A simple recursive least square algorithm of space-temporal joint equalizer

doi:10.1016/j.dsp.2012.06.003

NHUIR > College of Science and Technology > Department of Computer Science and Information Engineering > Periodical Articles > Item 987654321/25090

Please use this identifier to cite or link to this item: http://nhuir.nhu.edu.tw/handle/987654321/25090

Title:	A simple recursive least square algorithm of space-temporal joint equalizer
Authors:	吳建民;柳茂林;廖怡欽;Wu, Chien-Min;Leou, Maw-Lin;Liaw, Yi-Ching
Contributors:	資訊工程學系
Keywords:	Adaptive array;Space-temporal joint equalizer (STE);Co-channel interference (CCI);Inter-symbol interference (ISI);Recursive least square (RLS);Convergent rate
Date:	2012-12
Issue Date:	2017-07-19 15:54:35 (UTC+8)
Abstract:	The environment of the communication system will be more complicated in the future. Therefore, the adaptive array alone cannot cancel all the interferences in future communication system. Many space–temporal joint equalizers (STE) have been proposed to further enhance the capability of canceling the co-channel interferences (CCI) and the multi-path inter-symbol interferences (ISI). The recursive least square (RLS) algorithm of those joint methods will suffer the problems of system structure is complex, the convergent rate of performance is slow or number of calculations is large. In this paper, we propose a simple RLS algorithm (SRLS) of the STE to reduce the computation load, simplify the system structure and improve the system performance. The SRLS algorithm uses the output signal of STE to adjust the weights of adaptive array and equalizer simultaneously without using a transversal filter. Since there is no transversal filter required in the system, the system is simpler in comparison to the conventional STE. Besides, the RLS algorithm of conventional STE adjusts the weights of equalizer after the weights of adaptive array. The convergent rate of weights is slowly in those STE systems. The SRLS algorithm adjusts the weights of adaptive array and equalizer simultaneously. Therefore, the convergent rate of system will improve by the SRLS. Furthermore, the SRLS algorithm is found with input vectors of adaptive array and equalizer separately. Another STE method, which combines the input vectors of adaptive array and equalizer into a large dimension input vector, needs more number of calculations than the SRLS method. Thus, the SRLS of STE can simplify the complexity of system structure, reduce the computation load and improve the convergent performance of system effectively.
Relation:	Digital Signal Processing vol. 22, no. 6 pp.1145-1153
Appears in Collections:	[Department of Computer Science and Information Engineering] Periodical Articles

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	419	View/Open

Loading...