Convex Combination of SISO Equalization and Blind Source Separation for MIMO Blind Equalization

• Published in: Wireless personal communications Vol. 106; no. 3; pp. 1397 - 1409
• Main Authors: Sun, Yongjun, Zhu, Liangting, Li, Dongmin, Liu, Zujun
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.06.2019; Springer Nature B.V
• Subjects: Algorithms; Communications Engineering; Computer Communication Networks; Computer simulation; Convolution; Engineering; Equalization; Interference; MIMO (control systems); Networks; Probability density functions; Selective fading; Separation; Signal processing; Signal,Image and Speech Processing; SISO (control systems); Dual-mode; MIMO; Blind source separation; Low complexity
• ISSN: 0929-6212; 1572-834X
• DOI: 10.1007/s11277-019-06221-4

For Multiple-Input Multiple-Output (MIMO) system the received signal will suffer not only inter-symbol interference but also inter-antenna interference under frequency selective fading channel. This paper proposes a MIMO blind equalization algorithm consisting of the convex combination of Single-Input Single-Output (SISO) blind equalization algorithm and blind source separation (BSS). The main purpose of the SISO equalization algorithm is to convert the convolution channel into multiplicative channel, and the BSS algorithm is mainly used to separate the different sources. The SISO equalization algorithm used in the paper is the modified constant modulus algorithm (MCMA) because of its simplicity and effectiveness. The BSS algorithm is the constraint fitting probability density function algorithm (CFPA). The MIMO blind equalization algorithm is named as MCMA–CFPA. Moreover, a low complexity MCMA–CFPA and a dual mode algorithm based on soft switching are presented. Simulation results show that the proposed algorithms can simultaneously equalize and separate all transmission signals.