Hello everybody! I'm using a Barker sequence to achieve communication synchronization and a first AGC and phase detection. For the moment I use a classic 13 symbols long Barker sequence: [+1 +1 +1 +1 +1 −1 −1 +1 +1 −1 +1 −1 +1] The thing is that I use a QPSK modulation and I was wondering if this was the most efficient way or if there were Barker sequences for complex alphabets. Something like [1 -i i i 1 -1 1 1 i] (this is a random sequence). Any help or document/lead would be appreciated! _____________________________ Posted through www.DSPRelated.com
Barker Sequence for QPSK
Started by ●July 16, 2013
Reply by ●July 16, 20132013-07-16
On 7/16/2013 10:44 AM, Gracchus wrote:> Hello everybody! > > I'm using a Barker sequence to achieve communication synchronization and a > first AGC and phase detection.> For the moment I use a classic 13 symbols long Barker sequence: > [+1 +1 +1 +1 +1 тИТ1 тИТ1 +1 +1 тИТ1 +1 тИТ1 +1] > > The thing is that I use a QPSK modulation and I was wondering if this was > the most efficient way or if there were Barker sequences for complex > alphabets. Something like [1 -i i i 1 -1 1 1 i] (this is a random > sequence). > Any help or document/lead would be appreciated!P. Fan, M. Darnell. “Sequence design for communications applications” John Wiley & Sons ISBN 0-471-96557
Reply by ●July 16, 20132013-07-16
>Hello everybody! > >I'm using a Barker sequence to achieve communication synchronization anda>first AGC and phase detection. > >For the moment I use a classic 13 symbols long Barker sequence: >[+1 +1 +1 +1 +1 −1 −1 +1 +1 −1 +1 −1 +1] > >The thing is that I use a QPSK modulation and I was wondering if this was >the most efficient way or if there were Barker sequences for complex >alphabets. Something like [1 -i i i 1 -1 1 1 i] (this is a random >sequence). > >Any help or document/lead would be appreciated! > >_____________________________ >Posted through www.DSPRelated.com >I would just setup a Matlab script that generates a random sequence, then test it for autocorrelation, balance, and spectral flatness, and save off the best sequences you get - put that in a loop and run it over the weekend to get a good sequence to use. If it only has to be 13 symbols long, then you can probably check each and every sequence in a reasonable amount of time. -Doug _____________________________ Posted through www.DSPRelated.com
Reply by ●July 16, 20132013-07-16
On Tue, 16 Jul 2013 10:44:27 -0500, "Gracchus" <98125@dsprelated> wrote:>Hello everybody! > >I'm using a Barker sequence to achieve communication synchronization and a >first AGC and phase detection. > >For the moment I use a classic 13 symbols long Barker sequence: >[+1 +1 +1 +1 +1 −1 −1 +1 +1 −1 +1 −1 +1] > >The thing is that I use a QPSK modulation and I was wondering if this was >the most efficient way or if there were Barker sequences for complex >alphabets. Something like [1 -i i i 1 -1 1 1 i] (this is a random >sequence). > >Any help or document/lead would be appreciated! > >_____________________________ >Posted through www.DSPRelated.comIn practice what is often done is that the preamble/Barker/UW is transmitted in BPSK, regardless of the subsequent modulation type. This provides the best likelihood of detection/synchronization but at the expense of some additional symbols allocated to the UW. This also preserves the autocorrelation properties of the orginal sequence, whether it's a Barker code or some other flavor of UW. The easiest way to transmit BPSK with a QPSK modulator is to put the same sequence in both the I and Q channels. That makes good use of the available transmit power with the minimum fuss in the modulator logic. If you can't do that for whatever reason, then picking the best way to modulate the QPSK waveform to get good autocorrelation can be determined in simulation (as previously suggested). Eric Jacobsen Anchor Hill Communications http://www.anchorhill.com
Reply by ●July 17, 20132013-07-17
On Tuesday, July 16, 2013 11:44:27 AM UTC-4, Gracchus wrote:> Hello everybody! > > > > I'm using a Barker sequence to achieve communication synchronization and a > > first AGC and phase detection. > > > > For the moment I use a classic 13 symbols long Barker sequence: > > [+1 +1 +1 +1 +1 -1 -1 +1 +1 -1 +1 -1 +1] > > > > The thing is that I use a QPSK modulation and I was wondering if this was > > the most efficient way or if there were Barker sequences for complex > > alphabets. Something like [1 -i i i 1 -1 1 1 i] (this is a random > > sequence). > > > > Any help or document/lead would be appreciated! > > > > _____________________________ > > Posted through www.DSPRelated.comPerhaps complementary code keying (CCK) would be useful to you. Clay
