There has been a push in some quarters towards setting standards for Software-Defined Radio (SDR) based on a standard hardware architecture that can support a variety of software to support different RF communications techniques. The SDR Forum has long been a primary proponent of these standards, but the realization of systems built on these standards has been long in coming. The military has actively pushed the standards for a number of new radio system acquisitions, but the commercial world has to a large extent ignored the concept.

This may be changing. An EE Times report from the Software Defined Radio conference described commercial implementations of cellular base stations based on SDR technology. These base stations would be able to adapt to frequencies and protocols even after they are deployed. This is a long way from the ideal SDR implementation, which would be able to load wireless protocol software arbitrarily and practically immediately, but it is a step in the right direction. The cost savings for carriers would be significant, which would certainly be good PR for the SDR concept.

The biggest problems are on the software side, though. The later pages of the article talk about the Software Communications Architecture, which defines the APIs to be used for SDR. This standard has had a number of problems, including the 3.0 version being considered to be beyond the capabilities of the DSP and FPGA front ends to implement.

The best thing for SDR might just be if control of the standard were taken out of military hands and given to the SDR Forum. This would at least separate it from some of the politics that currently impedes its progress.

Larry Mittag