I have one doubt regarding how to analyze a constellation diagram. It is from a QPSK transmitter. The constellation diagram looks typically as books (as a square as you can see in Figure 1), but for shorts periods it changes as you can see in Figure 2. We don't know if it is a problem with the transmitter, a problem with the data that is being modulated, or if it is a bad configuration on the Spectrum Analyzer. What do you think about that? How can we interpret the constellation diagram and the transitions? Thanks in advance.

UPDATE: 27/02/2018 Please see Figure 3 that shows how we set up of the Vector Analyzer.

constellation diagram during most of the time (Figure 1)

Strange transitions for short times (Figure 2)

Setup of the Vector Analyzer

  • \$\begingroup\$ Symbol rate and spectrum analyzer sample rate? \$\endgroup\$ – alex.forencich Jan 31 '18 at 19:56

Note how the second image looks kinda-sorta like two rotated copies of the top image superimposed.

What has happened is that the receiver has temporarily lost synchronization with the signal. In fact, it looks like it is off by exactly ±22.5°, which suggests that there might be a problem with the phase detector specifically.

BTW, "spectrum analyzer" is not the correct term here — that would imply some sort of frequency-domain display. This is a time-domain display, with many cycles superimposed.

| improve this answer | |
  • \$\begingroup\$ Yes, the receiver PLL shifted, causing the phase shift in the constellation. You can see in the lower right of the following video how the received QPSK constellation will "spin" when the receiver PLL is not locked. In your case, it appears that the receiver PLL shifted, but then remained locked, or, it is jumping between -22.5, 0, and +22.5. youtu.be/bfqJHVmOGn0 \$\endgroup\$ – TimB Jan 31 '18 at 23:29
  • \$\begingroup\$ Perhaps a better term for this instrument would be a vector analyzer? \$\endgroup\$ – duskwuff -inactive- Feb 2 '18 at 5:48
  • \$\begingroup\$ Thank you for the explanation. It makes a lot of sense now. I see the "rotated copies". I still have one more doubt. When you refer to a problem in the phase detector o in the receiver PLL, are you referring to a problem in the Vector Analyzer? Could it be a problem in the Hardware or the setup? (I have updated the post including the setup of the Vector Analyzer during measurements). Thanks again. \$\endgroup\$ – Gabriel Feb 27 '18 at 20:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.