Do you know of effective and simple methods to compensate for Doppler due to movement for the CDMA protocol?

  • $\begingroup$ There's no "the CDMA protocol"; CDMA is a multiuser access scheme, not a specific protocol! $\endgroup$ – Marcus Müller Aug 6 '18 at 20:36

I sure wish I had a universal method for that that is any smarter, but:

Unless you have enough SNR and computational power so that you can treat the chips as symbols of a higher rate single-carrier system and do the usual frequency correction as if it was just a normal single carrier system (e.g. higher-order phase control loops), you'll have to brute force your way through frequency hypotheses.

That's among the reasons why high-bandwidth communications switched away from CDMA (as used in UMTS and 802.11b, for example); without very specific restrictions on the code, the best you can do is decoding a set of frequency-shifted versions of your signal. One of them will work best – the one with the correct frequency offset.

Thinking of systems, not all is lost – one could have a non-CDMA preamble where you do an initial frequency error estimate, and then periodically insert pilot symbols in your CDMA data payload to keep track of phase drift. This really happens in a very narrow window between channel coherency and desirable maximum spreading gain.

  • $\begingroup$ Once you are within the coherence bandwidth of your symbols (1/T where T is your correlation time), you can simply track the post correlation I and Q rotation. I think of the frequency hypothesis operations you describe as the acquisition operation and "Doppler correction due to movement" as an ongoing function during carrier tracking. Do you agree Marcus? $\endgroup$ – Dan Boschen Aug 8 '18 at 3:08

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.