DocumentCode :
1373507
Title :
Capacity Bounds for Wireless Optical Intensity Channels With Gaussian Noise
Author :
Farid, Ahmed A. ; Hranilovic, Steve
Author_Institution :
Dept. of Electr. & Comput. Eng., McMaster Univ., Hamilton, ON, Canada
Volume :
56
Issue :
12
fYear :
2010
Firstpage :
6066
Lastpage :
6077
Abstract :
Lower and upper bounds on the capacity of wireless optical intensity pulse amplitude modulation channels under nonnegativity and average optical power constraints are derived. A lower bound is derived based on source entropy maximization over a family of discrete nonuniform distributions with equally spaced mass points. A closed form for the maxentropic discrete input distribution is provided. Compared to previously reported bounds, the derived lower bound is tight at both low and high signal-to-noise ratios (SNRs). In addition, a closed-form upper bound is derived based on signal space geometry via a sphere packing argument. The proposed bound is tight at low SNRs and incurs a small gap to the channel capacity at high SNRs. The derived bounds asymptotically describe the optical intensity channel capacity at low SNRs, where a majority of such links operate.
Keywords :
Gaussian noise; channel capacity; optical communication; pulse amplitude modulation; wireless channels; Gaussian noise; capacity bounds; channel capacity; entropy maximization; optical power constraints; pulse amplitude modulation; wireless optical communications; wireless optical intensity channels; Channel capacity; Gaussian noise; Optical fiber communication; Optical modulation; Signal to noise ratio; Upper bound; Wireless communication; Conditional Gaussian channels; entropy maximization; optical channel capacity; sphere packing; wireless optical communications;
fLanguage :
English
Journal_Title :
Information Theory, IEEE Transactions on
Publisher :
ieee
ISSN :
0018-9448
Type :
jour
DOI :
10.1109/TIT.2010.2080470
Filename :
5625632
Link To Document :
بازگشت