AEROSPATIALE, leading a European team, has just conducted a successful study, under ESA contract, to demonstrate the feasibility of a spaceborne Doppler wind lidar instrument meeting the scientific requirements of wind velocity measurements from space with high spatial resolution. A first parametric investigation, based upon the initial set of mission requirements, and supported by dedicated models and detailed trade-off studies, took account of capabilities of most promising signal processing algorithms and calibration/validation constraints: it yielded a large conically scanned instrument deemed technologically risky. A risk analysis was then carried out to propose a less challenging instrument meeting most key mission requirements. The fixed line-of-sight concept with return signal accumulation appeared as most attractive. A second set of requirements agreed upon by scientific users was therefore issued, with relaxed constraints mainly on horizontal resolution, keeping roughly the same level of wind velocity measurement accuracy. A second instrument and subsystem trade-off was then performed to eventually produce an attractive instrument concept based upon a pair of small diameter telescopes each one associated to one scanning mirror rotating stepwise around the telescope axis, which drastically reduces the detection bandwidth. Following the main contract, studies of accommodation on the International Space Station have been performed, confirming the interest of such an instrument for wind measurements from space.

For the post 2000 time frame, the European Space Agency (ESA) has defined candidate missions for Earth observation. In the class of the Earth Explorer missions, dedicated to research and demonstration missions, the land-surface processes and interactions mission (LSPIM) involves a dedicated satellite carrying a single optical payload named PRISM (processes research by an imaging space mission). PRISM is a push broom multispectral imager providing high spatial resolution images (50 m over 50 km swath) in the whole optical spectral domain (from 450 nm to 2.3 micrometer with a resolution close to 10 nm, and three bands from 8 to 12.3 micrometer). It provides an access on any site on Earth within at maximum 3 days. In addition, the mission will be able to provide multi-directional observations by combining instrument depointing capabilities and satellite maneuvering. The instrument radiometric performance reaches a high level of accuracy by involving on-board calibration capabilities. This paper presents the results of one of the two pre-feasibility studies awarded by ESA, led by AEROSPATIALE and concerning the PRISM payload.

The SAGEM Group has been involved for more than 30 years in the field of remote sensing, especially via line-scanning sensors. Today, the SAGEM Group develops and manufactures optronic sensors with spectral bandwidths ranging from ultraviolet up to long-wave infrared (LWIR). Their name is CYCLOPE. Twenty five years ago, a four-channel infrared linescanner was delivered to the French Space Agency, CNES, for remote sensing evaluation and future specification of related spaceborne system. At the same time, a version was delivered to the French Administration for maritime oil pollution monitoring. This equipment is still in use and second-generation equipment was purchased in 1995 by the French Customs. The payload is described as well as the feasibility of such payload for spaceborne applications. Design-driving parameters and technologies are discussed. Emerging technologies make it possible now to propose such systems.

This paper introduces the 'complex instruments ranking with a new computational environment' or CIRCE software tool for aiding elaboration and exploitation of analytic models for performance management of optical instruments. CIRCE is currently developed by the optical instrument preliminary design team at the Aerospatiale Company's Cannes, France, Center, in cooperation with the Institut National de Recherche en Informatique et Automatique' or INRIA at the nearby Science Park of Sophia Antipolis. As a multiprogram tool, CIRCE incorporates the requirements at all stages of optical instrument development, from conception through manufacturing. It affords an original approach to creation and operation of performance models that facilitates know- how conservation through introduction of the notions of concepts (relations bases) and models (computation tree). It eases out the everyday tasks of engineers owing to generating capabilities for performance budgets or parametric analyses and to automatic numeric code generation.

The advanced spaceborne thermal emission and reflection radiometer (ASTER) is a research facility instrument to be launched on NASA's Earth Observing System AM1 (EOS AM-1) platform in 1998. ASTER has three spectral bands in the VNIR, six bands in the SWIR, and five bands in the TIR regions with 15, 30, and 90 m ground resolution respectively. The VNIR subsystem has one backward-viewing band for stereoscopic observation in the along-track direction. Because the data will have wide spectral coverage and relatively high spatial resolution, we will be able to discriminate a variety of surface materials and reduce problems resulting from mixed pixels. ASTER will provide the highest spatial resolution surface temperature and emissivity data of all the EOS AM-1 instruments. The primary science objective of the ASTER mission is to improve understanding of the local- and regional-scale processes occurring on or near the Earth's surface and lower atmosphere, including surface-atmosphere interactions. Specific areas of the science investigation can be listed as: (1) vegetation and ecosystem dynamics, (2) land surface climatology, (3) volcano monitoring, (4) aerosols and clouds, (5) carbon cycling and in the marine ecosystem, (6) hydrology, and (7) geology and soil. There are four categories of data; global data sets, regional data sets, and local data sets to be obtained by data acquisition requests (DAR) from scientists. Prioritization of data acquisition requests will be done using the factors such as observation category, user category, and science discipline.

The effect of aerosols on the upward radiance at the top of the atmosphere for VNIR (0.56, 0.66, and 0.81 micrometer) and SWIR (1.650, 2.165, 2.205, 2.260, 2.330, and 2.395 micrometer) of EOS/ASTER is investigated using the doubling/adding method. The realistic atmosphere-surface model includes five plane parallel sub-layers bounded by the Lambertian surface. The mid-latitude summer model of LOWTRAN 6 is adopted for molecular scattering and absorption. Three aerosol types are considered: 'dust like,' 'oceanic,' and 'water soluble' models of IRS standard aerosol with optical thickness of 0.0, 0.157, and 0.25 at 0.56 micrometers. The extinction coefficient, single scattering albedo, and phase function are computed by the Mie scattering theory. The surface albedo is 0.7 that is the practically maximum value except for ice/snow field. For the above surface albedo, the radiance at zenith direction generally shows its maximum value when the aerosol is not included, expect for the purely scattering aerosols at small solar zenith angles. Part of this study is utilized for the design of EOS/ASTER VNIR and SWIR sensors.

To derive the earth surface (ocean) parameters quantitatively, a contamination of the atmospheric constituents is to be eliminated. Especially VNIR on ASTER provides data with respect to surface characteristics with a fine spacial resolution of 15 m. In this case, the adjacency effect due to the inhomogeneous surface on satellite data should be included in the investigation. At present, right after the launch the version 3 of the atmospheric correction algorithm is scheduled to be inclusive of this effect. Based on the simulation of atmospheric effects on the emergent radiation over a checkerboard type of terrain, an operational procedure of the atmospheric correction is outlined. The present new version enables us to quantitatively discuss radiative transfer over the non- uniform surface. The look-up-table method is used for the derivation of parameters.

The ASTER scanner on NASA's EOS-AM1 satellite (launch: 1998) will collect five channels of TIR data with an NE(Delta) T of less than or equal to 0.3 degrees Kelvin to estimate surface kinetic temperatures and emissivity spectra, especially over land, where emissivities are not known in advance. Temperature/emissivity separation (TES) is difficult because there are five measurements but six unknowns. Various approaches have been used to constrain the extra degree of freedom. ASTER's TES algorithm hybridizes three established algorithms, first estimating the temperature by the normalized emissivity method, and then using it to calculate emissivity band ratios. An empirical relationship predicts the minimum emissivity from the spectral contrast (min-max difference: MMD) of the ratioed values, permitting recovery of the emissivity spectrum. TES uses an iterative approach to remove reflected sky irradiance. Based on numerical simulation, TES can recover temperatures within about plus or minus 1.5 degrees Kelvin, and emissivities within about plus or minus 0.015. Limitations arise from the empirical relationship between emissivity values and spectral contrast, compensation for reflected sky irradiance, and ASTER's precision, calibration and atmospheric correction. Improvements of TES before launch will focus on refining the MMD relationship.

In addition to acquiring multispectral data, the advanced spaceborne thermal emission and reflectance radiometer (ASTER) will also acquire along-track stereo data. ASTER is capable of acquiring 771 digital stereo pairs per day, each covering 60 by 60 km on the ground, at 15 m resolution with a base/height ratio of 0.6. According to present plans, approximately 30 digital elevation models (DEMs), with 7 - 50 m accuracy (RMSExyz) will be produced daily. During the 5 year mission on the EOS AM1 platform, ASTER has the potential to provide a coherent digital stereo dataset covering the Earth's land surface. At minimum, DEMs derived from these data will augment topographic data from other sources. These fundamental geophysical measurements will be a major contribution to interdisciplinary studies of the Earth as a planet.

Orthoimage products are planned to be produced at ASTER ground data system (GDS), this will be useful for general applications users. The major problem in orthoimage generation is a big computation load for terrain correction. In this paper, a high speed orthoimage generation method is proposed. The proposed method is based on the next four basic ideas: (1) The DEM products produced at ASTER GDS are used as fundamental information for terrain correction. ASTER DEM is generated by stereo matching on the basis of area correlation algorithm. Therefore the terrain correction for every pixel is ineffective. (2) As ASTER DEM products have ground coordinates corresponding to every other pixel on observation images, it is easy to transform observation image coordinates into orthoimage coordinates by using the ECR coordination. (3) In order to eliminate iterative calculations for determination of relations from orthoimage coordinate to observation coordinate, inverse transformations are directly obtained from relationship between a block on the observation image and its projection onto the orthoimage coordinate system. (4) Pixels in the block projected on orthoimage are transformed into the observation image coordinate. Then, image values are interpolated at the image observation coordinate. As a result, the proposed method can be executed with similar computation load to geometric transformation without terrain correction.

The recently developed atmospheric Wiener filter, which corrects for turbulence and aerosol blur and path radiance simultaneously, is implemented in digital restoration of AVHRR imagery over the five wavelength bands of the satellite instrumentation. Restoration is most impressive for higher optical depth situations, with improvement with regard to both smallness of size of resolvable detail and contrast Turbulence modulation transfer function (MTF) is calculated from meteorological data. Aerosol MTF is calculated from optical depth, measured with a sun-photometer. The product of the two yields atmospheric MiT which is implemented in the atmospheric Wiener filter. Image restorations with accompanying atmospheric MTF curves are presented. However, restoration results using a simple inverse MTF filter were quite similar. This indicates the satellite images were characterized by very low noise and that turbulence jitter was very limited which, in turn, indicates that the twbulencc MTFs integrated upwards over the path length were small compared to aerosol MTFs.

An algorithm is currently under development that will provide a classification mask for ASTER imagery obtained poleward of 60 N and 60 S. The classification mask will be a product available through EOSDIS and is called the ASTER polar cloud mask. Ten classes are currently in the mask and include six clear classes (water, slush/wet ice, ice/snow, land, shadow on land, and shadow on ice/snow) and four cloud classes (thin cloud over ice/snow, water, or land, and thick cloud). The algorithms is designed as a four stage process. In the first stage the data are median filtered, sampled to 30 m spatial resolution, normalized, and navigated to coastlines and ancillary Earth surface databases. In the second stage, through adaptive thresholding, simple decision surfaces, and ancillary data, the class ambiguity of each pixel is reduced from ten to two to four classes. In the third stage, additional features are utilized in a paired- histogram classification methodology to make the final pixel classification. And finally, in the fourth stage, a simple spatial consistency check is performed over the entire classification mask to detect isolated pixel classifications. Over 3700 samples have been extracted and labeled to date representing over one million pixels from 82 Landsat TM circumpolar scenes. Tests of the algorithm on the labeled samples indicate that the clear/cloud classification accuracy is greater than 90 percent and subjective evaluation of the classification masks supports that result.

This paper presents a review of how data from the advanced spaceborne thermal emission radiometer (ASTER) can be used to estimate the energy fluxes from the land surface. The basic concepts of the energy balance at the land surface are presented along with an example of how remotely sensed surface brightness temperatures can be used to estimate the sensible heat. The example is from the Monsoon 90 experiment conducted over an arid watershed in the state of Arizona in the United States.

We have developed two high performance 1024 multiplied by 1024 focal plane arrays for astronomy, spectroscopy, surveillance and conventional imaging. Each hybrid consists of a photovoltaic HgCdTe detector array, fabricated on Al₂O₃ substrate and having photoresponse cutoff wavelength optimized for each specific application, mated to a CMOS silicon readout via indium column interconnects. In addition to updating the performance of our 1024 multiplied by 1024 FPA for astronomy developed in 1994, we introduce a second 1024 multiplied by 1024 having capability for operation at TV-type frame rates. The latter device also has low read noise but at much higher bandwidth by virtue of its capacitive transimpedance amplifier input and pipelined readout architecture. Both devices have been shown capable of consistently achieving background-limited sensitivity at very low infrared backgrounds (less than or equal to 10⁹ photons/cm2-sec) by their low read noise, low dark current including negligible MOSFET self-emission, and high quantum efficiency. FPA pixel operability as high as 99.94% with mean peak D* of 10¹⁴ cm-Hz^1/2/W has been demonstrated. Proprietary hybridization and mounting techniques are being used to insure hybrid reliability after many thermal cycles. The hybrid methodology has been modeled using finite element modeling to understand the limiting mechanisms; very good agreement has been achieved with the measured reliability.

The requirements for ultrareliable miniature cryocoolers with affordable price is growing up for the military, space and commercial infrared applications. For scientific space application a MTTF in the range of 20,000 hours is targeted today by the potential users. To reach this goal, a technical breakthrough is necessary and a combination of an ultrareliable linear drive compressor based on flexure bearings and a pulse tube coldfinger is proposed. A technical description of the cryocooler is presented and a dynamic vibration control system is discussed.

Uncooled infrared sensors are significant in a number of scientific and technological applications. A new approach to uncooled infrared detectors has been developed using piezoresistive microcantilevers coated with thermal energy absorbing material(s). Infrared radiation absorbed by the microcantilever detector can be sensitively detected as changes in the electrical resistance as a function of microcantilever bending. These devices have demonstrated sensitivities comparable to existing uncooled thermal detector technologies. The dynamic range of these devices is extremely large due to measurable resistance change obtained with only nanometer level cantilever displacement. Optimization of geometrical properties for selected commercially available cantilevers is presented. Additionally, we present results obtained from a modeling analysis of the thermal properties of several different microcantilever detector architectures.

The wide-field infrared explorer (WIRE) is a cryogenically cooled spaceborne telescope designed to study the evolution of starburst galaxies. Scheduled for a September 1998 launch as NASA's fifth small explorer mission, WIRE will employ a 30 cm aperture Ritchey-Chretien telescope to image a 33 by 33 arcminute field simultaneously onto two Si:As BIB detector arrays covering broad bands centered at 12 and 25 microns. A three-part survey strategy calls for moderate- depth (about 15 minutes total integration time), deep (3-6 hours), and ultra-deep (24 hours) fields. For the deep fields, hundreds of background-limited exposures will be recorded by the WIRE instrument over many orbits, and rectified, registered, and combined on the ground. The sensitivity of these final images will be limited by source confusion and is expected to be less than 0.4 mJy (5-sigma) at 25 microns. The WIRE image simulator is being developed to simulate the exposures sent down from the spacecraft as closely as possible, including the effects of diffraction, background noise, source confusion, stray light, detector array characteristics, spacecraft jitter and roll, and others. We describe the design and implementation of the simulator, with particular emphasis on the generation of point-spread functions. The simulator is written in C for use on Unix workstations, and we assess its performance. Sample raw and combined images are displayed, and the image processing steps are outlined. The uses of the simulator to verify that mission requirements are met, to optimize observing strategy, and to test data analysis techniques are also described.

Japanese satellite-borne infrared telescope, the Infrared Telescope in Space (IRTS), was successfully launched on March 18, 1995 UT. The IRTS consisted of a 15 cm telescope cooled with superfluid liquid helium and installed onboard the space flyer unit (SFU) spacecraft. The IRTS mission started on March 29, UT and terminated on April 26 UT after liquid helium ran out. The cryogenic system operated as designed and held the telescope and the focal-plane instruments at a stable temperature of 1.9 K for 38 days. Four focal-plane instruments which together covered almost the entire infrared wavelength range observed a sky area of about 2700 deg² and returned a wealth of new data on a variety of objects, including the zodiacal light, interstellar gas and dust, near-infrared background light and point sources.

The near-infrared spectrometer (NIRS) is one of the focal plane instruments of the infrared telescope in space (IRTS). The NIRS is a simple grating spectrometer with two element InSb linear arrays, and was designed to measure the absolute sky brightness at the wavelengths from 1.4 to 4.0 micrometer with a spectral resolution of 0.13 micrometer and a beam size of 8 feet by 8 feet. The IRTS was launched on 1995 March 18. The NIRS worked well throughout the observation period from March 29 to April 25, and scanned about 7% of the entire sky. Multiple passage of bright stars through the NIRS field of view enabled us to reconstruct the beam pattern and to calibrate the sensitivity. Those flight data confirmed good performance of the NIRS on the orbit as was expected from the preflight measurements.

The mid-infrared spectrometer (MIRS) was one of four focal- plane science instruments that flew aboard the orbiting infrared telescope in space (IRTS). This telescope was a joint NASA/Japanese Scientific Space Agency (ISAS) project that was launched on March 18, 1995 aboard a Japanese HII expendable launch vehicle and was subsequently retrieved by the space shuttle. The telescope itself was liquid helium- cooled with a 15 cm aperture and surveyed approximately 7% of the sky over the course of its 26 day mission life before its cryogen expired and it began to warm up. The MIRS was developed jointly by NASA, the University of Tokyo, and ISAS and operated over a wavelength range of 4.5 to 11.7 microns with a spectral resolution of 0.23 to 0.36 microns. The MIRS has a conventional entrance aperture, so that spectral studies could be made of extended as well as point-sources. A cold shutter and an internal calibrator allowed accurate absolute flux determinations. The realized in-flight performance of the MIRS followed the pre-launch calibration performance as measured on the ground, with the exception of some degradation in the spectrometer throughput, some unanticipated detector behavior due to the passages through the South Atlantic anomaly, and to unavoidable observations of the moon.

The far-infrared line mapper (FILM) is a far-infrared spectrometer and in one of four focal plane instruments of the infrared telescope in space (IRTS), FILM was designed for wide area intensity mapping of far-infrared emission from interstellar gas and dust in the galaxy. The targets are the [CII] 158 micrometer line of the ionized carbon, the [OI] 63 micrometer line of the oxygen atom, and the continuum emission at 155 and 160 micrometer from the interstellar dust grain. A cylindrically concave varied line-space grating and a linear array of stressed Ge:Ga were successfully developed and allowed us to make a compact spectrometer compatible to severe limitations of the small cryogenic telescope. The IRTS, onboard the space flyer unit (SFU), was launched by a HII rocket on March 18, 1995 and was recovered by a STS on January 13, 1996. The FILM worked very well during four weeks allocated for the IRTS observation and produced a lot of valuable data. The sensitivity and the spatial resolution for the [CII] line are an order of magnitude better than the previous work.

We present the flight performance of the far infrared photometer (FIRP) onboard the infrared telescope in space (IRTS). The FIRP was designed to measure the absolute sky brightness in four submillimeter wavelength bands centered on 150, 250, 400, and 700 micrometer with a spectral resolution of lambda/(Delta) (lambda) equals 3, and with spatial resolution of 0.5 degrees. The bolometers were cooled to 300 mK by a ³He refrigerator, and an ac bridge readout circuit was used to achieve high sensitivity. The ³He refrigerator achieved a temperature of approximately 300 mK during each of three ³He condensations carried out in orbit. The hold time was confirmed to be at least 7 days per cycle. We observed approximately 7% of the sky during a mission time of 3 weeks. Some excess noise was observed in most of the channels. Adequate sensitivity was achieved in all channels for observations at low galactic latitudes, where emission from interstellar dust is relatively bright. The preliminary sensitivities (60 sec integration, 1 (sigma) ) are estimated to be 7.7 multiplied by 10^-7, 3.1 multiplied by 10^-8, 7.7 multiplied by 10^-8, and 3.7 multiplied by 10^-8 Wm^-2sr^-1 for the 150, 250, 400, and 700 micrometer channels, respectively.

In the given paper there is presented the theoretical analysis of the affects induced by a high-energy irradiation in low-background IR detectors during their operation. This problem is extremely important for the space measurements of the very low IR radiation of the cold Universe. In the paper special attention is given to the analysis of slow variations of the detector characteristics under high-energy particles, to the determination of the main mechanisms causing these variations and defining their magnitudes, to the classification of the typical phenomena being attendant to these variations. At the end of the analysis there are given the conclusions dealing with the use of the presented results for improving space measurements.

A six-feature, all-sky star field identification algorithm has been developed, integrated with a CCD-based imaging camera and tested at the Table Mountain Observatory. The calibration of the CCD-based camera to minimize the error in the measured star parameters due to instrument error and experimental conditions is described. A set of observatory tests on star fields with this intelligent camera are presented.

The on-axis image plane incidence of an extended object (sometime also called irradiance), radiating as a Lambertian radiator, is derived for an optical system with a central obscuration. It is then extended to off-axis image points to obtain a generalized form of image incidence for an extended source. For an object at infinity, under the assumptions of paraxial image formation, the newly derived expression reduces to that used traditionally in the radiometric system design.

With half of the world's population living within 50 km of the coastal ocean the coast and adjacent land areas are heavily used for recreation, and for frequently conflicting uses, such as, fisheries, oil and gas production, disposal of wastes, transportation and naval operations. Coastal ecosystems are sensitive, highly productive systems which are being heavily impacted by human activities, but which are not adequately sampled by any present or planned spaceborne remote sensing system. To remedy that situation we propose building a coastal ocean imaging spectrometer (COIS) with adequate spectral and spatial resolution and high signal to noise to provide long term monitoring and real-time characterization of the coastal environment. COIS would provide a snapshot of the effects of human activities and natural processes, including runoff, tides, currents and storms, on the distributions of phytoplankton, suspended sediments, colored dissolved organic matter, including sediment resuspension and changes in bathymetry. COIS will also be an excellent tool to assess changing land use practices and the health of corps and natural vegetation on the adjacent land areas. This paper reviews the scientific rationale for such an instrument, and the recent scientific and engineering innovations that make it possible to build a small inexpensive spaceborne instrument to meet these requirements.

AEROSPATIALE, prime contractor, presents the main results related to the activities performed in order to demonstrate the feasibility of a coherent 2 micrometer lidar instrument capable of measuring water vapor and wind velocity in the planetary boundary layer, and to determine the main subsystem critical items: (1) selected instrument configuration and associated performances, (2) 2 micrometer laser configuration with phase conjugation, (3) coherent receiver chain architecture, (4) frequency locking and offsetting architecture. The second phase of this study will be dedicated to breadboard the most critical elements of the instrument at 2 micrometers in order to technologically consolidate the feasibility of such an instrument.

AIRS is a key instrument in NASA's Earth Observing System (EOS) Program. Passive IR remote sensing is performed using a high resolution grating spectrometer design with a wide spectral coverage focal plane assembly (FPA). The hybrid HgCdTe focal plane consists of twelve modules, ten photovoltaic (PV) and two photoconductive (PC), providing spectral response from 3.7 to 15.4 micrometers. The PV modules use silicon readout integrated circuits (ROICs) joined to the detector arrays as either direct or indirect hybrids. The PC modules are optically chopped and led out to warm electronics. Operating at 58 K, the sensitivity requirements approach BLIP in the critical 4.2 and 15.0 micrometer bands. The optical footprint coupled with the support and interface components of the focal plane make it a very large assembly, 53 mm multiplied by 66 mm. Dispersed energy from the grating is presented to the modules through 17 narrowband filters mounted 0.2 mm above the focal plane in a single, removable precision assembly. With PV and PC devices on the same focal plane operating simultaneously, shielding and lead routing as well as ROIC design have been optimized to minimize any interactions between them. Multilayer carriers have been designed to lead out the closely spaced PC arrays and the entire focal plane itself. Multilayer shielded flex cables are used to interconnect the focal plane to a very unique dewar. The tightly spaced optical pattern, along with more than 50 components in the focal plane, make this a highly complex assembly. The vacuum dewar, while providing approximately 600 leadouts, is directly coupled to the cold spectrometer and operates at 155 K while cooling the focal plane to 58 K via a sapphire rod interfaced to a pulse tube cooler. This paper discusses the key features of the FPA/dewar assembly, modeling/analyses done in support of the design, and results of design validation activities to date.

A rotationally shearing interferometer may be used for a detection of wavefront asymmetry, even when it is very small. We derive the analytical expressions demonstrating that the asymmetrical part of the wavefront aberration function may indeed be detected using a rotationally shearing interferometer. The aberration coefficients up to the fourth order in the radial coordinate may be found using the rotationally shearing interferometer. The method of rotationally shearing interferometry is shown to be particularly suitable for the characterization of the asymmetrical optical systems.

Ray-tracing has found important applications in the analysis of optical systems. In this work, a special modification of the ray-tracing method is applied to the analysis of several different micro-optical structures, used as probes for refractive index measurements, with the aim of obtaining the structure that is practical and acceptable from the technological point of view. The structure which is studied here is a small size solid retroreflector of special shape and complex inner constitution. It serves to couple together a pair of optical fibers. One of these fibers is connected to the light source, the other one to the photometer. The operation of this structure as the refractometric sensor is based on the frustrated total internal reflection in the retroreflector, which is formed by the structure's surface. Therefore, the structure's internal reflectivity depends on the refractive index of the surrounding media. This media may be either a gas or a fluid. By use of this structure, the refractive index of the surrounding media can be determined, and the type of the particular fluid may be disclosed. However, for better performance, the sensor's shape has to be optimized taking into consideration the particular parameters of the optical elements and materials used for the sensor, and the refractive index value of the particular fluid. Because of some singularities of the problem, the special ray-tracing algorithm has been developed for the analysis of this structure. This algorithm has proved to be effective for the theoretical optimization of the structure used as refractometric optical sensor.