Paper Abstract and Keywords |
Presentation |
2022-06-16 14:30
Comparison of Performance and Complexity for different Search Methods in Stochastic MIMO Signal Detection Hiroki Asumi, Yukiko Kasuga, Kazushi Matsumura, Junichiro Hagiwara, Toshihiko Nishimura, Takanori Sato, Yasutaka Ogawa, Takeo Ohgane (Hokkaido Univ.) RCS2022-49 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
In large-scale MIMO signal detection, the computational complexity increases as the number of antennas increases. We have proposed several approaches to reduce computational complexity while maintaining detection performance as much as possible by applying a mixed normal distribution to the prior distribution in a stochastic problem setting to solve this problem. This study compares the computational complexity and detection performance of the Hamiltonian Monte Carlo method, variational Bayesian method, and L-BFGS method in this framework. The results show that the detection performance of each method differs due to the mechanism of initial value change and the handling of the gradient information for the log-posterior probability density in the algorithm. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
MIMO / Signal detection / Hamiltonian Monte Carlo / Variational Bayesian method / L-BFGS / / / |
Reference Info. |
IEICE Tech. Rep., vol. 122, no. 73, RCS2022-49, pp. 150-155, June 2022. |
Paper # |
RCS2022-49 |
Date of Issue |
2022-06-08 (RCS) |
ISSN |
Online edition: ISSN 2432-6380 |
Copyright and reproduction |
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
Download PDF |
RCS2022-49 |
|