Abstract
Optimization of iterative algorithms for channel state estimation in a sparse coding multiple-access system (SCMA) is performed to reduce computational costs of the receiver. It is shown that when a signal-to-noise ratio (SNR) does not exceed 10 dB, one iteration of the algorithm is sufficient, and an increase in the number of iterations does not lead to an increase in spectral efficiency. Simulation demonstrates a possibility of a reasonable choice of the total number of decoder iterations and their distribution between different stages of the channel estimation. For an uncoded system, iterative re-estimation of the channel is proposed, as well as ways to reduce computational costs during its calculation. In the coded system, at a low SNR the achieved spectral efficiency values are approximately similar to those with pilot-only channel estimation. The article provides recommendations for the placement of data symbols and pilot signals in re-source blocks to increase the system spectral efficiency.
Highlights
Optimization of iterative algorithms for channel state estimation in a sparse coding multiple-access system (SCMA) is performed to reduce computational costs of the receiver
По этой причине в алгоритме 2 при переоценке канала не используется турбокод – его применение требовало бы декодирования всех Q сигналов множественного доступа с разреженным кодированием (МДРК) в блоке после их оценки по пилот-сигналам
Вторым исследуемым методом оценки состояния канала являлся метод последовательной итерационной переоценки, при котором для каждого сигнала МДРК используется только однократное декодирование с применением алгоритма MPA
Summary
Optimization of iterative algorithms for channel state estimation in a sparse coding multiple-access system (SCMA) is performed to reduce computational costs of the receiver. Символов сигнала МДРК; D – матрица кодовых слов с размерами R L J для канала без разнесения и Для ортогональных пилот-сигналов дисперсия оценки канальных коэффициентов в системе без разнесения определяется границей Крамера–Рао:
Published Version (Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have