A Robust GLRT Receiver With Implicit Channel Estimation and Automatic Threshold Adjustment for the Free Space Optical Channel with IM/DD

• Published in: Journal of lightwave technology Vol. 32; no. 3; pp. 369 - 383
• Main Authors: TIANYU SONG, KAM, Pooi-Yuen
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.02.2014; Institute of Electrical and Electronics Engineers
• Subjects: Applied sciences; Channel estimation; Channel models; Complexity theory; Decision-feedback (DFB); Detection, estimation, filtering, equalization, prediction; Exact sciences and technology; free space optical (FSO); Information, signal and communications theory; intensity modulation/direct detection (IM/DD); Mathematical model; maximum likelihood (ML) detection; Measurement; Optical telecommunications; receiver design; Receivers; sequence detection; Signal and communications theory; Signal, noise; Silicon; Systems, networks and services of telecommunications; Telecommunications; Telecommunications and information theory; Transmission and modulation (techniques and equipments); Wireless optical communications; Performance evaluation; Parameter estimation; Error estimation; Optical telecommunication; Wireless telecommunication; Feedback regulation; Time variation; Implementation; Degradation; maximum likelihood (ML) detection; Viterbi detection; Feedback; Channel estimation; receiver design; Free space optics; free space optical (FSO); Decision-feedback (DFB); Receiver; intensity modulation/direct detection (IM/DD); sequence detection; Direct detection; Free space propagation; Atmospheric turbulence; Likelihood ratio test; Metric; Maximum likelihood; Intensity modulation; Gain
• ISSN: 0733-8724; 1558-2213
• DOI: 10.1109/JLT.2013.2293137

Atmospheric turbulence and pointing errors cause intensity fluctuation of free space optical communication signals and impair link performance. Several receiver structures which could mitigate the signal fluctuations were proposed in the past, but these existing receivers depend highly on the channel model and the model parameters. The performance deteriorates if the channel model or the model parameters are inaccurate. In this paper, we develop a Viterbi-type trellis-search sequence receiver based on the generalized likelihood ratio test principle that jointly detects the data sequence and estimates the unknown channel gain. This receiver requires very few pilot symbols, and therefore, does not significantly reduce the bandwidth efficiency. It is robust in that it continuously performs maximum likelihood (ML) estimation of the unknown channel gain without the knowledge of the channel model, and adapts the decision metric accordingly. It works well in a slowly time-varying environment and its error performance approaches that of ML detection with perfect knowledge of the channel gain, as the memory length used for forming the sequence detection metric increases. A new, decision-feedback, symbol-by-symbol receiver with lower implementation complexity and higher memory efficiency is obtained as an approximation to the sequence receiver. The performance improvement and implementation simplicity of our receivers compared to existing receivers are pointed out.