A model of decision feedback carrier recovery noise is presented as a function of the number of symbols over which the carrier is derived. For this model, quantitative results for performance degradation in BPSK, QPSK, 8-phase, 16-QAM, 32-QAM, and 64-QAM coherent demodulators are presented. When the carrier is derived from only the last symbol, these results reduce to the performance for differential detection. Carrier integration over 100 symbols is shown to be adequate to provide performance quite close to ideal for all the modulations considered.
No takes yet. Share an insight, caveat, or question.
Liskov et al. (1987) studied this question.
Synapse has enriched one closely related paper. Consider it for comparative context: