[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

Re: DSP Possibilities ?

I'm off by a factor of 2 in the bandwidth required. However, the codec
sampling rate will still limit the acheivable data rate. Given a fixed
bandwidth, as the modulation scheme becomes more complex, the intersymbol
interference increases and the window in which data can be sampled
decreases. This would require the demodulator sampling rate to be increased
to in order to reduce the timing error.

I've seen schemes described where the incoming signal is decimated down to
the minimum sampling rate allowed by the Nyquist limit, filtered, and then
interpolated before sampling the data, but this would seem to cause
distortion which would increase the required SNR. Since we are power
limited, we don't want to do anything that compromizes performance at low

The last DSP code that I wrote was in 1974 for filtering accelerometer data.
When I get my digital IF card assembled I will be able see what effect these
things have on real PACSAT signals.



----- Original Message -----
From: "Howie DeFelice" <hdefelice@petrexintl.com>
To: "'John Stephensen, KD6OZH'" <kd6ozh@AMSAT.Org>
Cc: <amsat-bb@AMSAT.Org>
Sent: Thursday, 21 September 2000 20:20 UTC
Subject: RE: [amsat-bb] DSP Possibilities ?

> I think your bits/hz calculations are with FEC. Uncoded QPSK would produce
> 2 bits/Hz, 8PSK without coding is 3 bits/Hz and 16QAM is 4 bits/Hz. This
> would reduce the transmitted symbol rate to the data rate divided by the
> number of bits/Hz. Of course, I can't imagine using these schemes on
> amateur satellites (or any other satellites for that fact) without some
> FEC.
>  I could be wrong, but I don't think changing the modulation effects the
> sample rate for the modulator, but it does increase the work load from a
> number crunching view. On the other hand, the demodulator's sampling rate
> should be lower due to the decreased symbol rate.  I would imagine that
> number crunching inside the demod would also increase. The net sum result
> is that if you take some of the burden off the sample rate, you use up the
> processing power somewhere else, and it's most likely NOT a linear
> tradeoff...
> Howie
> AB2S

Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org