Open AccessSimplified log likelihood ratio calculation in soft ordered successive interference cancellation for multiple‐input multiple‐output digital video broadcasting‐second generation terrestrial receivers
Author(s) -
Kang InWoong,
Lee YoungJun,
Kim Youngmin,
Seo Jae Hyun,
Kim Heung Mook,
Kim HyoungNam
Publication year - 2014
Publication title -
iet communications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.355
H-Index - 62
eISSN - 1751-8636
pISSN - 1751-8628
DOI - 10.1049/iet-com.2014.0510
Subject(s) - single antenna interference cancellation , computer science , mimo , interference (communication) , broadcasting (networking) , decoding methods , antenna (radio) , computational complexity theory , equaliser , algorithm , electronic engineering , telecommunications , computer network , beamforming , channel (broadcasting) , engineering
This study proposes a soft ordered successive interference cancellation (OSIC) receiver accompanied by a simplified log‐likelihood ratio calculation method for a dual‐polarised multiple‐input multiple‐output (MIMO) digital video broadcasting‐second generation terrestrial system. The future terrestrial transmission systems will be required to provide high data rates for ultra‐high definition television service. For this purpose, dual‐polarised MIMO techniques with a little cross‐antenna interference and high‐order modulation schemes have been considered. As the cross‐antenna interference needs to be removed, a soft OSIC receiver can be used effectively. However, the receiver has to endure high computational complexity for calculating soft values. To reduce it, the authors simplified the calculation based on the decision threshold (DT) method. The main contribution of this study is to extend the conventional DT method, which is restricted to only low‐order modulations, to an algorithmic process applicable to any modulation order. The proposed soft OSIC receiver has about 1 dB degradation from the soft OSIC receiver with the maximum‐likelihood calculation of soft values. The degradation can be fully rewarded by the significantly reduced computational complexity. Furthermore, this study presents an additional possibility of improving the decoding performance by modelling the OSIC equaliser as a biased estimator.