Re: [Discuss-gnuradio] Decoding constellation (0, 1-1) using gnuradio

[Marcus Müller]

The point is that what you're effectively doing is using the front 11 as a kind of preamble, and 01 or 10 being the symbol you send.
That doesn't sound like much. To be honest, PPM in an environment like that without lots of redundancy simply sounds unstable -- you just miss one of your preamble symbols, and your whole PPM symbol will be lost. You just misinterpret two "no-signal" 0s, and get a preamble.

Make your contiguous packets longer. The average PPM transceiver -- old garage door openers etc -- would use something  10s of symbol durations worth of preamble, and then 10s of repititions of the symbol. PPM isn't very popular these days, because with the same transmitter energy you can usually get better performance, as it doesn't work very well on the typical radio channels: It's sensitive to multipath (which is pretty obvious) or non-flat fading (which, is, for many practical aspects, usually equivalent to multipath), and timing recovery is unnecessarily hard if you don't have a lot of symbols.

So, maybe we should take a step back and ask: *what* is the *data* you're trying to transmit? Transmitting a single bit at a time sounds so unlikely.

Best regards,
Marcus

On 11/11/2015 08:26 PM, abhinav narain wrote:

Hi Marcus,

Will be really great if you could look at the last part of my mail.

My specific questions is -

Lets say I transmit two PPM frames... 1100 and 1101

say: 1100000000001101

The number of non-bold zeros are what I am filling in between the information frames at transmitter.

But at receiver, I get more number of non-bold zeroes than what I expect(=8).

Is this something that I cannot solve because of clock-drift/synchronization, or is my flow graph incorrect causing this ?

Thanks,

Abhinav

On Tue, Nov 10, 2015 at 12:58 PM, abhinav narain <address@hidden> wrote:

On Tue, Nov 10, 2015 at 12:45 PM, Marcus Müller <address@hidden> wrote:

Hi Abhinav,

sorry, I might just be tired right now, but I don't understand this
sentence:

On 10.11.2015 21:18, abhinav narain wrote:
> I have now fallen to doing PPM where I map {0,1} bits to {101,11}
> symbols on the transmitter side, where 0 in 101 is equivalent to x 1x1
> as I don't transmit anything in that slot too.

 

I'd expect Pulse Position Modulation symbols to have the same length,
but with the non-zero element being at a different position; maybe I'm
just misunderstanding?

Yes, sorry - lets say 1010 and 1100 as the two  PPM codes.

Thanks,

Abhinav

 

Best regards,
Marcus