0
$\begingroup$

I have a question about the data rate for a DSSS transmission in amateur radio. I give an example with a bandwidth $B=4$ kHz, a spreading factor $SF=64$, a DQPSK modulation and a convolutive code with coding rate $R_c=\frac{1} {2}$. The data rate is: $D \simeq 63$ bps.

My questions:

  1. Is the data rate obtained interesting for amateur radio?

  2. If we take into account the preamble and the possible pilot symbol for Doppler estimation, what is an acceptable data rate for DSSS in amateur radio?

$\endgroup$

1 Answer 1

1
$\begingroup$

Is the data rate obtained interesting for amateur radio?

That is impossible to answer. With a 10 W transmitter over 100m? Certainly not for me. Dangerously close to Shannon capacity by using a few microwatt over kilometers? Maybe!

Interesting is a human category. There's no objective answer to "is it interesting", but it's fair to point out that if you invent a new mode, it would make sense to make sure it's technically not "stupid", but gets reasonably close to what is possible over the range of channels you're planning to use it with.

If we take into account the preamble and the possible pilot symbol for Doppler estimation, what is an acceptable data rate for DSSS in amateur radio?

See above. Your question has no answer, because "amateur radio" doesn't describe a single use case, nor a channel that you "fight against". Some people want to talk through their radios – then 64 bit/s is too little. Other people will want to send text messages; then it might suffice, but one would be very justified to ask why you're using a 4 kHz wide channel.

a convolutive code with coding rate $R_c=1/2$.

This might belong in the "don't be technically much inferior to state-of-the-art" category: There's technically no reason any more for short messages (and you're not going to have long messages at that rate) being encoded with a convolutional code, unless you're going a more modern route and are thinking about Tail-Biting Convolutional Codes (TBCCs), which are technically block codes. Every microcontroller you can buy that could reasonably be interfaced with a receiver could decode a short block code in a soft fashion, reasonably well.

So, I think you should definitely investigate what you're doing further – but maybe start with an application that you design a system for, and not the other way around.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .