Gaussian Fading Interference Channels: Power Control

被引:11
作者
Tuninetti, Daniela [1 ]
机构
[1] Univ Illinois, Dept Elect & Comp Engn, Chicago, IL 60680 USA
来源
2008 42ND ASILOMAR CONFERENCE ON SIGNALS, SYSTEMS AND COMPUTERS, VOLS 1-4 | 2008年
关键词
D O I
10.1109/ACSSC.2008.5074498
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
In this paper we consider power allocation policies for 2-user Gaussian interference channels (IFC) with ergodic fading, perfectly known at all terminals. Based on recently found outer bounds on the capacity region of unfaded IFCs, we determine the power allocation that maximizes a sum-rate outer bound. With this power allocation both users are simultaneously active when the power of the interfering signals is small relative to the power of the intended signals. Otherwise, only the user with the largest fading gain is active, similar to multi-access and broadcast channels. By considering a simplified version of the Han-Kobayashi region, we derive the power allocation that maximizes a sum-rate inner bound. Numerical results for iid Rayleigh fading show that the inner and our bounds are very close to one another. The structure of the determined power allocation policies suggests that practical power allocations can be designed based on only a bit of channel state information.
引用
收藏
页码:701 / 706
页数:6
相关论文
共 15 条
[1]  
Akuiyibo E., 2008, P 2008 INT ZUR SEM M
[2]  
Boyd S., 2004, CONVEX OPTIMIZATION
[3]  
CADAMBE V, 2007, ARXIV07070323V2
[4]  
CADAMBE VR, 2008, ARXIV08022125
[5]   The capacity region of frequency-selective Gaussian interference channels under strong interference [J].
Chung, Seong Taek ;
Cioffi, John M. .
IEEE TRANSACTIONS ON COMMUNICATIONS, 2007, 55 (09) :1812-1821
[6]  
ETKIN RH, 2007, T INFO THE UNPUB JUN, P2181
[7]  
HAN TS, 1981, IEEE T INFORM THEORY, V27, P49
[8]  
KNOPP R, 1999, IEEE INT C COMM 1995, V1, P331
[9]   Outer bounds on the capacity of Gaussian interference channels [J].
Kramer, G .
IEEE TRANSACTIONS ON INFORMATION THEORY, 2004, 50 (03) :581-586
[10]  
SANKAR L, 2008, P INT S INF THEOR 20