Laser Communication and Propagation through the Atmosphere and Oceans XIII

This research presents the analysis of Resistance Temperature Detectors (RTDs) at the Townes Institute Science and Technology Experimentation Facility (TISTEF) 1 kilometer laser range. The RTDs produce nodal turbulence measurements that leverage the precision of PT100 RTDs. The Wave Propagation Research Group (WPRG) aims to complement TISTEF's current fleet of sonic anemometers with an additional precise nodal measurement of turbulence strength. The system consists of four, four-wire RTDs yielding differential temperature measurements used to estimate the Refractive Index Structure Parameter (Cn2). These data are then validated against a co-located SATIA-3A Sonic Anemometer to confirm accuracy.

This paper describes the design and implementation of a machine learning model for forecasting the refractive index structure parameter (Cn2). The method builds on previous work accomplished by Shaun Comino in using meteorological data to estimate Cn2. The AutoGluon framework is explored as a tool for both preparing data and the training process. Several standalone models as well as ensemble models are trained and tested against each other using locally collected data, and a BLS900 scintillometer was used as the ground truth. The model was trained and validated using data collected on the TISTEF 1 kilometer laser range between June 2023 and January 2024. Several numerical weather predication models are investigated and implemented to improve forecasts.

Free-space optics holds the potential for data communications at high bandwidth and security while minimizing size, weight, and power. However, the effects of atmospheric turbulence on an optical beam limits and degrades communication performance and bit-error-rate. Although degradation of beam quality occurs due to many factors, typically unwanted aberrations due to fluctuations in the refractive index along beam path causing scattering, absorption, and beam wander is the main cause. Randomly distributed cells called eddies are formed in the propagation medium giving rise to turbulence as well. In our first study we characterized the beam on a 3-meter link in a laboratory environment. In this paper, we report experimental results with similar test conditions on a 50-meter FSO path and compare to the shorter path length results. We also correlate scintillation data to beam wander and spread. A phosphor-coated silicon CCD is used to record and study the beam’s intensity profile.

Time-lapse imaging data has been used to solve the inverse problem of atmospheric refraction along multiple optical paths. Relative refractive index gradient values have been recovered from the shift results found in the field-imagery data. The gradient values derived from filed imagery were compared with ones predicted by Numerical Weather Prediction (NWP).

Awaiting Publication Release

This paper describes research conducted by the U.S. Army Combat Capabilities Development Command (DEVCOM) Analysis Center (DAC) to characterize the supercontinuum (SC) effect generated from interactions between ultrashort pulsed lasers (USPL) and commonly used optical materials. DAC successfully induced SC generation (SCG) in an outdoor setting through atmosphere – a feat that has not been documented nor attempted successfully until now. This work is the first instance resulting in calibrated spectral fluence data of SCG in bulk dielectric materials, from experiments conducted in an outdoor environment, where a USPL beam was propagated hundreds of meters downrange. The purpose of DAC’s USPL project is to develop analytic tools and methodology which are needed to assess the effects of USPL on the functionality of electro-optical sensors. Characterization of the USPL propagation in atmosphere is needed to understand the potential degradation and develop the associated analytic tools and to understand the influence varying atmospheric conditions have on the beam.

This presentation discusses the results of the analysis, numerical simulations, and experimental results of the physical phenomena related to the interaction of the laser pulse train (LPT) with air. This study considers the case when the intensity of the micro-pulses in the LPT is far below that of tunneling ionization. The focused laser beam ionizes the air forming a plasma filament. In our model, a low intensity LPT photo-ionizes background negative ions (produced by ambient ionizing radiation) and provides the seed electrons necessary to initiate collisional ionization. As the plasma density builds up on axis, the latter portion of the LPT gets defocused, resulting in scattering of the incoming laser radiation and shortening of the laser-plasma interaction length. The train of ultra-short laser pulses of adjustable repetition rate controlled by the intra-cavity longitudinal mode selector. The technique allows for generation of LPTs with a total length ~ 60 ns that consists of the micro-pulses with variable 10 ns to 0.45 ns separation time. In the performed experimental studies the intensity of the transmitted LPT depends upon the total energy of incoming LPT.

The propagation of laser pulse trains (LPT) through the atmosphere is analyzed and simulated. The effects of group velocity dispersion, focusing, chirping, and atmospheric turbulence are included in the analysis. We present a generalized form of Huygens principle to describe the LPT propagation. The model is general and is solved for pulses longer than several laser wavelengths. We present examples of the propagation of LPTs with parameters chosen to replicate recent experiments which use LPTs to generate rf radiation in the atmosphere. Propagation of LPTs over distances of ~1 km is of particular interest. The pulse length of the individual micropulses used in the simulations ranges from ~100 fs -100 ps. The rf generation mechanism that has been reported requires propagation of LPTs over these large distances for various applications. The rf frequency generated is determined by the micropulse repetition rate.

For various applications involving the propagation of light through the atmosphere, anisotropy of optical turbulence must be accounted for. At this point, however, there is no consensus about how to realistically model anisotropic turbulence. It is well established that at length scales small compared to a certain outer length scale, L_0, optical turbulence is locally homogeneous and isotropic and is well described by the Obukhov-Corrsin similarity theory. At scales large compared to L_0, however, the turbulence is usually anisotropic, and Obukhov-Corrsin similarity is no longer valid. In this paper, we discuss two questions: first, how to define and predict L_0; and second, how to model the 3D refractive-index spectrum, Φ_n (κ_x,κ_y,κ_z), for wavenumbers comparable to and smaller than 1/L_0.

An analysis of the coherent mode decomposition of an optical field after propagation through atmospheric turbulence is presented. The coherent modes represent an ideal basis by which to decompose the field for the design of mode-limited optical systems and systems which perform atmospheric mitigation based on mode-sorting. The similarity of the coherent modes to standard free-space and fiber modes is quantified. The eigenvalue spectrum yields sharp bounds on the efficiency of receivers using mode-sorting techniques; a general statistical model is presented for fading and loss in fiber-coupled free-space optical receivers based on the Weibull distribution.

We emphasize the requirement for a validated simulation environment to assess the impact of changes in atmospheric turbulence strength throughout the day and year on the performance of atmospheric electro-optics (EO) systems such as directed energy, power beaming, and laser communications. We show that continuous measurements of the refractive index structure parameter (Cn2) and meteorological data can be utilized in wave-optics simulations of atmospheric EO systems. We use continuous (24/7) Cn2 and meteorological data measurements on a 7 km atmospheric propagation path at the University of Dayton test range. We used this large dataset for predictive modeling of laser beam propagation parameters and their fluctuations during diurnal and monthly cycles. We analyzed a specific laser beam director system configuration, both with and without the implementation of adaptive optics beam control. The proposed technique provides a pathway to a comprehensive predictive performance evaluation of diverse atmospheric EO systems, allowing for statistical and interactive parametric analysis involving the simple adjustment of pre-selected system and propagation path parameters.

Recent work has shown that calculations of scintillation using wave optic simulation provides a much closer match to experiment than analytic approaches, such as extended Rytov theory. However, link budgets using analytic theories can be calculated much faster than those using wave optic simulation. In this work we pre-calculate wave optic simulations for uniform horizontal links using dimensionless parameters that allow application to a wide variety of cases. We then develop approximations to the pre-calculated values that allow quick computation.

This presentation explores the possibilities of using AO, which is widely applied to the correction of turbulent phases in astronomy but is now used in free-space communications (FSO). Optimal controllers have been demonstrated to be a helpful way of improving the error budget on extremely demanding AO systems in astronomy. Still, no recent study shows the full benefits of implementing an optimal control AO in a working FSO communication system. Here, we present our first results after implementing a closed-loop AO system in the Universidad de los Andes laboratory that allows us to test real-time controllers and implement the AO correction directly to FSO communication systems.

We propose the use of orbital angular momentum (OAM) states to form modulation symbols in a free-space laser communication link affected by atmospheric turbulence. A collection of vortex superpositions with 1 to 4 active OAM states are used for transmitting digital information. Sensing and decoding data is performed with three candidate spatial detection strategies, including a Mode Sorter and a Shack-Hartmann. We use concepts of Optimal Transport, particularly the Wasserstein distance, to finding optimal selection of OAM superpositions in which turbulence is taken into account. Symbol detection and classification is performed as a whole, without the need to individually detect the OAM states. We demonstrate the effectiveness of our approach through both simulation and trials using an outdoor experimental setup.

Laser based applications leveraging coherent laser sources are subject to optical distortions created by random media containing particles with radii comparable to the optical wavelength. Light propagation through such media can be approached using statistical optics by the implementation of phase screen simulations. A phase screen represents a statistical realization of the optical phase shifts accumulated during propagation derived from a medium’s optical phase power spectrum. The phase power spectrum of a turbid medium is connected to the medium’s particle distribution by the medium’s volume scattering function using radiative transfer theory. Such a model is limited to small-angle forward scattering and small propagation geometries to prevent aliasing by high angle scattered light. Using this relationship, the statistics of the generated phase screens are studied and applied to the propagation of optical vortex beams through turbid water. Experimental orbital angular momentum mode spectrum measurements and intensity attenuation data are presented showing good agreement with the model.

The explosive growth of satellites in low Earth orbit (LEO) demands advanced surveillance and communication capabilities. However, atmospheric turbulence hinders high-resolution imaging and high-speed communication through optical wavelengths, which remains the only viable option. SEETRUE (Sharp wavefront sEnsing for adaptivE opTics in gRound-based satellite commUnications and spacE surveillance), proposes a game-changing solution: cost-effective, AI-driven wavefront sensing for Adaptive Optics (AO) in optical ground stations. It features a unique ground station equipped with a 50 cm robotic telescope with AO capability, a multi-purpose 38 cm binocular telescope, and an atmospheric profiling system. AI-powered wavefront sensors (WFS) within the system leverage novel turbulence models and a revolutionary "end-to-end" design approach to maximize information extraction. This enables compact and low-cost AO solutions, overcoming a major barrier to widespread adoption. Paving the way for a future with accessible and affordable space communication and surveillance for all.

We present experimental results and analysis of collimated 2.09-micron laser propagation across a 1-km coastal range through diverse, clear-weather conditions. Tests were conducted at the University of Central Florida's TISTEF range. By tracking the evolution of the beam’s centroid position and power distribution, we calculate beam wander and profile statistics. Beam wander is correlated with optical turbulence. As the beam profile evolves from a small to a larger Gaussian spot, and eventually to an even larger beam characterized by multiple regions of high and low intensity, we correlate profile beam size and quality with scintillation. By collecting data throughout the day, diverse atmospheric conditions are explored. A suite of weather stations provide temperature, humidity, wind speed, and solar radiation. Data collected by sonic anemometers and scintillometers quantify optical turbulence by providing estimates of the refractive index structure parameter. A comparison between the present work at 2.09 microns and prior work at 1.535 microns will reveal unique beam quality dependencies between wavelengths and weather conditions.

In January 2024, the Wave Propagation Research Group (WPRG) successfully conducted an outdoor lasing test to gather data for developing predictive Artificial Intelligence (AI). This experiment employed state-of-the-art instrumentation and multiple cameras positioned at both receiver (RX) and transmitter (TX) sites, featuring both passive (non-emitting) and actively-illuminated (lasing) imagery. This study aims to advance WPRG's existing image-based methods for estimating variations in phase and scintillation resulting from optical turbulence. Additionally, an estimation of Greenwood's frequency will be conducted by analyzing sequential frames from both active and passive states.

A multiple-input single-output (MISO) configuration is used to study the effectiveness of spatial diversity on reducing single-photon losses caused by atmospheric turbulence for quantum communications over free-space. The system consists of two parallel transmitters and a single receiver, in which a single-photon signal is transmitted through one optical path at a time. An outdoor experimental campaign was conducted to obtain results under real turbulence conditions.

This study develops a machine learning framework to impute missing Fried parameter values, crucial for assessing atmospheric turbulence's impact on optical systems. By leveraging data from a sky monitor, including atmospheric conditions and imagery, the research addresses the challenge of continuous data collection. The approach, which separately imputes nighttime and daytime values, significantly improves accuracy, offering insights into atmospheric dynamics and aiding in the optimization of optical communications.

This presentation will depict recent results from a co-located multi-wavelength transmissometer. The system uses a single telescope to transmit and receive a temporally modulated pulse trans from multiple super-luminous LEDs (SLEDs). The signal return is facilitated by retroreflector(s) mounted in a static ground-based array or mounted to a dynamic Un-crewed Aerial System (UAS). The system measures EO transmission along propagation vectors to develop hemispherical ground truth timeseries datasets of wavelength dependent transmission and absorption parameters that will improve the accuracy and speed of test and evaluation of EO atmospheric propagation models.

This paper explores the fundamental phenomenology of weather-driven diurnal and nocturnal optical turbulence trends. Examining long duration persistent atmospheric measurements at Townes Institute Science and Technology Experimentation Facility (TISTEF), an outdoor laser range operated by the University of Central Florida (UCF), reveals key correlations between observed meteorological quantities and optical turbulence strength. A distributed set of meteorological instruments provide information on local conditions via temperature, pressure, relative humidity, net radiation, wind anemometers, cloud ceilometer, and a sky imager. The strength of optical turbulence is captured via a boundary layer scintillometer (BLS) and the delayed tilt anisoplanatism (DELTA) sensors. Additionally, the paper compares the turbulence measurements against the performance of a physical weather-driven turbulence model and a deductive machine learning (ML) based turbulence model. These models attempt to accurately capture the relationship and phenomenology between meteorological conditions and optical turbulence.

Laser communication link performance in free space depends heavily on atmospheric conditions present on the propagation path. Distortions due to atmospheric turbulence, such as scintillation and beam wander, greatly diminish signal detection and performance. In order to improve the communication link, experimental measurements and analysis of turbulence strength are presented as useful metrics in determining the system’s detection limits. Experimental optical links were tested in a 1-km horizontal propagation path in order to study and compare different low-regime turbulence scenarios through the Rytov approximation, and their spatial spectrum properties.

Free space optical communication (FSOC) provides high capacity and data security without the spectral allocation challenges of RF systems. Optical scintillation in atmospheric FSOC links causes both spatial and temporal distortions on the received beam, spanning many milliseconds and causing signal fading that can result in significant data loss. In-situ monitoring of scintillation experienced by a given FSOC system in operation can not only provide a better understanding of the underlying distribution functions of various scintillation conditions for link availability planning but also inform more efficient error correction protocol techniques that can better adapt to the changing atmospheric channel conditions. In previous work, we demonstrated a tone-based irradiance variance characterization approach that can provide real-time measurements of the scintillation index, power spectral density, and distribution function of scintillation using an FSOC system’s data modulation envelope. Here, we expand this capability to operate on modulation envelopes with data rates in excess of 1Gbps while simultaneous supporting high bandwidth data reception.

The development of quantum cascade lasers (QCLs) has enabled semiconductor based devices in the mid wave infrared (MWIR) region (8um-14um). The anticipated properties of this range with regards to photon transmission in foggy weather conditions is highly beneficial for communications through the atmosphere in the full range of weather conditions. In this paper we experimentally describe and demonstrate the performance of a 10um FSOC system using artificial smoke as a replacement for fog. We show successful communication with a high data rate in poor visibility. Additionally there will be a discussion of future requirements for components in the MWIR region.

During a series of tests in January 2024, a 1064 nanometer continuous-wave laser beam was focused and propagated on the TISTEF 1 kilometer range. The optical path was instrumented with a variety of meteorological instruments, and beam profile data was measured using multiple cameras. The results of the propagation data were explored as a function of measured atmospheric conditions.

Pending completion of the publication release process.

We investigate the atmospheric-fade mitigation properties from the combination of temporal interleaving at timescales comparable to atmospheric fade durations and repeat-coded waveforms with similar timescales. Defining L as the the ratio of statistically-independent fade events during each interleaved and framed code-word, and Q as the duplicative factor for a repeat-coded waveform structure, we analyze how these parameters interrelate to provide optimal immunity to atmospheric fading. This study includes the development of a model and validation on an experimental modem testbed where various scintillation conditions are emulated.

The use of free space optical communication from deep space promises to revolutionize the capabilities of human and robotic exploration, enabling modern-internet-like data rates throughout the inner solar system. Such an improvement will enhance the potential of space science and exploration by enabling higher rate instruments and imagers, lower latency operational capabilities, and less data lost due to limited on-board storage. The adoption of optical communication in National Aeronautic and Space Administration programs, however, has faced numerous budgetary and technical challenges. Building up the Earth-based telescope infrastructure to support high-rate and energy-efficient optical communication from deep space requires significant capital investment that may take years to become operationally significant. The use of many small telescopes in an array is one solution to this problem. We propose a scalable and cost-efficient telescope array architecture based on the Deep Space Optical Communication project, and analyze the primary communication impairments and performance capabilities for a standards-compliant downlink waveform.

We introduce an enhanced all-optical free-space optical (FSO) transceiver utilizing our polarization-independent liquid crystal on silicon (PI-LCoS) phase modulators. By actively correcting for atmospheric turbulence and beam wander, the FSO transceiver enables all-optical fiber-to-fiber signal transmission, minimizing latency and loss. The specialized analog PI-LCoS phase modulator, featuring a unique silicon backplane, provides polarization independence, high grayscale, low phase flicker, and frame-at-a-time update capabilities.

The effectiveness of free-space laser communications is limited due to wavefront deformations caused by atmospheric optical turbulence. To determine these deformations, we propose a wavefront sensor that utilizes the angular selectivity of an optical transmission filter to measure the first derivative of the wavefront, i.e., its local gradients. The transmission filter converts the gradients of the incident wavefront into an intensity distribution. For each direction (x and y) this distribution is captured twice, with different angles between filter and optical axis for each measurement. The contrast of both measurements, calculated for each pixel of the used detector, and the local gradients of the wavefront, integrated across each pixel, have a nearly linear relation. Algorithms developed for the Hartmann-Shack wavefront sensor can be used to reconstruct the wavefront from the local gradients, which are obtained from the contrast measurements. Simulations demonstrate the applicability of this wavefront sensor in atmospheric turbulence.

In problems involving optical wave propagation in the atmosphere it is common to assume turbulence statistics obey a Kolmogorov power spectral density model. It is well known that this model has a key weakness in that it does not model the outer scale. Also, some observations indicate a deviation from the Kolmogorv -11/3 power law often referred to as non-Kolmogorov turbulence. Understanding the effect of a change in power law has been a difficulty for some researchers and most assume the turbulence volume is homogeneous, which is unlikely. In this work, I describe how spherical wave anisoplanatic error can be described for inhomogeneous turbulence using the mean turbulence height and the generalized Fried parameter. I also consider a two scale outer scale model and describe its effect compared to the single parameter von Karman model.

Microplastic pollution is a salient problem in large bodies of water such as lakes and oceans. Study of that kind of pollution, therefore, is of interest. Some large drawbacks of current identification techniques are that they are expensive, time intensive, and lack standardization procedures. Here, a novel pollution detection method has been developed, which which records laser-light scattering induced by said microplastics at multiple angles. Adding a camera allows for additional measurement regarding size and shape. We believe this method of detection is not only useful for determining the composition and size of plastic, but also allows for temporal measurements regarding concentration.

Coherent light transmission through the atmosphere, as occurs with laser propagation, is significantly affected by fluctuations in the refractive index of air. The fluctuations depend on pressure and temperature. Conventionally, phase screens, which follow assumptions of thin turbulence regions and turbulence isotropy, are used to model optical propagation and imaging. An alternate model for optical turbulence is proposed, employing a spherical bubble packing approach to accommodate the effects of shear and inhomogeneity. The bubbles representing the broad spectrum of turbulent length scales are assigned to radii based on a linear-eddy model and refractive index values based on a refractive index spectrum. The effects of path, refractive-index structure constant, and length scales are investigated by ray tracing and imaging analysis. The imaging techniques are verified by a numerical wave propagation approach with phase screens and validated with field anemometer and scintillometer measurements for assessing model inadequacy and uncertainty.

Atmospheric turbulence can significantly impact the quality of an image caused by distortions during image acquisition. The turbulence can affect the image sharpness and contrast. We will quantify atmospheric impacts on image quality by using standard image quality metrics (such as the modulation transfer function (MTF), SSIM or PSNR). In this paper we will utilize meteorological and contrast targetboard observations, as well as synthetically-generated images, to train machine-learning models. We utilize approximately two years of optical turbulence data and meteorological data along a coastal path to train and test the models. A comparison between various ML algorithms, such as Bagged Trees and LightGBM (Light Gradient Boosting Machine), is performed to determine which best predicts the image quality metrics.

As the need for “data on demand” increases, the likelihood of users moving to hybrid systems is very high. Today, multiple users in both commercial and government sectors are looking to integrate both laser communication (lasercom) and RF solutions onto their platforms. The goal of this work is to train a Alexnet Convolution Neural Net (CNN) on Doppler radar imagery to make decisions on which system will provide the highest performance in differing atmospheric conditions.

This study evaluates an AI model utilizing differential imagery and meteorological data for real-time prediction of the Refractive Index Structure Parameter (Cn2). The Wave Propagation Research Group (WPRG) builds upon previous work performed by Anthony Eppley to abstract turbulence statistics from passive imagery in real-time using image subtraction. Developed using the TensorFlow Functional API, the model integrates these subtracted images with meteorological timeseries data, including temperature, humidity, and radiation metrics. Advanced techniques such as K-Fold Cross-Validation and Grid Search are employed to refine the model for greater accuracy and robustness. The meteorological data will supplement the model for feature importance analysis of the differential frames. To validate this integration, the model's accuracy is evaluated with TISTEF's current suite of instruments.

When an intense laser interacts with a gas of atoms, high-order harmonics are generated. In the time domain, this radiation forms a train of extremely short light pulses, of the order of 100 attoseconds. Attosecond pulses allow the study of the dynamics of electrons in atoms and molecules, using pump-probe techniques. This presentation will highlight some of the key steps of the field of attosecond science.