies or restricting their ability to transmit at certain frequencies. Evaluation and assurance models used by companies in trusted computing and safety services can also provide a sliding scale of assurance from nonexistent to the formal levels required for managing top secret documents or human life. This can reduce the certification burdens on non-critical devices, such as those for personal area networks (e.g., Bluetooth).
In certain bands it is conceivable that new services might be introduced through the use of interference-avoiding SDR technology. This may not be true for other bands, such as radio astronomy and aeronautical communications, where a higher level of non-interference assurance is required. If SDR-based devices are designed to communicate in these potentially off-limit bands, they will also need to undergo equivalent evaluations to preclude reciprocal interference. In closing, we believe that SDR offers great opportunities for improving the use of the radio spectrum and that with the proper precautionary steps we can ensure its introduction consistent with the needs of the military, industry and public safety.
John Giacomoni is a graduate student and Douglas Sicker is an assistant professor in the Computer Science Department at the University of Colorado at Boulder.