Re: [Discuss-gnuradio] BPSK transmitter - time domain waveform

[Tom Rondeau]

On Thu, Apr 5, 2012 at 9:33 AM, senthil murugan
<address@hidden> wrote:
> Hi all,
>
> I have some doubt on the time domain view of bpsk signal generated in
> Gnuradio and usrp. I used psk mod block in GRC (sps =2) and created bpsk
> (baseband )waveform with vector source producing 0,1,1 continuously.
>
> When I plotted the output of psk mod block using scope sink, I seen waveform
> of  constant 1 for some samples (for 0) , inverted root-raised cosine
> waveform (for 1) and another inverted root-raised cosine waveform (for 1).
> This is repeated continuously and i understand that it because of transmit
> pulse shaping.
>
> Now, the question is when I give this signal to USRP for transmission, how
> the time-domain view of that waveform look like?  Will it look like sinusoid
> signal with 180deg phase shift as we all seen in stand. communication text
> books? I guess it will not the case. I tried  multiplying  the baseband
> waveform with signal source (complex) -- mathematical equivalent of what
> happening in usrp -- and seen waveform. No smooth shift of 180deg.
>
> If anyone having usrp and digital oscilloscope, could you please send me the
> pic of time domain view of output of usrp?
>
> Thanks
> Senthil

If you want to see this in simulation, make sure that you are
upsampling your signal before doing the frequency conversion so your
sample rate can support the new frequency range you are interested in
seeing.

The answer is yes, the resulting waveform will be a sinusoid at the
carrier frequency with a lower frequency BPSK envelope. If you are
doing this over the air, remember that if the oscillators in your
transmitter and receiver are not locked together, there will be a
frequency offset between them.

Tom