[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Discuss-gnuradio] Draft letter to Cahners re GNU Radio

From: John Gilmore
Subject: [Discuss-gnuradio] Draft letter to Cahners re GNU Radio
Date: Sun, 13 Jan 2002 00:30:21 -0800

When we see press mentions like the one Bob Kelley found, we should
send them a letter-to-the-editor alerting them to our project.  This
will get us free publicity, probably leading to more users and
contributors.  What's below is too long to be printed verbatim, but is
offered as a start.  The trade press is used to getting such letters
from companies that have a product, but not from free software groups.


To: address@hidden, gnu
Subject: Free Software Defined Radios
I enjoyed reading your January 2002 survey "Software-Defined Radios
Enter the Limelight".  I was disappointed to notice that your
"Web-based resources" section didn't mention the Free Software
Foundation's SDR project, called "GNU Radio".  The Foundation 
created or catalyzed much of the software now known as the popular Linux
operating system.  We hope to have a similar effect on the software
infrastructure for building software-defined radios, enabling broad
research, development, experimentation, and production of SDRs, by
anyone from single individuals to major corporations.

GNU Radio is written in C++ and will run in any ordinary PC.  (It is
designed for easy porting to workstations and embedded processors as
well).  The software is freely available for public use, distribution,
and modification, under the standard GNU General Public License.  It
can make productive use of multiple CPUs and signal processing
instruction set extensions.  It is based on the free "PSpectra"
research software from MIT.

GNU Radio consists of a software backplane that allocates memory
buffers and processor time, and a set of individual DSP modules which
consume, process, and produce data samples.  For example, a tuner
module would accept samples of a wide band, and process it to extract
and output only the particular frequency band of interest.  An FM
decoder module might accept baseband samples, decode them, and produce
samples holding the modulating signal (e.g. audio).  An FM encoder
module would do the inverse.

The software currently works with off-the-shelf PCI hardware that
digitizes analog inputs at 20MHz with 12-bit samples.  It runs on
Linux.  Engineers are encouraged to port it to other hardware,
other Unix-like operating systems, and to write new decoder and
encoder modules for it.

Full details are available at the project's web site:



   Eric or me or somebody

reply via email to

[Prev in Thread] Current Thread [Next in Thread]