Loading...
Proceedings of COAT-2019 Workshop
Communications and Observations through Atmospheric Turbulence :
characterization and mitigation
02 to 03 DECEMBER 2019 / ONERA, Châtillon FRANCE
COAT-2019 has been organized in the context of "Journées Scientifiques ONERA (JSO)" by Jean-Marc Conan (ONERA) with the support
- of the following Scientific Committee :
Géraldine Artaud (CNES), James Osborn (Durham Univ.), Ramon Mata-Calvo (DLR), Szymon Gładysz (Fraunhofer IOSB), Aniceto Belmonte (Technical Univ. of Catalonia), Olivier Meyer (DGA), Anne Durécu (ONERA), Clélia Robert (ONERA), Laurent Mugnier (ONERA) ;
- of the following Organizing Committee at ONERA :
Marylène Baroin, Alain Broc, Frédéric Cassaing, Anthony Drouaire, Elodie Esperanssa, Caroline Lim, David Mariette, Aurélie Montmerle-Bonnefois, Nicolas Védrenne, Marie-Thérèse Velluet.
Sessions :
- Atmospheric turbulence : modeling, characterization & prediction
- Impact of turbulence on beam propagation
- Adaptative optics for space observation
- Novel turbulence mitigation strategies
- Ground-space optical links : concepts and applications
- Free space communication links : experimental demonstrations
Scope :
In a large variety of fields, the development of ever more demanding optical systems now requires a thorough understanding of the impact of atmospheric turbulence together with the development of dedicated mitigation strategies. Originally driven by astronomy, these issues now concern near ground imaging, satellite observation, ground-space and near ground optical telecommunications and frequency dissemination, laser focusing... The workshop aims at giving an overview of current research activities in this field. It will address turbulence impact and mitigation, especially when facing strong turbulence, strong scintillation, high apparent wind speeds induced by target tracking, anisoplanatism effects due to the field of view or point-ahead angle... The following topics will, therefore, be addressed during the workshop: knowledge on turbulence conditions (modelling, characterization & prediction of both Cn² and wind profiles); study of system performance with and without mitigation; mitigation strategies based on innovative optical solutions such as single aperture adaptive optics, multi-aperture or aperture diversity solutions, integrated optics architectures, new wave-front sensing approaches, sensorless solutions, laser guide stars, high performance control; lastly, image/signal pre- and post-processing techniques (deconvolution, PSF prediction, digital communications techniques), possibly in a co-design approach.
Proceedings
Session 1 : Atmospheric turbulence: modeling, characterization & prediction
Laser communication (lasercom) is influenced by atmospheric turbulence, a quality measured by the refractive index structure parameter Cn². This paper quantifies the degree of improvement to lasercom link budgets afforded by using ground-level measurements of turbulence in vertical turbulence models. Ground-level Cn² is measured with an off-the-shelf scintillometer for a path adjacent to DLR's optical ground station (OGS). Measurements are in agreement with literature on turbulence; nighttime Cn² is well represented by a log-normal distribution. Comparisons are drawn between profiles by comparing link budget parameter estimates generated by four turbulence profile models: HV-5/7, HV with Cn²(h0), and HV and HAP models with Cn²(h0) and fitting to downlink experiment data. Vertical turbulence profiles are converted to scintillation index sigma²_I by way of theory described in the literature on weak and strong turbulence. Normalised root-mean-squared-error is used to establish goodness-of-fit of modeled sigma²_I to downlink beam parameter measurements . Use of Cn²(h0) in a profile model improves upon the fit beyond HV-5/7 by ~8.3%. Improvements in the mean expectation from specific fits to satellite downlink experiments improve the NRMSE 30%. However, the variability in margin estimation due to changes in Cn²(h0) indicates fitting might not be a consistent improvement over the HV-5/7 model. This paper describes the setup of the scintillometer, six months of measurements, the use of Cn²(h0) measurements in vertical profile models to find the path integrated intensity scintillation index (sigma²_I), and a comparison of modeled-path integrated scintillation index to the scintillation index of downlink ground measurements at DLR's OGS.
Session 2 : Impact of turbulence on beam propagation
Small temperature variations carried by the turbulent velocity field in the Earth's atmosphere produce small phase perturbations in laser beam propagating through it. Perturbations in the phase cause wave-front errors and may significantly decrease the Strehl ratio of the laser beam. For compensation of atmospheric distortions, adaptive optics systems are used. Before the design of any adaptive optics system, the simulation of the atmospherically distorted laser beam is important. In this work, we describe the purpose, theory, implementation, and results of a wave optics propagation simulation code LAtmoSim and we investigate the effect of laser system parameters on the degree of the wave-front error caused by turbulence.
Session 3 : Adaptative optics for space observation
The Adaptive Optics Facility (AOF) is an ESO project started in 2005, which transformed Yepun, one of the four 8m telescopes in Paranal, into an adaptive telescope. This has been done by replacing the conventional secondary mirror of Yepun by a 1172 actuator Deformable Secondary Mirror (DSM) and attaching four Laser Guide Star (LGS) Units to its centerpiece, each of them featuring a high stability narrow-band 20 W laser. Additionally, two Adaptive Optics (AO) modules (GALACSI serving MUSE a 3D spectrograph, and GRAAL, serving Hawk I a wide field infrared imager) have been assembled onto the telescope Nasmyth adapters, each of them incorporating four LGS WaveFront Sensors (WFS) and one tip-tilt sensor used to control the DSM at 1 kHz frame rate. These WFSs are based on 0 Read-out Noise (RoN) detectors. The complete AOF has been operated on-sky for more than 2 years and is routinely delivering science. This paper presents the most important and amazing features of the AOF, focusing on cutting-edge technology in use, operational concept and overall performance. In the first part of the paper, the AOF design is recalled, with a focus on the DSM, the lasers, the WFS cameras and the Real Time Computer technology. Then, the acquisition sequence and overall on-sky operation efficiency will be detailed; finally, on-sky performance of AOF will be presented.
It is well-known that data-driven models for the tip-tilt modes significantly improve the performance of Adaptive Optics (AO) systems as it allows to compensate for vibration-induced disturbances. Whether identifying from data the temporal dynamics of more modes makes an impact on the performance has been studied on-sky with the CANARY demonstrator at the William Herschel Telescope in July 2019. In this brief paper, we report on these experiments using both Strehl ratios computed from the science camera in H band or by replaying the AO telemetry data in numerical simulations. We show that Linear Quadratic Gaussian (LQG) controllers that embed a data-driven model for the low-orders (that are not limited to the tip-tilt) improve the performance of the AO for two different dynamical behaviours of the atmospheric turbulence.
Space Situational Awareness (SSA) has become a key issue both for defense and civilian/industrial applications. Identification of potential or active threats and monitoring of key assets and operations are at stake. But it also includes follow up of dedicated satellites (such as telecommunication, observation), traffic handling, debris identification and tracking. Radar imaging is not enough for detailed characterization and identification of satellites, even for low earth orbits, and optical imaging can provide a powerful and complementary tool. We present current works performed at Onera on adaptive optics for Low Earth Orbit (LEO) satellite imaging. We briefly present our system and focus on the specific developments made to tackle the particular issue of fast moving targets observed through the atmosphere. We also discuss image post-processing in this particular framework, where turbulence conditions and correction strongly evolve along the observation, leading to a badly known point spread function (PSF). We have developed a robust and efficient post-processing method based on recent works for both astronomical and biomedical observations. We discuss further possible improvements.
Session 4 : Novel turbulence mitigation strategies
In this work we propose an optimal modal decomposition technique for gradient-based wavefront sensorless adap-tive optics. We implement statistically independent Karhunen-Loève functions for iterative blind correction, and analyze their optimal correction performance in various turbulence conditions in temporally evolving turbulence simulations.
Adaptive atmospheric compensation in laser communication systems can be iteratively addressed minimizing the loop bandwidth requirements. Here we evaluate in the laboratory a speckle-based iterative technique within the framework of a communication scenario. Experimental results show signicant coupling efficiency gain with a reduced number of iterations. Evaluation of the technique with an integrated 80 Gbps QPSK system shows an error probability of 1e-5 for highly turbulent atmospheric channels.
We have considered the problem of finding optimal transmission beams, or eigenmodes, for propagation through the random atmosphere with minimum distortion. We have found, making a realistic assumption that a transmitter knows atmospheric turbulence statistics, but not the instantaneous state of the atmosphere, that these stochastic eigenmodes are Laguerre-Gauss (LG) modes with beam waist chosen adequately relative to the field coherence length in the receiver plane. We have shown in our analysis that these eigenmodes have the remarkable capability of enduring coherence during propagation in turbulence. The ability to order the strength of individual stochastic eigenmodes notwithstanding turbulence is fundamental. It allows using of optimal transmission strategies that only consider the small number of most robust modes and significantly outperform conventional fixed mode sets under weak or strong turbulence conditions. Using these eigenmodes as transmit and receive bases minimizes signal degradation by turbulence, and minimizes the complexity of any signal processing for compensating the impact of turbulence. Adaptive optics can be replaced by adaptive multi-input multi-output signal processing, enabling compensation of fast fluctuations of both phase and amplitude.
Optical satellite-to-ground communications are disrupted by the effects of atmospheric turbulence which causes scintillation and phase distortion. By correcting the wave phase perturbations, adaptive optics allows the beam to be coupled into a single-mode optical fiber at reception with minimal fading. When the satellite is close to zenith, the scintillation rate is a few percent: conventional adaptive optics techniques are well suited. On the other hand, when the satellite elevation is low, the scintillation is high and conventional wavefront measurement techniques are no longer usable. We propose here a new approach to couple an incoming beam disturbed by turbulence with a set of propagation modes representative of a spatial multiplexer, including in the case of strong scintillation. We illustrate our approach by considering the case of the Laguerre-Gauss modes.
Long range horizontal path imaging through atmospheric turbulence is hampered by spatiotemporally randomly varying shifting and blurring of scene points in recorded imagery. Typical mitigation strategies employ software algorithms that combine optical flow to estimate and reduce tip/tilt aberrations with lucky patch identification and data fusion to filter higher order aberrations. In practice, accurate and fast optical flow estimation in turbulence faces is highly challenging. Here we investigate if these challenges can be overcome by using a neuromorphic camera. These sensors measure logarithmic changes in scene radiance in each pixel instead of the radiance itself. The changes are thresholded to produce a stream of events, which enables an ultrafast time resolution on the order of microseconds and enables efficient optical flow estimation at very high frame rates. Here we report on our initial experiments, where we have used a neuromorphic camera to image through turbulence in a controlled indoor setting. We analyze how the sensor responds to the turbulence induced apparent scene motion and propose an algorithm for computing high resolution images of the scene from the event stream in combination with intensity frames. Our initial results do not show a statistical relation between event counts and lucky patches. The image reconstruction based on the event stream did show promising improvements in output image sharpness and stability compared to the raw image stream, with many opportunities for further improvement.
Session 5 : Ground-space optical links : concepts and applications
The demand from satellite operators for ever increasing capacity during these years has led to the emergence of VHTS (Very High Throughput Satellites) system solutions. Such satellites exhibit as a minimum several times the capacity of conventional satellites and possibly up to Terabit per second and beyond. One of the main challenges associated to the implementation of these very high capacity systems is to feed the satellite in an efficient way to limit the cost of the system, in particular that of the ground segment. While RF bands are facing saturation and are submitted to strict frequency regulation, leading to a large number of ground stations, optical feeder links are considered as a promising technology to meet the future VHTS system requirements while strongly reducing the ground segment. Nevertheless, optical feeder links are still facing some implementation uncertainties, beyond the obvious issue of nebulosity which can be alleviated through site diversity approach. While several feeder link architectures are envisaged leading to significant implementation differences, atmospheric propagation impairments and their mitigation techniques together with high power generation and management as well as efficient modulation use are of primary importance in the design and sizing of the optical feeder link. During past years several experimentations on ground or in-flight have demonstrated part of these concepts and subsystems necessary to implement such high capacity systems. Simultaneous combination of all these concepts in comprehensive demonstrations has however not been implemented yet. This paper recalls the main drivers of satellite systems design based on optical feeder links and introduces the H2020 VERTIGO project that is specifically addressing the topic of the GEO-ground optical link and the associated technological challenges (high optical power generation, high efficiency waveforms, atmospheric impairments mitigation techniques).
Forward Error correcting Code design (FEC) and implementation for optical feeder link is reported. The designed LDPC codes present very good performances (BER < 1e-9) while requiring less than 20% hardware resources of FPGA compatible with space constraints. Protection scheme based on these LDPC codes combined with a physical layer bit interleaver is presented. Links sizing for GEO optical feeder link at 30° elevation angle under strong turbulence shows that uplink might require up to 100W per link.
High capacity GEO-Feeder optical links are very promising but constitute a real challenge for the whole transmission chain. The modeling and mitigation of turbulence effects is a particularly critical issue that drives the other subsystems design requirements. Building on SAOST, our reciprocity based performance evaluation tool, we study the GEO-Feeder link scenario: adaptive optics error budget and performance in terms of link availability threshold, for forward and return links, as a function of the number of corrected modes and of the ground telescope diameter. Adaptive optics correction is shown to be essential both for downlink and uplink. Choosing a large diameter is highly beneficial for downlink, pointahead anisoplanatism is however shown to level uplink performance for diameters above a few tens of centimeters. The article also demonstrates that the reciprocity, beyond providing powerful numerical tools, is a key principle for an indepth understanding of the physical effects impacting adaptive optics assisted ground-space optical links. These theoretical results are complemented by a brief presentation of FEEDELIO, the first experimental demonstration of uplink pre-compensation performed on a slant line of sight representative of a GEO-Feeder link.
Session 6 : Free space communication links : experimental demonstrations
This study characterizes a horizontal atmospheric telecom channel in terms of evolution time of the intensity and the phase of the electromagnetic field received. The channel temporal characterization comes from a turbulence data base recorded during a Cn² profiler experiment (SCINDAR). The coherence times of the intensity and phase for an overcast and unclear sky are computed, these values bring information on the bit rate achievable and the speed of an adaptive optics system.