Minimize Deimos-2

Deimos-2 Minisatellite Mission

Deimos-2 is a follow-on imaging mission of Deimos-1 Imaging S. L. U., an Elecnor company, Boecillo, Spain. The Deimos-2 mission is aimed at operating an agile minisatellite for high-resolution EO (Earth Observation) applications. The agile spacecraft can be steered to accurately point the pushbroom-type optical payload, which can provide 1 m panchromatic and 4 m multispectral images in a swath of 12 km at nadir, at an orbit altitude of ~600 km. The multispectral capability includes 4 channels in the visible and near-infrared spectral range (red, green, blue and NIR). 1) 2) 3) 4) 5)

The off-nadir tilting capability of the S/C is intended to improve the revisit time performance and enhance the operational flexibility by significantly reducing the time interval to take images on areas of interest. While the across-track tilt angle for nominal image acquisitions is ±30º, the satellite can be configured to achieve ±45º off-nadir pointing capability, i.e., an extended FOR (Field of Regard) for data collection with short revisit times in emergency situations. Moreover, high-quality observations can be performed close to nadir (small FOR) to enable background mapping.


Figure 1: Illustration of the Deimos-2 spacecraft (image credit: SI, Deimos Imaging)


A contract between Deimos Imaging Inc. of Valladolid, Spain and SI (Satrec Initiative) of Daejeon, Korea was signed in November 2010. The satellite and its payload are being designed and developed by SI. According to agreements, the spacecraft will be integrated and tested in Spain in collaboration with SI, in the new Elecnor Deimos Satellite Systems premises at Boecillo near Madrid. Deimos Satellite Systems is responsible for system engineering, procurement, testing, integration, launcher procurement, commissioning, in-flight acceptance, and operations of the system.

The project PDR (Preliminary Design Review) was in May 2011. The CDR (Critical Design Review) was completed in June of 2012.

The spacecraft bus is designed to fulfil all mission requirements by incorporating 3-axis stabilized, accurate and agile attitude control for precise imaging operations. The satellite attitude is based on the following main nominal operations modes (Ref. 1):

• Housekeeping sun-pointing mode in sunlight: During the sunlight stage, the S/C attitude shall be maintained or steered so that the solar panels are always pointing towards the sun for maximum power generation. When the S/C is in sunlight, the camera points in the direction opposite to the sun, unless it is in observation mode.

• Housekeeping nadir-pointing mode in eclipse: The S/C shall point towards the Earth during eclipse to minimize the usage of payload heaters. When the S/C is in eclipse, the camera points in the nadir direction.

• EO modes: The platform agility is exploited to take images not only close to nadir, but also with up to ±45º into any direction (cross-track or ±30º along-track). This allows single-strip imaging of areas of interest, multi-pointing imaging of close-enough targets, single-pass stereo imaging by along-track pitch angle maneuvers and tessellation imaging.
Note: In geometrical terminology a tessellation is the pattern resulting from the arrangement of regular polygons to cover a plane without any interstices (gaps) or overlapping. The patterns are usually repeating.

• Data download operation mode: When the link with a GS (Ground Station) can be set up, the S/C shall maneuver to point to the GS with maximum solar power generation. At the same time, the X-band antenna gimbal shall be steered towards the GS antenna to provide the needed gain. Consecutive image download operations are foreseen if the single-GS pass duration is not sufficient for data download and the following GS visibility interval is very close to the previous one.

• Orbit maintenance and control operation mode: In-plane and out-of-plane orbit maneuvers are enabled by setting firing attitude parameters.

Spacecraft: The Deimos-2 spacecraft is identical to the DubaiSat-2 minisatellite except for a battery capacity to account for a longer life. The nominal mission lifetime shall be 7 years, with a goal of 10 years. 6)

Deimos-2 features the SpaceEye-1 platform of SI (Satrec Initiative), a 300 kg class minisatellite bus (also referred to as SI-300). The SI-300 bus is of DubaiSat-2 heritage. Its architecture is designed to accommodate an Earth observation and/or science payload. The HEPS (Hall Effect Propulsion System) is installed for orbit control and maintenance. Accurate and agile three-axis attitude control supports precise imaging operations. Dual redundancies are adapted where necessary in the system architecture design to increase reliability of the satellite system.

The SI-300 platform, whose size is 1.5 m in diameter and 1.95 m in height, features a deck-and-longeron type structure permitting easy assembly and disassembly. A Li-ion battery and four deployable solar panels are equipped for power supply. The interface with the launch vehicle is made through an adapter bolted to the bottom of the structure.




Spacecraft launch mass

~310 kg

Payload inclusive

Spacecraft size

Ø 1,500 x 1,950 mm

Hexagonal shape

Agile platform

±45º pointing in cross-track (max)
±30º pointing in cross-track (nominal)

High-performance AOCS for pointing accuracy & stability

Power generation

450 W @ EOL

4 GaAs solar panels

GSD (Ground Sample Distance)

1 m (PAN), ~ 4 m (MS)

@ 600 km altitude, nadir view

Swath width

> 12 km


Spectral bands

PAN + 4 MS




Telemetry, telecommand & tracking

Image transmission

X-band with 160 Mbit/s downlink rate

QPSK modulation

On-board recording capability

256 Gbit

Equivalent to 1400 km observation strip

Table 1: Summary of Deimos-2 performance parameters

RF communications: The TT&C data are transmitted in S-band. The payload imagery is transmitted in X-band at a data rate of 160 Mbit/s (QPSK modulation). The imagery is stored on a high-capacity solid-state recorder (256 Gbit). The recorder compresses, encrypts and encodes the data in real-time during transmission. CCSDS encoding during transmission.


Figure 2: Two views of the deployed Deimos-2 spacecraft (image credit: SI)


Figure 3: Photo of the Deimos-2 flight model during integration in the clean room of Elecnor Deimos Satellite Systems (image credit: Deimos)


Launch: A launch of Deimos-2 is scheduled for the first half of 2014.

Orbit: Sun-synchronous near-circular orbit, nominal altitude 620 km, nominal LTAN (Local Time of Ascending Node) at 10:30 hours. Ultimately, a 4-day revisit time performance is enabled by the ±45º maximum off-nadir pointing capability.



Sensor complement: (EOS-D)

EOS-D (Electro-Optical Camera-D):

The EOS-D imager was designed and developed at SI (Satrec Initiative) of Daejeon, Korea.


Pushbroom type imager


- Korsch 4-mirror telescope of 42 cm aperture diameter
- Focal length = 5.7 m
- FOV = 1.2º

GSD (Ground Sample Distance)

0.75 m PAN, 4 m MS @ 590 km altitude

Spectral bands

Pan: 450-900 nm
MS1: 420-510 (blue)
MS2: 510-580 (green)
MS3: 600-720 (red)
MS4: 760-890 (NIR)

FPA (Focal Plane Assembly)

CCD detector with TDI capability

TDI detector exposure levels of the various bands

PAN: 6, 12, 24, 48
Blue, green: 4, 8, 16
Red, NIR: 2, 4, 8

Data quantization

10 bit

Swath width

12 km at nadir (24 km in wide area mode)

Data handling

Lossless compression, real-time encryption

Downlink data rate

160 Mbit/s (QPSK)

Table 2: Some parameters of the HiRAIS instrument (Ref. 5)


Figure 4: Illustration of the EOS-D pushbroom camera (image credit: SI, Deimos Space, Ref. 2)

Calibration: The imager underwent extensive pre-launch calibration activities (Ref. 5) involving geometric, spectral, and radiometric calibrations to characterize the sensor.

The on-board radiometric calibration techniques follow closely the CEOS CAL/VAL recommendations and sites:

• PRNU (Photo Response Non-Uniformity) measurements using uniform and high reflectivity LNES (Land Non Equipped Sites)

• Dome-C around winter solstice

• Greenland around summer solstice

• Dark signal measurements using night images over the north pacific, where light sources (natural or anthropogenic) are unlikely.

MTF measurement:

• The MTF estimation will be performed using the slanted edge methodology. Same as Deimos-1 but using artificial targets instead of field transitions

• The basis of this methodology is to oversample the edge taking advantage of its tilt along the track, assuming that it is straight

• Source data will be raw data after DS and PRNU correction, which is simply applying a linear function to each column to remove stripping. No convolution or resampling will be applied

• Viewing geometry and line rate will be taken into account.



Ground segment:

The ground segment has been completely developed in-house by Elecnor Deimos Space, in Madrid. Deimos Imaging (DMI), a company of the Elecnor Deimos Group, will be operating the Deimos-2 mission from its premises in Spain. The link between the space and ground segments will be performed thanks to the main GS (Ground Station) Puertollano (Spain) for both telemetry and telecommand and payload data (10 m antenna dish), and optionally to the GS of Svalbard (Norway) for payload data download only. 7) 8) 9)


Figure 5: EO ground system infrastructure (image credit: DEIMOS Space)

Deimos ground segments are based on the gs4EO (Ground Segment for Earth Observation) suite of state-of-the-art products. These products are the result of the know-how gathered for more than a decade of work for ESA (European Space Agency), customized to small Earth Observation missions.


Figure 6: gs4EO suite of products (image credit: Elecnor DEIMOS Space)

Ground segment architecture: Deimos-2 is the first mission where the complete suite of gs4EO products is being used. The Deimos-2 ground segment includes the complete on-ground facilities to control, monitor and commercially exploit the mission. Figure 7 shows the D2 GS high level architecture decomposition, a simplified view of the relations between the different ground segment elements, as well as the specific initial Ground Station setup for the Deimos-2 mission, with one main station in Puertollano, Spain, and a polar station (Svalbard). It is to be noted that nearly all the GS infrastructure is running in a virtualized HW environment. Archive & catalog, as well as data processing servers are easily scalable in this framework.



user4EO user services

Tool for user image acquisition request assessment with or without automatic connection to central plan4EO, together with advanced front-end for archive and catalogue query, visualization and dissemination

plan4EO mission planning

Mission plan generation for imagery acquisition and related data downlink activities to ground stations; as well as platform operations planning

fly4EO flight dynamics

Provides full orbit monitoring and maintenance capabilities, including the monitoring of space debris that would allow S/C operators to implement any required collision avoidance maneuver scenario(s) to be uplinked and then executed by the on-board SW

control4EO mission control system

Performs the monitoring and control of the satellite platform and payload, including real-time and back orbit HK (Housekeeping) TM reception, processing and visualization, telecommanding and onboard SW management

track4EO ground station control

Configuration of the ground station equipment and monitoring its operation during satellite passes

archive4EO archive & catalog

Hierarchical storage solution and meta data catalog storing archived files meta data. Catalog Web services (CSW) are used to allow access via HTTP interfaces through Description, Discovery & Publication ops.

process4EO image processing chain

Processing the payload raw data from the satellite to produce image products (radiometric and geometric corrections, up to fully ortho-rectified images)

monitor4EO monitoring & control

Data circulation, Ground Segment health monitoring and management of the complete automatic processing chain execution (orchestration)

calval4EO calibration & validation

Facility for updating and validating on-ground configuration parameters to ensure the image data quality requirements.

Table 3: Description of the gs4EO products


Figure 7: The Deimos-2 ground segment (image credit: Elecnor DEIMOS Space)

Two main chain are identified in the ground segment:

1) the flight operations chain, and

2) the data processing chain.

In both chains the user services element is involved, either as initiator or as final destination of the chain.


Figure 8: Flight operations chain of Deimos-2 (image credit: Elecnor DEIMOS Space)

Flight operations chain: The gs4EO FOS (Flight Operations Segment) addresses one of the main challenges for small commercial missions, the reduction of operations costs. One of the main costs drivers for commercial missions that required uninterrupted exploitation of the S/C payload, is the amount of operators required in the control center during working and non working hours to perform operations tasks or even to supervise that ongoing operations are performed according to plan. This is critical to prevent any unforeseen stop on the mission data return.

For this reason it is important for small commercial missions to reduce to a minimum the amount of man power required to operate the S/C but ensuring the maximum mission return. The gs4EO FOS addresses this critical aspect by two key features:

• Automation of nominal operations

• Grant secured access to Control Center functions from remote sites.

The GS4EO FOS provides the auto4EO component, a powerful automation infrastructure that allows all nominal satellite operations to be performed without operator intervention.

All gs4EO FOS components export their functions via programmatic APIs (Application Programming Interfaces). Those APIs can be called from scripts to access those functions, in a similar way that the operator works via the component HMIs (Human Machine Interfaces). Using the various APIs , scripts can be implemented to automate all required tasks related to satellite operations.

Scripts could, in principle, be implemented in any available scripting language, though for the gs4EO, support is provided only for a number of them that are of common use by the computing community such Python, Ruby and Javascript. Satellite operations related activities (as scripts) can be scheduled for execution by an operator, or even by other gs4EO FOS component within the auto4EO infrastructure as required. Scheduled scripts can then be triggered by time or even by the occurrence of predetermined operational events (e.g. AOS, LOS, etc). This makes the gs4EO FOS automation environment very dynamic, where the system can react to whatever occurs as result of the operations activity execution.


Figure 9: Data processing chain of Deimos-2 (image credit: Elecnor DEIMOS Space)

Data processing chain: The data processing function has been designed to operate automatically and in near-real time. Several levels of processing have been defined, from the stream of raw data produced by the instrument up to fully annotated, ortho-geolocated images of standard size. Two independent processing chains are defined:

• The radiometric processing chain, which translates the instrument counts to scientific units (radiances). This chain relies on specific auxiliary information provided by the instrument, and on a calibration and characterization data base that must be updated regularly during the mission lifetime.

• The geometric processing chain, which registers the different bands, geolocates every pixel and resamples the images to a variety of standard cartographic projections.

Other product derivatives, such as pan-sharpened images, three-dimensional scenes, and domain-specific products, are also supported as higher level processing levels. Multi-temporal analysis tools and methods, especially in the field of environmental monitoring, are a strong asset at Deimos.

Although the processing chain integrates well-known transformations that are generally applicable to most optical missions, the specificities of the Deimos-2 mission have been taken into account when selecting between alternative algorithms. In addition, the modular design has made it possible that for some of the processing steps, more than one alternative module is provided. For example, the geolocation and ortho-rectification can be accomplished by using either a physical sensor model, or a RFM (Rational Function Model). Performance considerations will be used to select the best configuration during the commissioning phase of the mission.

Notwithstanding the system’s capability to automatically ortho-rectify the product images, Deimos’ ground segment also supports human-driven ortho-rectification. In this approach, a human operator manually identifies ground control points in both the acquired image and a reference image, using tools specifically designed for this purpose. In the context of the Deimos-1 mission, this approach has shown to be more reliable and to achieve better accuracy than the fully automatic chain.

The data processing chain incorporates different standards such as:

• Catalog Web services following OGC (Open Geospatial Consortium) standards

• EO GML meta data for products

• OGC WPS for cluster distributed processing

• ESA’s Generic Interface for Instrument Data Processing (via job orders).


CAMP (Capacity Analysis and Mission Planning):

CAMP is an automated operational mission planning tool optimized for agile satellites; it provides both enhanced long-term mission return analyses and a sound prototype for a fully-automated mission planning chain, working on short-term operational horizons. 10)

The CAMP tool is constructed around two main modules (Figure 10). On top of the inputs shown on the diagram (orange blocks), both modules share a wide set of inputs describing the orbit, the ground segment and the 3 modelled platform resources: AOCS, power and data handling.


Figure 10: General architecture of the CAMP tool (image credit: Deimos Space)

The first module generates feasible MTLs (Mission Timelines, sequence of operations). From a set of AoI (Areas of Interest), it analyses the orbital geometry to find observation opportunities and builds full MTLs that respect the constraints imposed by the system resources (modelled with some approximations). It is also able to repeat the scheduling exercise and select the best-performing MTL from the point of view of mission return (taking also cloud forecast into account). Finally, it provides results and plots about the coverage performance of the selected MTL over the simulation time.

The second module simulates the execution of the selected MTL by the spacecraft, thanks to a high-fidelity system simulator. It combines the MTL and the high-fidelity orbit propagation (based on operational orbit determination) to derive the pointing angles to be uploaded to the satellite. It then uses state-of-the-art models to simulate the AOCS, the power management and the data flow. It produces a thorough reporting of the satellite system state at any moment and detects any possible resource conflict. An MTL coming from the MTL generator has no reason to create any conflict under nominal conditions, as it uses fair models of the on board resources plus security margins. But unexpected changes in the MTL (operator manual edit for contingency reasons) or in the orbit (emergency maneuvers) might overload the system. This final crosscheck is critical for the security of the system.

Automated mission timeline generation: The tool receives as input the users’ requests, represented by AoIs (Areas of Interest), and transforms them into targets. Through a geometric analysis, involving the propagated satellite orbit, the targets’ position and some user defined constraints (including cloud coverage forecast) all the possible observation events are selected. They are ordered by priority and then passed to a scheduler that, thanks to approximate models of the satellite resources, generates a mission timeline satisfying the platform constraints. The MTL performance is evaluated by means of a FoM (Figure of Merit) and the ordering and scheduling process can be repeated in order to maximize the mission return. The architecture of the automated mission timeline generator is shown in Figure 11.


Figure 11: Architecture of the CAMP automated MTL generator (image credit: Deimos Space)

Overall, the CAMP tool is able to automatically translate a set of customer areas of interest, with complex characterization and constraints, into a detailed sequence of operations. This MTL (Mission Timeline) includes all system events: station keeping, various modes of Earth observation, data downlink and orbit maintenance, plus all the satellite slews and tranquillization intervals between them. It is always compliant with on-board resources: agility, power and memory. Finally, it is chosen as the best-performing MTL amongst a set of peers generated during a process of optimization that can be driven by various algorithms.

1) Stefano Cornara, Blanca Altés-Arlandis, Matthias Renard, Stephania Tonetti, Fabrizio Pirondini, Roberto Alacevich, Annalisa Mazzoleni, “Mission Design and Analysis for the DEIMOS-2 Earth Observation Mission,” Proceedings of the 63rd IAC (International Astronautical Congress), Naples, Italy, Oct. 1-5, 2012, paper: IAC-12-C1.4.9

2) Fabrizio Pirondini, Enrique Gonzalez, “DEIMOS-2: Cost-Effective, Very-High Resolution Multispectral Imagery,” Proceedings of the 11th Annual JACIE (Joint Agency Commercial Imagery Evaluation ) Workshop, Fairfax, VA, USA, April 17-19, 2012, URL:

3) Fabrizio Pirondini, “The DEIMOS Earth Observation System,” , The Synergy of High Technology, Proceedings of Sovzond VI International Conference of Remote Sensing, Moscow, Russia, April 25-27, 2012, URL:

4) Diego Lozanno, “DEIMOS Optical missions,” 3rd GS Coordination Body Workshop, ESA/ESRIN, Frascati, Italy, June 6-7, 2012, URL:

5) Jorge Gil, Alfredo Romo, Cristina Moclan, Fabrizio Pirondini, Enrique Gonzalez, Jesus Quirce, “The Deimos-2 Mission: Pre and post-launch calibration and data validation,” 12th Annual JACIE (Joint Agency Commercial Imagery Evaluation) Workshop, St. Louis, MO, USA, April 16-18, 2013, URL:

6) “Satrec Initiative- Challenging Space Smart,” URL:

7) A. Monge, S. Negrín, O. González, J. A. González, A. Ortíz, “gs4eo: a new ground segment for earth observation missions,” Proceedings of the 64th International Astronautical Congress (IAC 2013), Beijing, China, Sept. 23-27, 2013, paper: IAC-13-B1.4.6

8) Diego Lozano, “DEIMOS optical missions,” 3rd GS Coordination Body Workshop ESA/ESRIN, Frascati, Italy, 6-7 June 2012, URL:

9) “Elecnor Deimos inaugurates its Puertollano Integration and Satellite Operation Centre, and presents its DEIMOS-2 Satellite,” Elecnor Press Release, October 8, 2013, URL:

10) M. Renard, S. Tonetti, B. Altés-Arlandis, S. Cornara, F. Pirondini, “Fully Automated Mission Planning Tool for DEIMOS-2 Agile Satellite,” Proceedings of the 64th International Astronautical Congress (IAC 2013), Beijing, China, Sept. 23-27, 2013, paper: IAC-13-B4.3.2

The information compiled and edited in this article was provided by Herbert J. Kramer from his documentation of: ”Observation of the Earth and Its Environment: Survey of Missions and Sensors” (Springer Verlag) as well as many other sources after the publication of the 4th edition in 2002. - Comments and corrections to this article are always welcome for further updates.