Imperial College London
Portrait Picture

ProfessorBrunoClerckx

Faculty of Engineering, Department of Electrical and Electronic Engineering

Professor of Wireless Communications and Signal Processing
 
 
 
//

Contact

 

+44 (0)20 7594 6234b.clerckx Website

 
 
//

Location

 

816Electrical EngineeringSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@article{Joudeh:2016:10.1109/TSP.2016.2591501,
author = {Joudeh, H and Clerckx, B},
doi = {10.1109/TSP.2016.2591501},
journal = {IEEE Transactions on Signal Processing},
pages = {6227--6242},
title = {Robust Transmission in Downlink Multiuser MISO Systems: A Rate-Splitting Approach},
url = {http://dx.doi.org/10.1109/TSP.2016.2591501},
volume = {64},
year = {2016}
}
Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - We consider a downlink multiuser MISO systemwith bounded errors in the Channel State Information at theTransmitter (CSIT). We first look at the robust design problemof achieving max-min fairness amongst users (in the worstcasesense). Contrary to the conventional approach adopted inliterature, we propose a rather unorthodox design based on aRate-Splitting (RS) strategy. Each user’s message is split intotwo parts, a common part and a private part. All commonparts are packed into one super common message encodedusing a public codebook, while private parts are independentlyencoded. The resulting symbol streams are linearly precodedand simultaneously transmitted, and each receiver retrieves itsintended message by decoding both the common stream andits corresponding private stream. For CSIT uncertainty regionsthat scale with SNR (e.g. by scaling the number of feedbackbits), we prove that a RS-based design achieves higher max-min(symmetric) Degrees of Freedom (DoF) compared to conventionaldesigns (NoRS). For the special case of non-scaling CSIT (e.g.fixed number of feedback bits), and contrary to NoRS, RS canachieve a non-saturating max-min rate. We propose a robustalgorithm based on the cutting-set method coupled with theWeighted Minimum Mean Square Error (WMMSE) approach,and we demonstrate its performance gains over state-of-the artdesigns. Finally, we extend the RS strategy to address the Qualityof Service (QoS) constrained power minimization problem, andwe demonstrate significant gains over NoRS-based designs.
AU  - Joudeh,H
AU  - Clerckx,B
DO  - 10.1109/TSP.2016.2591501
EP  - 6242
PY  - 2016///
SN  - 1053-587X
SP  - 6227
TI  - Robust Transmission in Downlink Multiuser MISO Systems: A Rate-Splitting Approach
T2  - IEEE Transactions on Signal Processing
UR  - http://dx.doi.org/10.1109/TSP.2016.2591501
UR  - http://hdl.handle.net/10044/1/34520
VL  - 64
ER  -
Download