Integrating independent component analysis with Hopfield recurrent neural network to estimate the channel of MIMO-OFDM system

doi:10.1145/3271553.3271593

	Integrating independent component analysis with Hopfield recurrent neural network to estimate the channel of MIMO-OFDM system
	Jie, Hao
	2018-08-27
会议名称	2nd International Conference on Vision, Image and Signal Processing, ICVISP 2018
会议录名称	ACM International Conference Proceeding Series
会议日期	August 27, 2018 - August 29, 2018
会议地点	Las Vegas, NV, United states
出版者	Association for Computing Machinery
摘要	Channel estimation is very important for MIMO-OFDM system, and because of the complexity of the electromagnetism circumstance in multipath communication environment and the sensitivity of OFDM modulation mode, which brings great difficulty for the channel estimation. In order to solve the difficulty of channel estimation problem in MIMO-OFDM system, this paper propose a semi-blind estimation method based on independent component analysis (ICA) algorithm combined with improved Hopfield recurrent neural network (HRNN) as a hybrid approach named ICA-HRNN. Then use the ICA-HRNN algorithm to estimate the channel information of MIMO-OFDM system. The simulation results show that, the ICA-HRNN algorithm can better adapt to the nonlinear characteristics of MIMO-OFDM system, and increase the estimation accuracy and the estimation speed, especially when the system has low SNR. © 2018 Association for Computing Machinery.
关键词	Gain control Image processing Independent component analysis MIMO systems Orthogonal frequency division multiplexing Recurrent neural networks Signal to noise ratio Channel information Hopfield recurrent neural networks Improved HRNN Independent component analysis(ICA) MIMO-OFDM systems Multipath communications Nonlinear characteristics Semi-blind estimations
DOI	10.1145/3271553.3271593
收录类别	EI
语种	英语
EI入藏号	20185106273487
EI主题词	Channel estimation
来源库	Compendex
分类代码	716.1 Information Theory and Signal Processing - 731.3 Specific Variables Control
引用统计	被引频次：10[WOS] [WOS记录] [WOS相关记录]
文献类型	会议论文
条目标识符	https://ir.lut.edu.cn/handle/2XXMBERH/118000
专题	兰州理工大学
作者单位	Northwest Minzu University, School of Electrical Engineering, China. Lanzhou University of Technology, School of Computer and Communication, Lanzhou, China
第一作者单位	兰州理工大学
推荐引用方式 GB/T 7714	Jie, Hao. Integrating independent component analysis with Hopfield recurrent neural network to estimate the channel of MIMO-OFDM system[C]:Association for Computing Machinery,2018.