Since the mid-1990s, software-defined radio, or SDR, has been receiving considerable attention because of its potential for addressing a variety of communications needs both for today and tomorrow.

While the military's Joint Tactical Radio Systems, or JTRS, program traditionally has spearheaded interest in SDR, more recently the public-safety community's attention to this technology has increased dramatically. For example, the SDR Forum's Public Safety Special Interest Group, or PSSIG, has been conducting a comprehensive investigation of the technology's potential for meeting present and future public-safety communications requirements. Among the items being examined are interoperability and radio/network requirements derived from Project SAFECOM and Project MESA Statements of Requirements (SORs).

This article discusses the challenges and tradeoffs in realizing SDR's potential benefits for public safety. This is not intended to represent the consensus opinion of the PSSIG, of which the author is an active member, because the PSSIG's analysis of SDR's benefits is ongoing. (The PSSIG plans to release later this year a report detailing the results of this analysis.)

Not an interoperability panacea

In both the public-safety and military communities, SDR usually is mentioned as a potential solution for interoperability because of its ability to achieve flexible, multiple modulations, modes and frequency bands with a single hardware platform. However, to achieve ubiquitous interoperability, public safety must address a three-pronged challenge (Figure 1). SDR technology cannot reach its full potential for supporting interoperability unless solid management and regulatory structures also are in place.

Probably the greatest challenge to interoperability relates to people, not technology. Management barriers to interoperability include funding, operational policies, institutional barriers and getting the nation's approximately 60,000 first-responder agencies to work together for a common solution. Also, regulatory issues are important and include user licensing for interoperability scenarios, radio certification and security of over-the-air software downloads.

SDR provides support for the technical leg depicted in Figure 1, which is a composite of two ingredients: the radio's capabilities and the infrastructure/network over which the radio communicates. Limited direct radio-to-radio interoperability is possible today using talk-around operation with analog-FM or Association of Public-Safety Communications Officials Project 25-compliant conventional modes.

Also, the latest-generation public-safety radios offer software control of multiple protocols and dual-band operation. However, to achieve all categories of interoperability (day-to-day, task force and mutual aid) and to address the Project SAFECOM and Project MESA SORs, the infrastructure/network aspect of the technical solution is equally, if not more, important.

SDR gains traction

Radio manufacturers already are using, and will continue to use, the latest SDR technologies that make sense, given cost and functionality tradeoffs. Today's latest generation of public-safety radios — such as M/A-COM's M/P7100 and M/P7200 series — allow operating modes, modulations, radio features and the radio's “personality” (e.g., specific site frequencies, permissible modes of operation, groups, etc.) to be changed via software without modifications to the hardware platform.

Digital hardware has much greater precision and repeatability than analog circuitry, so the substitution of digital for analog circuitry in public-safety radio designs has reduced the need for costly recalibrations of the radio. Also, the calibration is typically built into the software and has the potential to be completely automated.

Consideration also is being devoted to special SDR (as well as cognitive radio) applications that might be judiciously incorporated into the radio to assist the user, such as voice command capability. However, one must resist the urge to get carried away by relinquishing too much operator control to the software because this actually could complicate the operation of the radio and could jeopardize first-responder safety should the software lack sufficient robustness. As just one example, a speech-recognition algorithm that can be voice commanded and which would be used to declare an emergency must be extremely reliable.

Regarding specific SDR devices for public-safety radios, major technology enablers include the following:

  • faster, lower power and denser digital processors and memory;
  • faster, lower power and higher dynamic range A/D and D/A converters;
  • wideband radio front ends and amplifiers;
  • multiband antennas.

Moore's Law-like advances in digital processors and memory have been the most influential factor in the proliferation of new SDR architectures. The processors can be implemented individually or used in combinations of numerous devices, including digital signal processors (DSPs), general purpose processors, configurable computing machines, field programmable gate arrays and application-specific integrated circuits.

For today's public-safety portables, size, weight, cost, performance and battery life tradeoffs tend to favor state-of-the-art, fast, low-cost, low-power DSPs for performing high-speed processing, typically in combination with a general purpose processor for slower radio control and user/external interface functions.

The placement of the A/D and D/A converters is a delicate balancing act in the design of a public-safety radio (Figure 2). The closer these converters are placed to the antenna, the more SDR-like the radio becomes, with the associated advantages of increased signal precision, repeatability and, perhaps, the digital enabling of multiple frequency bands without replication of analog circuitry.

However, these advantages must be weighed against cost, reduced dynamic range and reduced battery life associated with the higher-speed A/D and D/A devices that would be required. For example, to meet stringent public-safety specifications on adjacent channel rejection, the dynamic range of the A/D converter must be sufficient to simultaneously accommodate weak on-channel signals and strong off-channel interference — as much as 70 dB stronger. Available higher sampling rate A/D converters tend to have lower dynamic range capability.

As such, A/Ds and D/As are probably the most significant factor (at least for a portable implementation) limiting the progression toward the ultimate software radio, which would have the converters attached to the antenna (the leftmost placement depicted in Figure 2), so the entire radio would be digital. Although some improvements in size, weight, cost, sampling rate, dynamic range and power consumption have occurred for A/Ds and D/As in the past several years, today's SDR sampling devices simply do not support the requisite combination of dynamic range, speed, low power consumption (for a portable) and cost for anything other than second intermediate frequency or baseband sampling implementations.

The gain-bandwidth product of amplifiers and the antenna at the radio frequency (RF) front end also will need to improve greatly for efficient operation over multiple bands in one radio. For portable operation, the combination of numerous RF bands into one antenna must be accomplished in a small form factor, which traditionally has lowered the antenna's efficiency and thus reduced the battery life. There is hope that software-controlled microelectromechanical devices will facilitate the requisite gains in efficiency for such implementations.

SDR equals better, not always cheaper

Given the stringent requirements placed on public-safety radios, SDR will not likely become a magic pill that will reduce the acquisition cost of a public-safety radio to that of an inexpensive cell phone. For starters, there are many cost drivers in a typical public-safety radio for which SDR technology offers minimal benefit, including stringent environmental/ruggedness specs, higher reliability requirements, more stringent audio and RF specifications, and higher transmit power.

Also, as stated previously, public-safety radio manufacturers already have been using SDR technologies in their radios to the maximum extent that makes sense from cost, size, weight, battery life and performance tradeoffs.

There are other requirements that have a major influence on cost (as well as power consumption), including:

  • Dynamic range: It is driven by stringent public-safety adjacent channel rejection and intermod rejection requirements, which creates a need for A/D converters with many bits.

  • Transmit spectral purity requirements: They place substantial demands on the transmit modulator (including D/A converters), power amplifier, frequency synthesizer design, and transmit filtering.

  • Stringent receive/adjacent channel rejection requirements: They drive highly selective receive filtering as well as high dynamic range.

  • Total occupied bandwidth: In combination with the high dynamic range requirements, this is one of the most important drivers of cost as well as power consumption. Multi-band radios require multi-band antennas, wider bandwidth amplifiers, and either high dynamic range, high-speed digitizers or replication of analog circuitry.

  • Multimode baseband processing: Although a cost driver, this is less so than the above factors, since its cost (as well as power consumption) has received the most benefit from Moore's Law-like increases in speed and density.

  • Amount of software overhead required: Software architectures must be extremely efficient, and must exhibit low latency to meet public safety's demands for extremely fast radio response times and low power consumption.

In summary, utilization of SDR technologies already is progressing naturally as a result of the cost/functionality tradeoff optimization process used by public-safety radio manufacturers. But several tradeoffs still exist concerning SDR technologies, with the most crucial being those that tend to have the most impact on cost and power consumption. In addition, while SDR can provide substantial technical benefits for interoperability, other non-technical issues must also be addressed before interoperability can be fully supported.

Richard Taylor is a senior principal engineer at M/A-COM and has 31 years engineering experience in radar and land mobile radio systems. Taylor is a regular attendee of the SDR Forum and has given presentations related to SDR for APCO and the IEEE. He has a bachelor's degree in Electrical Engineering from Michigan State University and a master's degree in Electrical Engineering from Syracuse University.

on the web:

For related coverage of the PSSIG's recent activities, see the story “The promise and peril of SDR for first responders,” by Doug Mohney at