I.P.Torshina, Y.G.Yakushenkov
Moscow
State University of Geodesy and Cartography (MIIGAiK)
Gorokhovsky
by-str., 4, Moscow 105064, Russia
General Methodology of Modern
Electro-Optical Devices’ Computer Modeling
ABSTRACT
General methodology of modern
electro-optical devices’ computer modeling is described. Some peculiarities of
computer modeling for 3th generation of electro-optical devices are marked.
Analytic method of
evaluating adequacy and
robustness is offered.
Key words:
electro-optical device, computer modeling, model adequacy
1.INTRODUCTION
Electro-optical systems are the devices which
use optical radiation (optical signals) as a carrier of information on the
object being investigated or observed and convert the energy of radiation into
electric energy for preliminary data processing. The electro-optical devices
(EOD) engage with wide variety of applications, as example, ground, air, and
space based systems. Setting of the range performance for the complex
target-background situation is a difficult and expensive task enough. Many
traditional methods of designing don’t take account of complexity and changeability
of the target-background situation, the particular feature of the concrete
system, and other factors. Therefore it
has been proposed to utilize a computer modeling for assessment.
Role of computer modeling for projecting,
testing, and investigation modern EOD is well-known. Computer modeling is used
both for parametrical analysis and for the new EOD’ synthesis. It permits
modeling the variety structures of the systems with necessary adequacy,
calculating their figures of merit for different target-background situations,
evaluating effectiveness of modern element base, deciding many another tasks,
what permits to shorten time and cut down elaboration cost. Modeling permits
shortening volume of expensive and convincing experiment essentially, and
refusing against it often.
Now
the many electro-optical systems’ computer models are known. But the most of
them are used for synthesis and analysis of the special EODs or particular
cases their working conditions. Many models are counted on information users
only but not designers. Therefore the general methodology of computer modeling
for such complex system as EOD is represented very useful.
A certain version of generalized EOD computer model (EOD CM) has been
worked out by the Electro-Optical Design Department, the Moscow State
University of Geodesy and Cartography (MIIGAiK). Our general methodology
of EOD CM construction has been stated
in [1,2,3]. The peculiarities of multispectral EOD modeling based on general
EOD CM has been marked in [4,5].
The functional program ensuring of EOD CM may be divided into two parts:
1. composition of EOD CM, and 2. appraisal of its adequacy.
2. SOME GENERAL PRINCIPLES OF EOD CM
COMPOSITION
The structure EOD CM is the aggregate of its main blocks (submodels).
These submodels are: “Starting Data”, “Figures of Merit”, “Scenario”, “EOD
Structure”, “Result of Computer Modeling”, “Data Base of Computer Model”
(Fig.1). Due to modular structure of the
model it is possible to simulate a wide range of different EOD and interactions
between the EOD, the observed scene, its environment, and the date processing
algorithms.
|
Interface
of computer modeling program
|
Fig.1.
Structure of EOD general model
Software has been given as descriptions of the phenomenology submodels
and data base.
The model executive allows user to realize
various conceptions of EOD operating and to modify these conceptions working
out in its details, or simplifying the model in accordance with the available
data bases and the requirements to model adequacy.
The
submodel “Starting Data” contains a list of information what computer program’s user must know for EOD modeling. It is formed on the
basis of technical assignment for EOD designing. This list may be defined
finally after forming others submodels, in particular “Scenario” and “EOD
Structure” submodels.
The submodel “Starting Data” has a direct
connection with all submodels and feedback connection with submodels “Result of
Computer Modeling” and “Date Base of Computer Model”. A feedback with submodel
“Date Base of Computer Model” permits to select starting data to be demanded
from corresponding block of data base. A feedback with submodel “Result of
Computer Modeling” helps to optimize the EOD structure if it needs to correct
starting data.
A basis of “Figures of Merit” submodel is the
formulas of demanded indexes (figures of merit) of EOD working efficiency which are presented in general parametric
form at the first designing stage. During modeling these indexes are
calculated. If they don’t be satisfied the optimization of EOD structure and correction of its
elements’ parameters may be used.
The submodel “Scenario” takes
account of reciprocal disposition of objects, backgrounds and interferences to be observed in EOD field-of-view as well as outer conditions of
EOD operating (environmental
characteristics). The submodel has been developed to promote setting of
radiant flux or irradiance values from objects and other radiators at the EOD
entrance pupil in working spectral band. The submodel expresses the radiances
of objects, backgrounds and interferences as functions of their temperatures,
emissivities, reflectivities as well as the coordinates and shape factors of
the radiators. This submodel makes it possible for the concrete EOD and its
scenario of operating to formulate a basic radiometric equation which enables
the user to define the most important geometrical parameters of the optical
system, such as the entrance pupil area, focal number, field-of-view, and to
determine the noise equivalent power or detectivity of a detector [2].
Individual subsystems of the “EOD Structure” submodel (“Optical System”,
“Detector”, “Electronic Block”, “Display”, etc.) may be represented by their
modulation transfer functions (MTF). The functions as well as specific
parameters of the EOD elements may be packed into special blocks of submodel
“Data Base of Computer Model”. The MTF of the entire EOD is composed of a
number of partial MTF. These functions may be used to determine a noise
bandwidth and others figures of merit of EOD, for example, “Signal-Noise Ratio
at EOD Exit”.
The submodel “Data Base of Computer Model”
concludes a set of different blocks: “Radiators”, “Atmosphere”, “EOD Elements
and Algorithms Base Data” “EOD Figures of Merit”, and many others. These blocks
(data bases) may be realized as analytic forms, for example, transfer functions
of individual EOD blocks, as numeric, graphs, tables, etc. Parameters and
characteristics of many possible radiators and the earth atmosphere have been
packed into separate blocks. This block
architecture has been successfully implemented using object-oriented
engineering software techniques.
3. SOME
PECULIARITIES OF COMPUTER MODELING FOR 3th EOD GENERATION
Usually EOD 3th generation are named the devices working in two- or
multi-spectral ranges in which matrix focal plane arrays detectors are used.
These devices may be both two- or many-channels when optics of each channel are
divided and one-channel when there is one optical system but spectral ranges
are divided in two- or multi-spectral detectors [5,6]. A general structure
scheme of multi-stage EOD 3th generation (EOD-3) is showed at Fig.2.
Fig.2.
Structure of multispectral EOD
Computer model of EOD-3 may be formed on the basis of the general EOD CM
by means of addition special module “Correction of General EOD CM”
supplementing of “Data Base of Computer Model” by parameters and
characteristics of optical materials with wide spectral transmission, optical
elements realizing spectral decomposition (prisms, diffraction gratings,
filters, etc.), two- and multispectral detectors, systems of images’
combination.
For EOD-3 modeling it needs a more detailed data base including spectral
characteristics of optical radiation, transmission, and detection. The question
about simplification of spectral characteristics and presentation of the model
adequacy is formed at the first stages of modeling already.
Calculation algorithms of definition the reflection, the transmission,
and the absorption for various surfaces of targets, backgrounds, and obstacles
must take into consideration changing these physical quantities for different
wavelengths and orientations of the surfaces.
User can use such specific EOD-3 figures of merit as the spectral
contrast ratio, the difference of individual spectral signals, the spectral
ratio, the logarithmic spectral ratio.
4. ADEQUACY
AND ROBUSTNESS OF EOD CM
Adequacy is one of basic requirements for
EOD CM. It is characteristic of authenticity of any EOD properties
representation and conditions of its work. It is advisable to have adequacy
assessment as an errors function or criterion of adequacy (CA) which may be
quantitative measure of difference between the model and object-original
characteristics, for example, the EOD figures of merit [5].
At the first stages of EOD designing it is advisable CA evaluation
through the rejection of EOD figures of merit from the meaning to
have been assigned. It may be convenient to evaluate an adequacy with the help
of numeral methods. So the meaning is depended on EOD parameters and
characteristics changes it is possible to use the method of exact differential
for CA calculating. Separate parts of the perfect differential (the partial
derivatives) are determined by
parameters of individual EOD blocks. Forming the histograms of figure of merit differences as a functions of the input EOD parameters it may be possible to find the partial
derivatives of a figure of merit function
and calculate the perfect differential as a base for finding of the error
function and CA.
Testing of adequacy and robustness of EOD
CM may be both analytic and experimental. Analytic method is used at the first
stages of EOD designing i.e. the object-original is absent. At next stages it
is expedient to conduct the experimental testing of adequacy as physical
experiment.
The general structure of EOD CM, forming as the aggregate of some blocks (submodels) permits its using for
different practical cases, accounting
EOD structure’s and exploitation conditions’ changing. Computer model of EOD-3
may be formed on the basis of the general EOD CM by means of addition special
module “Correction of General EOD CM” supplementing of “Data Base of Computer
Model”. At the first stages of EOD
designing it may be possible to find the partial derivatives of a figure of merit function and calculate
the perfect differential as a criterion of
the model adequacy.
BIBLIOGRAPHY
1. I.P.Torshina
. “Computer Modeling of Electro-Optical Preliminary Data Processing Systems”. -
Moscow, Logos Publisher, 2009.– 248 p. (rus.),
2. Y.G.Yakushenkov. Electro-Optical
Devices: Theory and Design. - Moscow, Logos Publisher, 2011.– 568 p. (rus.),
3. I.P.Torshina and
Y.G.Yakushenkov. “Structure of General
Electro-Optical Systems’ Computer Model”.- Science-Technical Herald of Sankt-Petersburg State University
IFMO, 2009, ¹ 6(64). P.5-9 (rus),
4. V.V.Tarasov, I.P.Torshina and
Y.G.Yakushenkov. “Infrared
Systems of 3-th Generation”.- Moscow, Logos Publisher, 2011.– 240 p. (rus.),
5. I.P.Torshina and
Y.G.Yakushenkov. Peculiarities of
Computer Modeling for Electro-Optical Systems of 3th generation. – Optical
Journal (Journal of Soviet Optical Technology), 2009, ¹ 2, P. 87-89 (rus)
6. V.V.Tarasov and Y.G.Yakushenkov. Two-
and Multispectral Electro-Optical Devices with Focal Plane Arrays. - Moscow,
Logos Publisher, 2007.- 192 p. (rus.),
I.P.Torshina – Dr.Sc., Dean of Optical-Information
Systems and Technologies Faculty, the Moscow State University of Geodesy and
Cartography (MIIGAiK)
Y.G.Yakushenkov – Prof., Dr. Sc., Chief of
Electro-Optical Department, Optical-Information Systems and Technologies
Faculty, the Moscow State University of Geodesy and Cartography (MIIGAiK)