Improving a Satellite Mission System by means of a
Semantic Grid Architecture
Manuel Sánchez-Gestido1, María S. Pérez-Hernández2, Rafael González-Cabero3,
Asunción Gómez-Pérez3

1Deimos Space S.L.,
Ronda de Poniente 19, Edificio Fiteni VI, P2, 2º
28769 Tres Cantos, Madrid. Spain.
manuel.sanchez@deimos-space.com

2Departamento de Arquitectura y Tecnología de Sistemas Informáticos, Facultad de
Informática.Campus de Montegancedo s/n, Universidad Politécnica de Madrid.
28660 Boadilla del Monte, Madrid. Spain.
mperez@fi.upm.es

3Departamento de Inteligencia Artificial, Facultad de Informática.
Campus de Montegancedo s/n, Universidad Politécnica de Madrid.
28660 Boadilla del Monte, Madrid. Spain.
{asun,rgonza}@fi.upm.es
Abstract. The use of a semantic grid architecture can make easier the
deployment of complex applications, in which several organizations are
involved and diverse resources are shared. This paper presents the application
of the architecture defined in the Ontogrid project (S-OGSA) into a scenario for
the analysis of the quality of the products of satellite missions.
1. Introduction
In the last years, complex applications have arisen in multiple domains. Many of
these applications have been solved by means of traditional techniques. Although
these solutions are feasible, they are also error prone and difficult to implement.
In these scenarios, where a huge number of resources are shared and even several
organizations are involved, the use of grid computing can help to deploy flexible
solutions [5]. If there are complex interactions between all the components and
participants of the applications, it is possible to take advantage of the use of a
common vocabulary and conventions, which provide semantics and interoperability.
The “Quality Analysis of Satellite Missions (QUARC)” use case of the Ontogrid
project1 is one example of this kind of scenarios, since the current operational system
for the selected Satellite (EnviSat) has been developed through the use of traditional
techniques. This use case is intended to improve the possibilities in the analysis of the

1
www.ontogrid.net

2 Manuel Sánchez-Gestido1, María S. Pérez-Hernández2, Rafael González-Cabero3,
Asunción Gómez-Pérez3
Formatted: Spanish
(Spain-Modern Sort)
Formatted: Spanish
(Spain-Modern Sort)
Field Code Changed
Field Code Changed
quality of the products for satellite missions, whose main goal is obtaining measures
of the Earth observation. On the other hand, Ontogrid has as main goal the explicit
sharing and deployment of knowledge to be used for the development of innovative
Grid infrastructures and for Grid applications, that is, the Semantic Grid.
We consider that the use of a semantic grid architecture [2] for solving this use
case can provide a large number of benefits.
This paper describes this use case and its adaptation to the semantic grid
architecture developed in the Ontogrid project, S-OGSA [3].
2. QUARC Use Case Definition
Earth Observation can be defined as the science of getting data from our planet by
placing in orbit a HW/SW element with several observation instruments and
processing the scientific data obtained in order to get meaningful information (i.e.
images).
In a nutshell and putting aside other aspects, basic working of an Earth Observation
Satellite System consists on a simple process that is repeated over time. The
instruments on board the satellite act like cameras that can be programmed (very
complex cameras nevertheless), taking "pictures" (images) of specific parts of the
Earth at predefined times. Parameters for taking this pictures (like any camera would
need to operate) and also parameters for the satellite general configuration, constitute
the information included in the Planning issued by the Mission Planning System
(MPS), sent to the FOS (Flight Operation Segment), which also sends an equivalent
information to a Ground Station (GSt) located, for the Envisat satellite, in Kiruna
(Sweden) and from there to the satellite.
FOS converts the information in the Planning into the shape of MCMD´s. This
means a translation from one format to another that is meaningful to the satellite. The
Ground Station translates the MCMD information into a RadioFrequency link to
communicate with the satellite antenna in the Service Module of the satellite.
MCMD´s in a radiofrequency link are generically called TeleCommands, TC.
Computer on board the satellite will store the list of MCMD´s, each of them with
a time tag that marks the execution time of that MCMD. Taking one "picture" would
mean, for instance, the execution of a MCMD that copies those parameters for taking
the picture, from the memory of the satellite computer to the instrument computer
memory, and then the execution of the MCMD that triggers the camera shot.
Each MCMD is a command (constituted by different pieces of information) that
asks an instrument or any other part of the satellite to perform an action (load a table
or trigger an operation).
Some MCMD´s are also included by the FOS (in the whole bunch of MCMD´s
sent to the Ground Station) in order to get information from the Satellite internal
status and configuration at a particular moment. When this information is sent by the
satellite to the Ground Station is called Telemetry (TM).
"Pictures" from each of the instruments are stored onboard (in the satellite
computer memory) as a kind of raw data and when the satellite over-flies the Ground
station that information is sent to the Ground Station antenna (Data downlink). In that

Improving a Satellite Mission System by means of a Semantic Grid Architecture 3
Formatted: English (U.K.)
Field Code Changed
Ground Station a preliminary conversion from raw data to a so called "Level 0"
product is performed (basically adding a identification label to each of the pictures).
These "Level 0" products are sent to the PDS (Payload Data Segment) that produce
the Level 1b and Level 2 products that are made available to the final user community
(scientist, environmental organizations, etc).
QUARC is a system that checks off-line the quality of the product instrument
files. This process needs as input the product files, the MCMD and the mission
planning, which other facilities provided to the system. QUARC returns reports and
plots, which allows the operator to produce a new planning. Therefore, the QUARC
system is designed to help to take decisions when an instrument or the whole system
begins to malfunction and detect that something incorrect has occurred in one part of
the data generation and circulation system. A more detailed explanation of the whole
system is described in [4]. Figure 1 picture shows graphically the overall scenario:


Figure 1 A general overview of the use case.

4 Manuel Sánchez-Gestido1, María S. Pérez-Hernández2, Rafael González-Cabero3,
Asunción Gómez-Pérez3
Formatted: Spanish
(Spain-Modern Sort)
Formatted: Spanish
(Spain-Modern Sort)
Field Code Changed
Field Code Changed
3. A Grid Infrastructure for the QUARC process
The QUARC system involves a complex process in which distributed data
belonging to different organizations must be queried, processed and transferred.
These operations must be made by using suitable access control mechanisms.

Currently, the location of the resources and the processing of data are made in a
wired way, according to filenames and content of these files in an “ad-hoc” format.
The use of a grid framework will provide a flexible way of locating required
resources and the virtualization of these resources by means of (Semantic) Grid
Services.

Besides, there is a need for managing lifetime of certain data resources linked to
the used databases in the system, since there are database changes along the lifetime
of the system.

A notification scheme can help in the Satellite Mission system due to the
interactions between the processes of different organizations.

Regarding security, the involvement of several organizations implies the
establishment of different access policies and the definition of virtual organizations.
The role of each specific actor within an organization also defines its privileges as a
member of the virtual organization.
4. Semantics in the QUARC process
Many resources used in this use case can benefit from metadata.

Annotation is needed to link file resources and their actual content meaning.
Provenance information and temporal information (time stamp) of the processes
should also be annotated (it can be done even automatically). The formal relationship
between parameters of the satellite instruments should also be done (in this case in a
manual fashion). These relationships should be established by means of axioms;
therefore heavy weight ontologies are required and the WebODE Knowledge
Representation ontology [1] will be used for that.

Figure 2 describes an example of the use of the components of the semantic grid
architecture the files annotation for the first scenario:

Improving a Satellite Mission System by means of a Semantic Grid Architecture 5
Formatted: English (U.K.)
Field Code Changed

Figure 2 Components of the Semantic Grid Architecture involved in the files
annotation scenario.
5. S-OGSA in the QUARC process
S-OGSA is the reference semantic grid architecture developed in Ontogrid. The
most important elements of S-OGSA applied in the QUARC process are:
• Annotation Services, more precisely ODESGS and Grid-KP, in order to obtain
the information through metadata associated to resources in the system.
• Workflow engine, in order to add new functionalities in the system without
software updates.
Acknowledgements
This work has been partially financed by the Ontogrid Project (FP6-511513) and
by a grant provided by the Comunidad Autónoma de Madrid (Autonomous
Community of Madrid).
References
1. Arpírez J.C., Corcho O., Fernández-López M., and Gómez-Pérez A 2003.: WebODE in a
nutshell. AI Magazine.
2. D. De Roure, N. R. Jennings, and N. R. Shadbolt, "Research Agenda for the Semantic
Grid: A Future e-Science Infrastructure," National e-Science Centre, Edinburgh, UK
UKeS-2002-02, December 2001.
3. Ioannis Kotsiopoulos et al. Specification of a Semantic Grid Architecture. September
2005. Ontogrid Deliverable D2.1. http://www.ontogrid.net/
4. Mediavilla Garay AI, Sánchez Gestido M. Business cases and user requirements analysis
and test set definition for quality analysis platform. February 2005. OntoGrid Deliverable
D8.1. http://www.ontogrid.net/
5. I. Foster and C. Kesselman, “The Grid: Blueprint for a New Computing Infrastructure”.
Morgan Kaufmann, 1999.