Remote sensing of the atmosphere and the earth surface can aid in the solution of some fundamental scientific and national economic problems. These problems include meteorology, weather forecast and climate evolution, studies of the natural resources of the earth and environmental control. A satellite borne infrared scanning spatial radiometer is an important tool for the study of different natural and anthropogenic objects from space. A new infrared scanning radiometer having an efficiently cooled detector forms a part of the `Meteor-3' meteorological satellite. In this report the technical characteristics and principles of its operation will be described. Results of the data gathered during flight experiments will be presented. The technical problems in obtaining a cooling system with a long life-time and reliable operation will be described. Results obtained from `Meteor-3' show that these radiometers can provide global and local hydrometeorological information and can also help in the solution of ecological problems such as those found in the region of the Aral sea.

Recent developments in a spectral sensing method for the determination of IR atmospheric parameters are reported. Theoretical and practical aspects of the minimum information type inversion for the retrieval of temperature and water vapor profiles from IR emission spectra are presented. Two evalua- tion procedures developed to estimate first guess profiles of temperature and humidity based solely on the measured spectrum are described. The retrieval methodology is tested on new measurements performed at a resolution of 1 cm^-1 and zenith angle of 45 degrees. The potential of the approach is verified through several MODTRAN2 calculations performed to compare transmittances generated with radiosonde profiles and retrieved profiles. Finally, it is proposed to use the spectral sensing method as an aid for predicting slant path transmittances.

The utility of optical satellites depends heavily on latitude and cloud cover. In the north-eastern corner of the Fenno-Scandinavian landmass, extreme seasonal variations in sunlight and the clashing of weather fronts severely limit the season for optical satellites. In this paper, we examine the distribution of cloud-free Landsat images of the Barents Region during 1984-93. Based on cloud-cover data from ground station archives, we derive the observed frequency of cloud-free images (having less than 10 per cent cloud cover) of each of 10 scenes over 10 years. The data show that cloud-free images are very rare, and that they occur as isolated events in time. If the cloud-free images are essentially randomly distributed in space and time, they can be modelled by a Poisson distribution with unknown mean. Invoking the method of maximum likelihood, we estimate this parameter by the sample mean and hence compute the estimated expected frequencies. We find that they agree well with the observed frequencies. Next, we perform a X2 test to check the validity of the model. The computed test statistic falls below the critical one. Hence we conclude that the cloud-free images are randomly distributed in space and time.

A computer system for predicting and taking into account the effect of the totality of linear optical phenomena in the real atmosphere on the accuracy and energy characteristics of optoelectronic systems and devices is presented. Computer system of estimation of atmospheric effect on propagation of optical radiation is a set of application programs relying on the engineering techniques and summarizing the results of fundamental scientific studies in atmospheric optics. The system can provide versatile service and makes it possible to present tables, diagrams, and graphs promptly. It can be used a s training program for students. The computer system consists of two packages of programs for IBM PC/AT.

The U.S. Air Force Phillips Laboratory is developing a differential absorption light detection and ranging (DIAL) system for use in long-range remote sensing of trace atmospheric species. A wavelength-agile (WAL) transverse-electric-atmospheric (TEA) carbon dioxide laser operating on P- and R-branch transitions in the 9.4 micrometers bans is used as the pulsed radiation source for the system. A master oscillator-power amplifier (MOPA) WAL configuration is planned for the future to achieve the necessary increase in pulse energy required for extended range operation. The key system components, including the WAL source and power amplifier, transmitter, and receiver optical systems, and data collection equipment are described. The results of preliminary tests using and SF₆ absorption cell in a 50-m laboratory path and filed measurements using a 3.2 km path at the Phillips Laboratory Starfire Optical Range are presented. Performance predictions for operation using the WAL source alone and for the MOPA configuration for extended ranges are presented and discussed.

Some peculiarities of measurement of atmosphere temperature profiles with help of Doppler method of radio-acoustic remote sensing in short-wave range are investigated. For this purpose the expression of power spectrum of received signal is investigated at the point where the wind drag is compensated. We suppose that the periodical series of electromagnetic pulses is scattered on the single acoustic pulse and take into account the mutual influence of turbulence and wind in atmosphere. We find that turbulence and wind lead to widening of the frequency spectrum of the received signal. Naturally this circumstance decreases the exactness of this method. It is also shown that for the determination of local atmospheric temperature in this range of wave it is necessary to use the particular sound frequency for each altitude. Moreover, we find that this method depends heavily on the magnitude and direction of the wind. Calculations show that for obtaining an exact temperature profile it is necessary to use the short-wave narrow-angle antennas.

The possibility of the atmosphere remote sensing by the radio signals is considered. The base of this approach is contained in the solution of the inverse problem for the electromagnetic wave propagation from the point stationary or non stationary source in the spherical layered atmosphere. The inverse problem procedure is found by the imbedding method. This procedure have been developed earlier for many dimension media and have been adapted in this work for media with spherical symmetry. The procedure allows to restore the height profiles of the atmosphere refraction index by the measuring of the spatial time or spatial frequency distribution of the field. The cases of the impulse and monochro- matic sources and the method of the date obtaining are discussed. The proposal remote sensing may be realised by two satellites moving relatively each other. One of them transmits down and simultane- ously other receives the radio signals which are reflected by the tested area of atmosphere. The various polarisations of the source and the inversion of the atmosphere dissipation are discussed.

The theoretical basis for wind velocity field parameters' measurements using CW Doppler Lidars (DL) is outlined. The instant power spectrum of photocurrent is shown to be histogram of velocity projection field with a weight depending on Dl's parameters. A non-destructive long-range method is presented for measuring the structure constant of the wind velocity field. The essence of the method is (the constancy of the backscatter coefficient is assumed): the average square width of the DL photocurrent power spectrum is the averaged structure function of the wind velocity weighted with a function depending on the DL parameters, the measurement time and the average wind velocity. Consequently, it is proportional to velocity structure constant with a factor depending on the DL parameters, the measurement time, the average wind velocity, the internal and external scales of turbulence. At certain DL parameters and measurement time the factor is shown to be practically independent from the average wind velocity, the internal and external scales of turbulence, thus a priori estimate can be successfully used instead of their real values. The differences between structures of CW and pulsed DL signal are discussed. The described above method is applied for pulsed DL.

The Planck law holds that brightness of a real thermal radiation source can be expressed in tenns of radiant emittance M(A,T) of absolutely black body (ABB) and function c(2) which establishes similarity of emitting properties between a real body and ABB. If in the spectral interval . the selective properties of the radiation source can be neglected, then the coefficient of body blackness c is often used in place of function c(A). During the past several years when the spacecraft measurements were carried out to investigate the Earth's natural resources the approach based on the assumption that each element of the underlying surface represents a secondary radiation source was employed1. If the selective properties of radiation of this source are neglected, then its spectral function of brightness can be represented as the Planck relation by replacing there the energy temperature T by the color temperature T. In the visible spectral range the Earth's atmosphere can be considered nonselective since several narrow water-vapor absorption lines and an ozone absorption line (Chappear) do not strongly affect its general form. In this case it is also possible to extend the approach used for investigating the natural resources to studying the Earth's atmosphere. In our paper we discuss some calculational results of determining T and c of the solar radiation brightness field scattered in the atmosphere.

In the case of design of atmospheric optical remote sensing systems functioning with dense clouds it is necessary to take into account both multiple scattering and absorption of electromagnetic waves on water aerosol. These factors have an especially strong effect on the propagation of infra-red light beam on large distances. In the report a thick layer of chaotically distributed absorbing particles with narrow indicatrix of scattering is considered. A theory of radiation transfer is developed for the case of oblique lighting of the medium boundary. As a result it is managed to analyse some peculiarities of evolution of angular and spatial structures of the wave beam. The behaviour of the last one is anomalous in comparison with a case of normal illumination of the boundaries. Then occurred a very quick: increasing of widths of angular and spatial distributions of light intensity in some interval of distances. So they depend on the height nonmonotonicaly. Also maximums of the angular and spatial distributions anormalously quick shift themselves towards the boundaries normal direction. The investigations are made both analytically and numerically.

When observing a remote source through the Earth atmosphere the light scattering gives rise to a spurious signal which dims the source and compromises the measurement interpretation. In this paper this problem is re-examined. Our work has been limited to find a theoretical modelling for the optical radiation collected by aerospace detectors for image remote sensing. The model relies on the assump- tion that 'large-angle-scattering' and 'small-angle-scattering' are quasi-independent phenomena. It is qualitatively shown that large-angle-scattering is mainly originated on the first optical path (from the top of the atmosphere to the ground), while the second one principally contributes to the small-angle- scattering. The radiative transfer equation (RTE) that accounts for this phenomenon (path) is assessed. It holds for a turbid atmosphere with a free size-distribution of aerosol whose density is allowed to change with the quota. Due to the use of an unknown reflectance distribution of the ground, which is assumed to be uniformly lighted, the radiation field is allowed to be non-uniform for each layer of the accepted plane-parallel geometry of the atmosphere. The RTE so far obtained is solved by using the 'successive-order-method'. The solution is expressed by a superposing integral of the observed reflectance field, which is not space-shift-invariant. Finally, an approximate technique is described which allows the 'inversion' of the measured radiance to be achieved, and so the retrieval of the reflectance field.

We report on the entropic properties of a stochastic radiation field. The degree of polarisation P of light, in the form of plane waves, is of the nature of an order parameter. The radiation entropy takes an analogous form to the entropy of one-dimensional Ising (two-level) spin systems in contact with a heat bath. On the basis of this analysis the degree of polarisation has a new thermodynamic significance. It is argued that within this representation, one may define an effective polarisation temperature and we show how it depends on the degree of polarisation. The results are illustrated by two examples: (i) the computation of the degree of polarisation of an incoherent mixture of partially polarised light beams and (ii) the problem of entropy production due to multiple scattering of light by a spatially random medium composed of uncorrelated and noninteracting spherical dielectric particles. Light transmitted through a multiple scattering medium is depolarised by decorrelation of the phases of the electric field components and its polarisation entropy increases. The effect of size of the spherical particles and of the optical depth on entropy production are studied numerically, using the Mie theory, via the Monte Carlo method.

We present the results of statistical simulation of upward flux of thermal scattered radiation in a wide range of variations of characteristics of aerosol extinction and scattering both in the atmospheric boundary layer and in the stratosphere. We carry out this simulation in case of spatially homogeneous Lambertian surface using the Monte Carlo method. We consider the effect of aerosol upon scattered radiance, characteristics of the adjacency effect as well as the accuracy of both the single scattering approximation and the conservative scattering model.

Due to the big distance between a spaceborne lidar and the sounding targets a great volume of the atmospheric domain is caught within the lidar receiver field of view and the multiple scattering highly affects the lidar returns. Variance reduction Monte Carlo method and analytical extension of the Mie theory are used for the calculation of spaceborne lidar returns from multilayered cloud systems. A main advantage of Monte Carlo techniques is that they allow the calculation of the solution with the desired accuracy. The analytical scattering extension of the Mie theory leads to analytical expressions of the n-fold scattered electromagnetic field and then to a generalisation of the optical parameters. The performance capabilities of identification of cloud layers from space has been evaluated. The retrieval of the extinction and optical depth in clear and cloudy atmosphere has been carried out by single scattering inversion methods. For clear inhomogeneous atmosphere in visible spectral region the use of Klett's inversion method leads to a reduction of about 10% of the retrieval values in lower 30 km atmospheric slab as compared with the initial data. Influence of transparent multilayered cirrus clouds results in a reduction of the retrieval extinction coefficient varying between 30-50%.

The subject of coherent optical propagation and imaging through atmospheric turbulence, which was extensively investigated after the invention of the laser, has received renewed attention in recent years, both as a tool for investigating the physics of the atmosphere and in view of a number of applications in adaptive systems, in remote sensing and environmental applications. In the present paper some recent progress will be reviewed, with particular attention to double pass imaging and to the problem of laser scintillation. In double pass imaging, methods have been developed and tested whose aim is the production of images free from turbulence effects. Progress towards the solution of the problem of scintillation includes approximate theory, modelling by means of numerical simulations, and simultaneous measurements of both laser intensity fluctuations and all relevant parameters.

A general analysis is presented for the scintillation index of a Gaussian beam wave propagating through a turbulent medium confined to a portion of the propagation path between input and output planes, the limiting case of which defines a thin random phase screen. For a plane wave incident on a phase screen located midway between input and output planes, the phase screen scintillation index agrees with that of a plane wave propagating through extended turbulence when the phase variance and spectrum for refractive index of the screen and extended medium are the same. However, for a finite Gaussian beam at the input plane, our analysis reveals that the distance from input plane to the phase screen must be many times greater than the distance from the screen to output plane to simulate a plane wave propagating through an extended medium.

Experimental observation of coherence enhancement phenomenon are presented and described. We have illuminated an optically rough moving target through turbulence using two coherent point sources and observed the interference of the scattered fields in the region close to the sources (the region of coherence enhancement) with the help of a special interferometric system. Our study shows that the intensity distribution in the region of coherence enhancement depends on the inner scale of turbulence.

A sharp peak is shown theoretically to exist in angular distribution of light scattered by a layer of ran- dom medium with weak refractive index fluctuations in the vicinity of backward direction. The peak width may be estimated by the ratio of incident radiation wavelength and the turbulence correlation scale. The effect is found to be formed by coherent addition of contributions from elementary layers, and may be considered as an analogue to 'weak photon localisation' phenomena in random media. When the width of scattering layer is large enough compared with the correlation scale of refractive index inhomogeneities, the differential scattering cross-section is composed by the 'coherent' component and 'incoherent' one, obtained earlier by V. I. Tatarski. In the case of visible or IR radiation (in contrast to microwaves) the 'coherent' constituent determines the scattering for large angles close to 180 deg. Quantitative estimates show the backscattering by turbulent layers in atmosphere to produce noticeable contributions to signal registered in remote sensing monostatic lidar experiments, and thus the effect under consideration has to be taken into account when interpretation of laser remote sounding data is carried out.

We discuss a technique based on nonlinear and adaptive optics for simulation of phase distortion effects in imaging systems. This technique uses a nonlinear two-dimensional optical feedback system to produce a controllable spatially and temporally varying chaotic intensity distribution. The intensity pattern is converted into a spatially varying thin phase screen using a spatial phase modulator. A chaotic phase distortion is then introduced into an imaging system's output image by propagation through the phase screen. A deformable (adaptive) mirror with computer control is used for simulation of large-scale phase distortions.

We describe recent results explaining the behavior of optical turbulence effects on the propagation of initially convergent (focused) laser beams. First we discuss the diffractive characteristics of an initially convergent Gaussian beam based on the geometric properties of the diffractive beam parameters. Some behaviors previously attributed to optical turbulence effects can instead be explained by the diffractive nature of the beam itself. We give expressions for the location and size of the beam waist, maximum possible distance from the transmitter to the beam waist, and the optical radius of curvature for focusing in turbulence. Laser, we present turbulent beam parameters consistent with our previously defined diffractive beam parameters.

The generalisation of phase screen model is applied for studying light scattering on atmospheric paths and extended light source image properties. The random phase at the screen by integration along a broken line between source, point on the screen and observer is obtained. The optimal position of the screen surface and its connection with altitude dependence on atmospheric turbulence is discussed. The limiting conditions for phase screen model application and some possibilities of its improving are considered. The random light field in the observing plane as a sum of light beams scattered by different parts of screen is represented. The local coherent structures, connected with each beam, are described. The random (partial averaged) correlation function for light field local properties studying is applied. The second moment of this function is calculated and its dependence on model parameters is studied. The connection between random correlation function and short exposure image of light source intensity is shown. The second moment of image intensity is studied and conditions of maximal contrast image are obtained. The approach to phase on the screen reconstruction is considered.

Previous study of imaging of a coherently illuminated object through turbulence has shown that the average energy spectrum contains diffraction-limited information of the object. It is generally assumed that the object to be imaged lies within a single isoplanatic patch. In this paper the turbulence-induced anisoplanatism and its effect on the imaging of a coherently illuminated object which is larger than the atmospheric correlation length is discussed. It is shown that the dependence of the wave structure function on the object co-ordinates may lead to the characteristics of superresolution, and the average energy spectrum may contain more information than diffraction-limited information in vacuum.

The directional diagrams of the coherent systems using the self-mixing effect for the stable and Fabry- Perot resonators and the directional diagrams of the coherent systems based on the heterodyne detection are compared. It is shown that a number of lobes of the coherent systems using the self- mixing effect depends on the index of the mode generation. The directional diagram of the coherent system for the lowest mode generation has one major lobe. The width of the major lobe for the lowest mode generation in the cases of the stable resonator with the plane output mirror and the Fabry-Perot resonator is determined by the diffraction angle as for the heterodyne detection. The major lobe of directional diagram of coherent system based on the heterodyne detection is wider than the lobe of directional diagram of coherent system using self-mixing effect for the stable resonator with the plane output mirror and narrower than the major lobe of directional diagram of coherent system using the self-mixing effect for the Fabry-Perot resonator. In the case of the stable resonator the side lobes are absent in the directional diagrams of the coherent system for the lowest mode generation. In the case of the Fabry-Perot resonator the first side lobe is 0.5% of maximum value and of the order of the fourth side lobe of the directional diagrams of the coherent systems based on the heterodyne detection.

The aim of this communication is to show how the Richardson-Lucy deconvolution algorithm can be applied to the blind deconvolution problem. After a brief description of the R.L algorithm itself, we start from the basic papers of Ayers and Dainty (1988) and Lane (1992) and introduce in their approach the R.L algorithm in several different ways. We show that the general behaviour of the proposed methods is analogous to that of the error-reduction algorithms and that good solutions can be obtained. The unregularized behaviour of the RL algorithm is overcame by a limitation of the iteration number. Moreover we compare the structures of the various algorithms proposed here and emphasise the main differences. The proposed algorithms are used to blindly deconvolve two types of objects (point-like and extended objects) blurred by simulated point spread functions similar to those observed at the focus of a small telescope in presence of at- mospheric turbulence. The error reduction term is given as a function of the iteration number.

The US Air Force Phillips Laboratory recently completed the Floodbeam Experiment (FBE), recording the first ever spatially resolved, coherent laser returns from non-augmented (non- retroreflectors), low earth orbit satellites. The experiment broadcast a near-IR, coherent laser at a selected set of low earth orbit satellites using a beam director and visible tracking system located at the Phillips Lab Starfire Optical Range (SOR). Tracking was accomplished during terminator periods when the satellite was illuminated by the sun and the transmitting/receiving site was in darkness. Thirty eight different satel- lites were illuminated during the experiment. The reflected laser return was collected with the 1.5m telescope at the SOR and focused on a low noise IR camera. The Floodbeam experiment gathered radiometric data, data on depolarization effects, and spatially resolved coherent speckle patterns. This paper will discuss the experimental hardware and the field results.

The US Air Force Phillips Laboratory's Retro-Assisted Imaging Laser Experiment (RAILE) demonstrated high-resolution imaging of a space object from the ground using an active illumination imaging technique. Results were obtained using a technique known as Imaging Correlography where images are constructed from measurements of backscattered (non-imaged) laser speckle intensity patterns. The target for the experiment was the Relay Mirror satellite that was designed and launched for a separate experiment known as the Relay Mirror Experiment (RME). The satellite carried four retroreflectors, arranged in a pattern, that comprised the actual imaging target. The satellite was tracked from a ground site and illuminated with a low-power, coherent, visible laser beam. The receiver was a simple linear array of intensity detectors, or light buckets. The use of an inverse synthetic aperture scheme allowed a modest effective receiving aperture to be realized (approximately 2 x 2 m). Although the effective receiving aperture for this experiment was no larger than that of a moderate-sized astronomical telescope, the light bucket concept is scaleable to much larger apertures. Image reconstruction was done in the computer after the sampled intensity data was collected by the receiver array. The paper contains a discussion of the imaging correlography technique, the experimental hardware, and the field results including reconstructed imagery.

An improved TeO₂ and Te infrared acoustooptic tuneable spectrometer has been analysed, using infrared fibres, a high speed frequency synthesiser and optimised algorithms. A comparison is made with the next best AOTF infrared materials, Tl₃AsSe₃, HgCl₂ and PbBr₂. A design study of the TeO2 and Te AO imaging spectrometer is also presented, operating in the two thermal bands, 1-5micrometers and 6-12micrometers , using an interchangeable fore-optics and a multiplexed electronically scanned infrared array cooled at 77 degrees K. Some initial experimental results indicate that these systems can perform well, an increase in the dynamic range in the 8-12 micrometers and is obtained compared to the 3-5*m band. It can be very useful in chemical process control, medical diagnostics, aerospace and earth remote sensing. Based on recent imaging spectrometer development, a design study of the TeO₂ AO imaging spectrometer in the 0.4-1 micrometers band, for simultaneous spectroscopy at every pixel, is presented, using a CCD camera and fast data processing technology.

IASI is an Infrared Atmospheric Sounding Interferometer devoted to the operational meteorology and to atmospheric studies and is to be installed on board the second ESA Polar Platform called METOP-1, planned to be launched in the year 2000. The main purpose of this high performance instrument is to record temperature and humidity profiles. The required lifetime is 4 years. This paper presents the characteristics of the LW IR detection arrays for the IASI spectrometer which consist of HgCdTe de- tectors. SAT has to develop the Engineering Model, Qualification Model and Fight Models of detectors, each having 4 pixels and AR-coated microlenses in a dedicated space housing equipped with a flexible line and a connector. An array is composed of HgCdTe photoconductive detectors. For this long wavelength the array is sensitive from 8.26 micrometers to 15.5 micrometers . The detectors, with sensitive areas of 900 x 900 micrometers ², are 100 K operating with passive cooling. High quality HgCdTe material is a key feature for the manufacturing of high performance photoconductive detectors. Therefore epitaxial HgCdTe layers are used in this project. These epilayers are grown at CEA/LETI on lattice matched CdZnTe substrates, by Te-rich liquid phase epitaxy, based on a slider technique. The Cd content in the layer is carefully adjusted to meet the required cut off wavelength on the devices. After growth of the epilayers, the samples are annealed under Hg pressure in order to convert them into N type mate- rials. The electrical transport properties of the liquid phase epitaxied wafers are, at 100 K, mobility (mu) over 150,000 cm²/V.s and electrical concentration N of 1.5 10¹⁵ cm^-3, the residual doping level being 1014 cm^-3 at low temperature. On these materials the feasibility study of long wavelength HgCdTe photoconductors has been achieved with the following results: the responsivity is 330 V/W. The bias voltage is Vp=300 mV for a 4 mW limitation of power for each element. The resistance of an element is around 30 (Omega) .The detectivity is: D* at (lambda) pic (FOV, F, (Delta) F)=2x1010 cm HZ^1/2W^-1 and NEP=0.5 nW. Measurements are made under Earth observing flux corresponding to the conditions of the PPF sun-synchronous orbit.

The theoretical treatment of the Earth's atmosphere effect on coherent measurements includes the analysis of radiation, propagation, scatter, and reception of arbitrary signals in a random medium in the presence of complicating factors such as regular refraction, finite sizes of receiving and transmitting apertures, and non-linear character of received signal processing. The complexity of the problem demands seeking approximate approaches which, on the one hand, provide acceptable accuracy and, on the other hand, are sufficiently simple to enable received signal processing analysis. These requirements may be satisfied only taking account of the specific character of each practical situation. In particular situations, the spatial structure of the radiation field can be described in terms of a finite beam wave or infinite plane or spherical waves. As regards a medium, it is necessary to include in consideration not only its random inhomogenities but also the regular variation of its refractive index in height. Under these conditions, the extended Huygens-Fresnel principle, the method of geometrical optics, the Rytov method, and the hybrid method are advantageous to use for a wave propagation description. The main advantage of these techniques is a possibility to represent the random wave field in an explicit form that is very important from the standpoint of the distorted signal processing analysis. Using the above-mentioned methods, the general scheme of measurement error calculation for a variety of coherent measuring systems can be developed on the basis of the consistent analysis of measurement process. This enables us not only to get simple formulas for the measurement errors but also to determine the limiting sizes of the receiving aperture and the limiting duration of the signal which are admissible in coherent signal processing.

The problem of the oblique incidence of a wave on the half-space of layered randomly inhomogeneous medium without absorption is considered. We analyse the influence of boundary conditions on probability distribution of the reflection coefficient phase, and statistical behaviour of the wavefield intensity moments inside the medium.

We consider a model three-dimensional stationary problem of wave propagation in a fluctuating weakly dissipative layered half space. The statistical analysis of this problem is carried out in the framework of the imbedding method for different boundary conditions. We are interested in the statistical behaviour of the averaged intensity, which is expressed via a two-frequency correlator of waves incident at different angles. The correlator is calculated both analytically in the framework of the diffusive approximation and numerically with the help of the statistical simulation method.

In the article the relationship between the accuracy of random spatial signal restoration and information quantity about signal, that is inherent in the random observing function, is considered on the base of information and statistical decision theory notions. It is shown that in special case of statistically homogeneous strong correlated signal in the observing area the obtained lower bound of signal restoration error dispersion is analogous to the lower bound defined with Cramer-Rao inequality which is widely used in statistical estimation theory for analysing of the discrete parameters potential accu- racy. The obtained results are used for accuracy analysis of satellite information systems which are employed in the optical wavelength region. The image restoration accuracy is considered in dependence of the influence of small particles ejected from satellite and the quantum noise being, the result of irradiance-photodetector substance interaction.

The high energy laser beam propagation though atmosphere is investigated by means of numerical simulation. The specific goal of the report is to develop and to approve the new algorithms of adaptive compensation of beam distortion associated with non-linear effects and atmospheric turbulence. The principle of aperture sounding is used for beam phase control in numerical model of adaptive array. The simplex method is chosen as a basis for control algorithm development. It is found that in regime of stationary wind refraction the simplex method ensures the greatest convergence rate over the gradient procedure. A comprehensive analysis of phase distortion structure in regime of nonstationary wind refraction allows to carry out adequate choice of control basis and to reduce number of controllable co-ordinates without degradation of correction quality. It is established also that it is possible to optimise the simplex size which is found to be depending on control duration and parameter of nonlinearity. For regime of wind velocity pulsations the alternating control strategy is worked out to compensate random wandering of the beam without control instability. It is also found that in presence of large-scale refractive index fluctuations the application of the simplex method does not degrade the compensation quality.

The purpose of this work is to develop and to test the approach combining path integral technique and complex-valued Monte Carlo method for calculation highest moments of the Green function of stochastic wave equation for a media with random small-scale inhomogenities in the background of large-scale inhomogenities. Calculations of the second and forth moments of the Green function and scintillation index have been performed for 1D and 2D case in the framework of three models: model of stochastic wave equation and models of parabolic and Markov approximations. The finiteness of the correlation radius of inhomogenities has been shown to be the reason of the significant difference between Markov approximation and two others. Comparison has been made in a good agreement with reliable results for 1D media. The Monte Carlo results have shown the existing singularities at the localisation centres and forming exponential decay of the second moment from the distances of about wave length. The unexpected sharp oscillations interrupting the exponential decay of the Green function moments have been obtained at the several tens average distances between scatterers from centre localisation. The effect of weak large-scale inhomogenities on behaviour of second moment have been also investigated.

We have constructed wavefront compensated system that measures wavefront tilt of laser through the generated turbulence of the convective tank in laboratory. The wavefront tilt power spectral density has been measured. The measurements of optics characteristic in tank have been also measured simultaneously. The spectra show distinct asymptotic slopes, in general, agree with theoretical pre- dictions based on the Kolmogorov model. This paper reports on the results of some experiments made to evaluate the performance of the system, and to improve characteristics of turbulence in the con- vective tank. Spectral analyses of these signals allow to study bandwidth of this system. We study the behaviour of image distortion in order that the compensation system can be established.

An infrared Transmittance database for fast radiative Transfert model To assimilate the satellite information, meteorological forecasting models need, at each iteration, an accurate radiative transfer model. This model must be very fast and line by line models are inadequate. So much faster models adapted to predefined filters, have been developed: Meteo-France has chosen the fast RTTOV model which uses, for working pressure levels, 3 independent polynomial modelisations with non-linear terms functions of the temperature, the specific humidity and the ozone of the input atmospheric profile. The coefficients of the polynomial equations are off- line calculated by linear regression on atmospheric transmittances, computed with a line-by-line model for specific spectral filters and for a small number of diverse atmospheric profiles. At each new satellite, the regressed coefficients have to be recomputed. The purpose of this study is to develop the tools for quickly calculating these coefficients: To avoid the problem of large computing time inherent in line-by-line calculations, we have generated a database of elementary infrared transmittances for a set of 32 profiles representative of the earth. The transmittances have been computed with the line by line FASCOD3 model for separately uniformly-mixed gases, water vapour and ozone for 40 pressure levels and 5 scan angles. The results are stored at a spectral resolution of 0.1 cm-1. This large computing time step has to be made only one time and for each new radiometer, the elementary transmissions are convolved with the new filters and then used to compute the set of filter adapted coefficients. This second step takes less than 2 hours on working station. In a first section, we will present the elaboration of the transmission database. Then, we will evaluate the quality (through RTE calculations) of the elementary transmissions and of the regressed coefficients by comparison to the direct line-by-line results, for the NOAA7 and NOAA11 filters. In a last part, we will compare the radiances computed with this RTTOV version and another fast radiative model ITPP, for a set of more than 4700 collocated satellite and radiosonde situations, for NOAA11 and NOAA12 filters.

The multi-frame blind deconvolution algorithm is considered for processing astronomical speckle images when only a few frames of data are collected. It has been noted that when the speckle data contains even moderate amounts of shot noise the algorithm often converges to the trivial point solution. In this paper we consider a 'penalized' blind deconvolution algorithm in which the penalty function is based on the Knox-Thompson algorithm.

Sliding wave propagation over a sloping rough surface considered. Well-known Kirchoff approximation has some drawbacks. We can't consider rereflection from surface by this method. It is impossible to investigate a boundary of applicability of this method. In the case of the low grazing angles we can use the Parabolic equation. Procedure that reduces some boundary value problem for Parabolic equation to Cauchy problem was suggested by A.I.Saichev and V.N.Koshelev. We find exact solution of last problem in the form of Path Integral. The calculation of this Integral exactly is very difficult problem. We can calculate the Path Integral on the base of short wave approximation applying stationary phase method. Now we have approximate solution, which gives us opportunity to take into account a second- order (or more) reflection and consider a boundary of applicability of this approximation. The amplitude coefficients from our approximation described the reflection from surface with curvature in comparison with the reflection from tangent plane in Kirchoff approximation. Numerical results are illustrate an example of our approach application for the case of periodic surface. Some effects of scattering are considered.

The clouds play a significant role in the atmosphere particularly in the determination of the radiant balance. Since their type, structure and height considerable influence the transmission, absorption and re-radiation of the IR radiation they are of great importance for the local meteorology and the global climatology. In small spatial scales (micro- and meso- ones) the clouds considerable change in space and time. The ECLIPS (Experimental Cloud Lidar Pilot Study) program in which our lidar group also takes part is well described. In general, the main purpose is a ground-based lidar observation of the clouds simultaneously with the NOAA 10 and NOAA 11 meteorological satellites overpassing the same areas to be performed. Essential lidar data are the recorded profiles from which the information about the clouds height, optical depth and vertical extinction can be derived. The experimental data required for the ECLIPS Phase II measurements were recorded during the period from 21 May to O9 July 1991 at 30 observations by 3 hours each i.e. 1 hour before and 1 hour after the NOAA 10 and NOAA 11 satellites overpasses. The lidar data are completed with certain meteorological information obtained by the conventional means. The presented experiment was conducted using triple- beam aerosol meteorological lidar developed at the Institute of Electronics of Bulgarian Academy of Sciences, which are well described previously.

In this paper we propose a tomographic algorithm for reconstruction of the energy profile of an optical beam from the data measured with linear bolometer. This algorithm is based on a discrete Fourier transform (DFT). When reconstructing the DFT values and making linear interpolation we used a rather in the form of concentric squares, which essentially decrease of the error of reconstruction and saved a computation time. In a computer experiment we have estimated the accuracy of the reconstruction method and determined the optimal number of projections. The proposed algorithm has been successfully used for reconstructing the beam intensity distribution from the experimental data. The result of reconstruction were compared with the data of direct measurements. It is shown that the proposed algorithm allows most fast reconstruction of spatial energy structure of an optical pulse to be done. The accuracy of reconstruction is not worse than that of the existing algorithms when the noise level is no more than 20%.

Using the semiclassical approach the simulation of the light pulse generation by a thin layer of quantum two-level dipoles has been performed. The state of each dipole is governed by Bloch equations for individual density matrix elements under condition of phase relaxation absence. Identical dipoles are located at the nodes of 2-D grating with corresponding sizes Lx, Ly and are considered as pointlike sources of classical field. At the initial moment all the dipoles are in an inverted state Initiation of the pulse was carried out by either incoherent spontaneous decay (superfluorescence mode simulated by setting of small random polarisation with *-correlation) or coherent initial state. Based on the numerical solution the kinetics of dipoles characteristics and evolution of the radiation directivity pattern have been calculated. The influence of the initial state and short-range Coulomb dipole-dipole interaction on the pulse characteristics have been analysed. Tending of axis X to be a direction of primary amplification of pulse with increasing of inequality Lx > Ly has been traced. The comparison of the results obtained with the results of paraxial approximation that uses the concept of slowly varying along axis X amplitudes of the electric field and medium polarisation has been done.

It has been recognized for some time that full order adaptive optics systems (one or more actuators per r0) can provide considerable improvement in the performance of astronomical imaging systems. However as four meter class telescopes become the standard for astronomical work it will become more difficult in terms of cost and complexity to build and operate full order AO systems, especially if one is interested in working in the visible. For the past several years we have been developing simula- tion and analysis tools to study AO systems using laser guide stars. This analysis indicates that even low order correction can provide improvements in image quality especially when used in conjunction with computer post processing algorithms. In this paper we expand on past work with computer simulation studies to include the effect of tilt and focus anisoplanitism.