Uplink Pre-Compensation for Ground-to-Space Optical Communications Using Downlink Tomographic Reconstruction Journal: opticajournal Publication type: Research Article

Alex Frost, 1,* Nicolas Maron, 1 Shane Walsh, 1 Benjamin Dix-Matthews, 1 Sascha Schediwy 1 and Michael Hart 2 · 2025

Read Uplink Pre-Compensation for Ground-to-Space Optical Communications Using Downlink Tomographic Reconstruction Journal: opticajournal Publication type: Research Article by Alex Frost, 1,* Nicolas Maron, 1 Shane Walsh, 1 Benjamin Dix-Matthews, 1 Sascha Schediwy 1 and Michael Hart 2 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-08866.txt.

Introduction

Alex Frost and 5 co-authors present uplink pre-compensation for ground to space optical links in a 2025 paper. Correcting a laser beam before it travels up through air is hard, because the uplink path error is rarely measured. The downlink beam, by contrast, is easy to sense.

Repeated downlink measures are used to rebuild a three-dimensional picture of the shared air volume. From that picture, the uplink wavefront error is estimated. Simulations under typical conditions show good performance, especially at large point-ahead angles. Residual error falls by up to about nine times versus using downlink phase directly. The method needs no extra hardware beyond the existing sensor.

Catalog record

CREATORAlex Frost, 1,* Nicolas Maron, 1 Shane Walsh, 1 Benjamin Dix-Matthews, 1 Sascha Schediwy 1 and Michael Hart 2
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE30 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-08866

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Uplink Pre-Compensation for Ground-to-Space Optical Communications Using Downlink Tomographic Reconstruction Journal: opticajournal Publication type: Research Article Alex Frost, 1,* Nicolas Maron, 1 Shane Walsh, 1 Benjamin Dix-Matthews, 1 Sascha Schediwy 1 and Michael Hart 2 Abstract Uplink pre-compensation in ground-to-satellite optical links remains a problem, with an appropriate measurement of the wavefront error in the uplink path often not being available. We present a method that uses successive measurements of the readily accessible downlink beam to perform a tomographic reconstruction of the volume of atmosphere common to the downlink beam and the ground-station field-of-view. These measurements are done through the existing downlink wavefront sensor. From here, an estimate of the wavefront error along the uplink path can be obtained. We evaluate this method in simulation over representative atmospheric conditions and find good performance, especially for situations where the satellite point-ahead angle is many times greater than the atmosphere’s isoplanatic angle. Compared to pre-compensation directly using the downlink phase, we find that this method estimates the uplink path with a residual mean-square wavefront error that is up to 8.6 times less. The hardware simplicity of this method makes it a promising solution for uplink pre-compensation implementations targeted at optical communications. 1 Introduction Free-space optical (FSO) laser links will revolutionise the speed and security of wireless signal transfer, promoting major advancements in the field of communications [ 1 ] . However, higher optical frequencies have an increased sensitivity to atmospheric turbulence compared to radio- and microwave-frequency counterparts [ 2 ] . Propagation through atmospheric turbulence causes rapid fluctuations in the intensity and phase of the beam, leading to frequent signal dropouts. Astronomy-derived adaptive optics (AO) systems are often used in FSO links to alleviate these effects through active phase compensation of the received beam [ 3 ] . Such systems continuously measure the wavefront of some reference signal and correct it to a required shape. In an FSO link between an orbiting satellite and an optical ground station (OGS) , AO compensation on the satellite-to-ground downlink path can be done at the ground station using the downlink laser beam directly. However, the uplink path poses significant complications. Most atmospheric turbulence is concentrated in the first 20 km above the Earth’s surface [ 4 ] . This means the uplink beam will interact with turbulence immediately and then propagate a further ¿500 km to the satellite receiver. The resulting large spatial scale intensity fluctuations will cause fades over the entire satellite aperture. To overcome this, the uplink beam can be pre-compensated at the OGS with a dedicated AO system such that propagation through the atmosphere removes the turbulence-induced wavefront perturbations at the satellite. An experimental demonstration of pre-compensation has been conducted over a horizontal link, showing significantly improved received power statistics at the receiver [ 5 ] . The choice of reference signal for uplink pre-compensation is critical and situation-dependent. All satellites in orbit have their uplink and downlink paths separated by their point-ahead angle (PAA) , a result of the satellite’s orbital speed and the finite speed of light. Due to their low PAA , Geostationary-earth orbit (GEO) satellites generally have strong correlation between the phase perturbations in the uplink and downlink paths in low to moderate turbulence conditions. In these cases, the downlink signal serves as an excellent pre-compensation reference signal [ 6 ] . However, as the satellite’s orbital speed increases from the perspective of the OGS , or the atmosphere becomes more turbulent, eventually the PAA exceeds the atmosphere’s isoplanatic angle, \theta_{0} , and the two paths are considered uncorrelated [ 4 ] . The terrestrial FSO link pre-compensation demonstration from Brady et al. [ 5 ] confirms that using the downlink signal for pre-compensation becomes less beneficial as the ratio \mathrm{PAA/\theta_{0}} increases. For \mathrm{PAA/\theta_{0}>2} , even the tip/tilt of the paths becomes uncorrelated, causing use of the downlink signal to ultimately worsen the received power statistics. For low-earth orbit (LEO) satellites, where the uplink and downlink paths will effectively always be uncorellated, a signal at the PAA is preferred. This could be achieved through a laser guide star; however the inherent tip/tilt indetermination leaves a major contributor to the overall phase perturbations unsensed [ 7 ] . Alternatively, the atmospheric structure at the PAA can be estimated by exploiting the readily available downlink beam. Simulation work has been conducted by Logone et al. [ 8 ] which uses the downlink phase and intensity measurements and the overhead \mathrm{C_{n}^{2}} profile to estimate the atmospheric structure at the PAA . We propose a new method that uses successive measurements of the downlink beam to perform a tomographic reconstruction of the atmosphere. The method assumes an atmosphere stratified into several equivalent layers with discrete apparent wind vectors, influenced by the orbital speed of the satellite. The net downlink phase is decomposed onto each layer, from which the phase at the PAA can be estimated. We present simulation results of this method, showing the reconstruction accuracy of the uplink path for a variety of different ground-to-space FSO link configurations. We additionally explore the method’s sensitivity to errors in its input parameters. 2 Method A schematic of the method is shown in Fig. 1 A. We assume an uplink beam transmitted from within a subaperture of the ground station’s main aperture. The uplink beam is transmitted in such a way to maximise overlap with the downlink beam during propagation. This is because this method is only able to measure the volume of atmosphere above the telescope that has been illuminated by the downlink beam. Figure 1: (A) Schematic of this method. PAA – point-ahead angle, \theta_{0} – isoplanatic angle, \vec{v} – equivalent layer velocity, h – equivalent layer height. (B) Above: visualisation of a single ‘swept mode’ for a single equivalent layer, defined by the wind vector \vec{v}_{\mathrm{layer}} . Below: the sampling of the swept mode with the telescope aperture for lookback frames 1 and N. This method uses solely the apparent layer wind velocities (i.e., the combination of the actual layer wind and the pseudo-wind due to satellite tracking) to decompose the net downlink phase onto each equivalent layer. We assume a constant and homogeneous layer velocity over the measurement period, and as such are restricted to working within the frozen flow hypothesis timescale, \tau_{fs} , of \sim 20 ms [ 9 ] . This method allows us to map a series of consecutive wavefront sensor (WFS) measurements to one volumetric distribution of phase over the equivalent layers. We call these measurements ‘lookback frames’. The method is structured as follows: 1. We define an area of atmosphere swept over the telescope aperture by the layer’s wind vector over N lookback frames. This forms the envelope of the swept mode in Fig. 1 B. 2. Within this envelope, we generate an atmospheric phase perturbation basis by performing principal component analysis on 5000 realisations of Kolmogorov turbulence. We also explicitly add tip and tilt to this basis. 3. We sample across each mode of this basis with the telescope aperture over the N lookback frames. These samples give us information about how each mode evolves over the lookback frames. 4. By measuring these samples with the WFS , we build an influence matrix with

Read the complete text