Fast gradient algorithm for complex ICA and its application to the MIMO systems
Author:
Open Access
- 19 July 2023
- journal article
- research article
- Published by Springer Nature in Scientific Reports
- Vol. 13 (1) , 1-16
Abstract
This paper proposes a new gradient-descent algorithm for complex independent component analysis and presents its application to the Multiple-Input Multiple-Output communication systems. Algorithm uses the Lie structure of optimization landscape and toral decomposition of gradient matrix. The theoretical results are validated by computer simulation and compared to several classes of algorithms, gradient descent, quasi-Newton as well as complex JADE. The simulations performed showed excellent results of the algorithm in terms of speed, stability of operation and the quality of separation. A characteristic feature of gradient methods is their quick response to changes in the input signal. The good results of the proposed algorithm indicate potential use in on-line applications.Keywords
This publication has 40 references indexed in Scilit:
- Blind adaptive receiver based on RLS-CMA for TH-UWB systemThe Journal of China Universities of Posts and Telecommunications, 2011
- Blind source separation based on constant modulus criterion and signal mutual informationComputers and Electrical Engineering, 2008
- A Simplified Constant Modulus Algorithm for Blind Recovery of MIMO QAM and PSK Signals: A Criterion with Convergence AnalysisEURASIP Journal on Wireless Communications and Networking, 2007
- Geometrical methods for non-negative ICA: Manifolds, Lie groups and toral subalgebrasNeurocomputing, 2005
- The Scaling and Squaring Method for the Matrix Exponential RevisitedSIAM Journal on Matrix Analysis and Applications, 2005
- Blind source separation of instantaneous MIMO systems based on the least-squares Constant Modulus AlgorithmIEE Proceedings - Vision, Image, and Signal Processing, 2005
- Globally convergent blind source separation based on a multiuser kurtosis maximization criterionIEEE Transactions on Signal Processing, 2000
- A blind signal separation method for multiuser communicationsIEEE Transactions on Signal Processing, 1997
- Blind beamforming for non-gaussian signalsIEE Proceedings F Radar and Signal Processing, 1993
- Self-Recovering Equalization and Carrier Tracking in Two-Dimensional Data Communication SystemsIEEE Transactions on Communications, 1980