Robust Data Detection for the Photon-Counting Free-Space Optical System with Implicit CSI Acquisition and Background Radiation Compensation

TL;DR

This paper introduces a robust, implicit CSI acquisition method for free-space optical systems that effectively compensates for background radiation and turbulence, improving detection accuracy without relying on pilot symbols or perfect background knowledge.

Contribution

It proposes a novel sequence receiver and a simplified symbol-by-symbol receiver that implicitly estimate CSI and background radiation, enhancing robustness and efficiency in FSO detection.

Findings

Performance approaches that of perfect CSI knowledge as window length increases

Receivers are robust against turbulence and pointing errors

Simplified receiver maintains high accuracy with reduced complexity

Abstract

Since atmospheric turbulence and pointing errors cause signal intensity fluctuations and the background radiation surrounding the free-space optical (FSO) receiver contributes an undesired noisy component, the receiver requires accurate channel state information (CSI) and background information to adjust the detection threshold. In most previous studies, for CSI acquisition, pilot symbols were employed, which leads to a reduction of spectral and energy efficiency; and an impractical assumption that the background radiation component is perfectly known was made. In this paper, we develop an efficient and robust sequence receiver, which acquires the CSI and the background information implicitly and requires no knowledge about the channel model information. It is robust since it can automatically estimate the CSI and background component and detect the data sequence accordingly. Its decision metric has a simple form and involves no integrals, and thus can be easily evaluated. A Viterbi-type trellis-search algorithm is adopted to improve the search efficiency, and a selective-store strategy is adopted to overcome a potential error floor problem as well as to increase the memory efficiency. To further simplify the receiver, a decision-feedback symbol-by-symbol receiver is proposed as an approximation of the sequence receiver. By simulations and theoretical analysis, we show that the performance of both the sequence receiver and the symbol-by-symbol receiver, approach that of detection with perfect knowledge of the CSI and background radiation, as the length of the window for forming the decision metric increases.