
Low‐complexity equalisation of doubly selective channels for multiple‐input–multiple‐output transmission systems
Author(s) -
Qaisrani Muhammad T.N.,
Barhumi Imad
Publication year - 2013
Publication title -
iet signal processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.384
H-Index - 42
ISSN - 1751-9683
DOI - 10.1049/iet-spr.2012.0073
Subject(s) - computer science , transmission (telecommunications) , transmission channel , channel (broadcasting) , speech recognition , algorithm , telecommunications , control theory (sociology) , artificial intelligence , control (management)
Equalisation of multiple‐input–multiple‐output doubly selective channels is studied and a technique for low‐complexity equalisation of doubly selective channels is proposed. This paper studies low‐complexity techniques for linear equalisation as well as two different architectures for non‐linear equalisation. The first non‐linear architecture only involves feedback of decisions on previously detected symbols to cancel intersymbol interference. The second non‐linear architecture separates the interference into intersymbol interference between successive transmissions and co‐channel interference (CCI) between the multiple transmit streams. The ISI is then handled with a combination of both interference suppression via filtering and interference cancellation via decision feedback. However, the CCI is handled by an ordered successive interference cancellation mechanism. Both receiver structures involve significant design complexity because of the channel time variations. Therefore low‐complexity implementations are proposed. The time‐varying nature of the channel requires time‐varying equalisers. The proposed low‐complexity scheme thus involves designing equalisers at uniformly spaced intervals within the transmit block and then interpolating to obtain the time‐varying equalisers over the remainder of the block. The proposed low‐complexity approaches to equalisation are analysed in terms of both performance and complexity.