[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [linrad] Linrad, some questions
[mailto:owner-linrad@xxxxxxxxxxxxxxx]On Behalf Of Larry Kayser
Sent: 09 September 2002 00:45
Subject: [linrad] Linrad, some questions
Greetings all on Linrad:
I am a Linrad potential newbie. I have no interest in Linux by itself and
I am interested in Linrad for what it can do for a project I am working on
- I am trying to limit my interest in Linrad itself, ie trying to hold a
focus on my project. Not easy it seems.
I have some questions about Linrad.
My interest is to use Linrad as part of a tool box of software resources
that I can use to examine weak signals, very weak signals. I would like to
examine .wav files using Linrad as well as other FFT based tools with a
bandwidth of up to 20 KHz, + and - 10 KHz. For most of the time I would be
looking at the data some time after it has been recorded and other
resources have discarded up to 98% of the data that has been collected.
If I have it right, Linrad can look at .WAV files, but I have no idea if
the Linrad SDR can be used to search through a wide band recording and look
for weak signals over the whole bandwidth or does one loose detectability
through the recording process? My sense is that this should be possible if
the data is recorded at the full bandwidth and not constrained in some way
by the Linrad software.
Next, what might the impact be of transcoding from .WAV to MP3 and then
transcoding back to .WAV from MP3 and then using Linrad to look at the
data? What might the transcoding degradation be and how would it manifest
If anyone out there in Linradland has done any work with wider bandwidth
applications I would appreciate some thoughtful discussion in this area.
Larry I think that the MP3 algorithms will be very lossy in this
application, it would be far better to avoid any kind of data compression.
MP3 algorithms were designed for music applications and the algorithms
decide what is valid and what is not, if you try to listen to a MP3 encoded
classical music piece or anything with dynamics and compare it to even a 16
bit PCM original file you will hear that the algorithm gets it very wrong.
The artefacts are obvious even to the untrained ear. These errors will be
cumulative each time the process occurs.