Page 1 of 1
Interest in software defined radio continues unabated. The multiple programs of the U.S. Military’s Joint Tactical Radio System for software radio are progressing through their respective development and testing phases. Believing in its future, companies are investing in hardware and software products designed to allow newer and better software radio products—both military and commercial—to come to market faster and easier. This trend should continue in 2005.
Many software defined radio (SDR) efforts are focused on two hardware areas. FPGAs are foremost, with a number of firms creating systems-on-a-chip (SoCs) by combining their FPGAs with other functionality—small, powerful soft-core microprocessors or other intellectual property (IP) running on the FPGA fabric—to provide more powerful processors. These are designed to replace general-purpose processors (GPPs) or DSPs. Other companies are focusing on enhancing systems or subsystems for rapid prototyping or on providing more integrated components for inclusion in systems.
What’s New in SDR Hardware?
FPGAs, and the many ways they can be used in SDR, continue to attract the most interest (see accompanying articles by ICS on page 32, Altera on page 38 and Lattice Semiconductor on page 46). Echoing this trend, the SDR Forum recently announced a workshop on new chip architectures for software radio to be held at their January meeting.
Recently, Xilinx and iSR Technologies (ISR) demonstrated the use of a commercially available FPGA to run multiple Software Communications Architecture-compliant waveforms on a single chip. This architecture supports dynamic sharing of radio resources through partial reconfiguration—the ability to dynamically create or tear down an application (one waveform) in one area of the FPGA while simultaneously running another application (another waveform) in another. This architecture should lead to a much lower number of components, less board space and less power consumption, resulting in lower cost.
Meanwhile, Spectrum Signal Processing announced “the industry’s first integrated rapid-prototyping MILCOM platform”, the flexComm SDR-3000 MRDP (Figure 1), based on the Joint Tactical Radio System (JTRS) Software Communications Architecture (SCA). They claim the SDR-3000 to be the industry’s first “RF to Ethernet” COTS solution specifically targeting military communications programs.
In subsystems, Pentek introduced the Model 6822, the first VME board to combine VXS I/O with dual 215 MHz ADCs and two Xilinx Virtex II Pro FPGAs for on-board signal processing. VXS is the switched serial backplane fabric for VMEbus and this VXS interface provides two 1.25 Gbyte/s switched serial fabric ports to the VME backplane.
And Interactive Circuits and Systems (ICS) announced the PMC571 (Figure 2), an SDR mezzanine available in five different ruggedization levels. ICS believes the PMC571 is the first rugged PMC to offer wide bandwidth ADC and DAC conversion at software radio frequencies. The PMC 571 features a four-million-gate FPGA, is compatible with VMEbus PowerPC and CompactPCI SBCs and can use the Xilinx Virtex II software development environment.
Looking further out, Vanu received a development contract from the U.S. Army Communications-Electronics Command Research, Development and Engineering Center (CERDEC) to build a prototype mobile GSM cellular communications system for secure, rapid, field-deployable applications. It will demonstrate use of encrypted GSM handsets, supported by a vehicle-mounted Vanu Software Radio Base Station, to provide communications between dismounted soldiers. And HYPRES, developer of Superconducting MicroElectronics (SME), was awarded contracts by CERDEC to develop ADCs and direct digital synthesizers for JTRS using SME technology that will address current obstacles to size, weight and power consumption.
On the Software Side
Software defined radio will always emphasize software—to replace traditional hardware, to allow systems to reconfigure quickly or adapt them to new challenges, and to provide development tools. Among more recent important announcements were several JTRS SCA implementations, as well as the roll out of software tools to aid in the development of SDR. It’s reasonable to assume that these are not the last of a stream of similar products.
PrismTech introduced the first COTS implementation of the SCA, providing a pre-integrated and optimized product that addresses the SCA’s RTOS, ORB and Core Framework (CF) layers. Called OpenFusion SCA OE, PrismTech’s product is the first to provide a complete operating environment (OE) supporting the development and deployment of advanced SCA-compliant waveform applications for SDR in a military radio, satellite communications, commercial wireless infrastructure and public safety.
Taking a different approach, Virginia Tech’s Mobile and Portable Radio Research Group (MPRG) released an open-source C++ implementation of the SCA. Attempting to lower the barriers to entry into the software radio research arena in general and the JTRS community in particular, MPRG is providing a software framework that is free, easy to use and written in C++, a language known to most wireless developers. This open source version of the SCA is called OSSIE (Open-Source SCA Implementation::Embedded).
In tools, the Communications Research Centre Canada (CRC) announced its SCARI software suite to help manufacturers accelerate the roll out of SDR technology. SCARI provides an easy-to-use blueprint for designing and implementing SDR, helping to ensure interoperability between SDR devices from different manufacturers. It includes several platform-independent software tools to simplify the SDR development life cycle, from the creation of SCA components to assembly in applications or nodes to debugging and operation of the radio.
And Zeligsoft CE (Component Enabler)—formerly Waveform Builder—allows developers to model, configure and validate SCA-compliant radio platforms and waveform applications through a UML 2.0 interface. CE then automatically generates and validates the complete set of SCA-compliant XML descriptor files—reducing development time from months to days.
On the Commercial Side
Not everything is for the military. In the commercial space, Vanu’s Software Radio GSM Base Station became the first device to successfully complete the FCC’s certification process governing software radio devices, after the FCC’s recent recognition of SDR as a new category. These radios afford greater flexibility than traditional ones, offer wireless network operators the ability to lower capital and operating expenditures and allow more efficient use of the spectrum. Certification by the FCC opens the way for commercial sales in the U.S. of Vanu’s Software Radio base station.
But for SDR, a key question is, “Does the market size justify all this development
effort?” Steve Jenkins of PrismTech suggests that a prime SDR objective
should be to rapidly leverage military technology into the commercial field so
that the SCA does not become an orphan like the Ada language. His thesis is, while
the current primary market for SCA is the JTRS and related military market, SCA
is not specific to SDR and can be much broader than JTRS. In addition to the military—SCA’s
early adopter—he sees applicability for the SCA in 3G cellular base stations,
handsets and wireless networking.
Jenkins also believes that the short-term focus should be on lowering costs and removing obstacles, while self-optimizing or cognitive radio systems can wait. However, not everyone agrees. The SDR Forum just announced creation of a Cognitive Applications Special Interest Group and a Cognitive Radio Working Group to leverage SDR in the development of cognitive radio technology.
Ottawa, Ontario, Canada.
Digital Receiver Technology
Interactive Circuits and Systems
Ottawa, Ontario, Canada.
Montreal, Quebec, Canada.
JTRS Joint Program Office
Spectrum Signal Processing
Burnaby, BC, Canada.
Virginia Tech MPRG
San Jose, CA.
Gatineau, Quebec, Canada.
Nice article, Dave. I just shared it on Twitter with a small but interested audience. Regards, John