Abstract

This paper presents the DSP technologies for controlling & processing as well as advanced technologies for high speed processing of data downlink system of Earth Observation (EO) payloads. These technologies are under development by OHB-System for multiple small and medium EO-satellites (e.g. the DIAMANT ≈200kg satellite class, or the KOMPSAT-II ≈700kg satellite class). Future missions for high resolution (spatial, spectral and radiometric) earth observation will have significant increased performance requirements for onboard controlling and data processing. High performance of data storage and down-link systems is already required for typical missions today. Advanced technologies for user specific and theautomatic on-board data processing and downlink services are driven by new applications and smarter concepts. Applications for data pre-processing are for example autonomous object/event detection using a wide field sensor. After on-line detection of events, the selected area will be investigated in more detail using a high resolution sensor steered automatically to the area of the event by the spacecraft itself. Further a modular design is desired to provide maximum flexibility in order to insert other processing elements into the processing chain, e.g. to truncate data in front of the compression or the data storage is not required as the mission wishes online downlink only. OHB-System provides the following products for this applications: n DSP Board (VSOP) for high performance controlling and pre-processing / compression n Data Storage Unit (DSU) for high speed and capacity storage. n Channel-Coding Unit (CCU) for high speed image data processing and formatting. n EGSE & SCOE for high-speed sub-units check-out and system tests. In the following chapters the status of the current implementations and the applications of the above described elements. 54th International Astronautical Congress of the International Astronautical Federation, the International Academy of Astronautics, and the International Institute of Space Law 29 September 3 October 2003, Bremen, Germany IAC-03-B.3.09 Copyright © 2003 by the International Astronautical Federation. All rights reserved. 2 Figure 1: VSOP DSP PROCESSOR BOARD & CHANNEL CODING – REALTIME ENCRYPTION BOARD HIGH SPEED PROCESSING Future missions for high resolution (spatial, spectral and radiometric) earth observation will have significant increased performance requirements for onboard data processing. High performance of data storage and down-link is already required for typical missions today. The following presents a modular system architecture as being used for DIAMANT/MSRS but also other project running at OHB-System. Major technology driver are: 1) high speed data processing 2) modular & scaleable design 3) cost effectiveness Data rates can easily reach 70 up to 1.800 Mbit/s per spectral channel for a ground resolution of about 5 to 1 meter at 10 bit spectral resolution. Assuming 5 meter ground resolution and a scene size of 50x700km, one would have to store already 84 Gbit per spectral channel on-board the satellite. For 12 channel that aims into a storage capacity of 1.008Gbit, which is enormous in terms of power issues. Assuming further a down-link capacity of 280Mbit/s (actual technology) one would need 1 hour for the downlink. This short example shows, that high speed data processing on-board the spacecraft is necessary to reduce the data amount to the really needed information part. One example for this pre-processing is data compression, other examples are data evaluation and truncation to the area of interest. A modular design is desired to provide maximum flexibility in order to insert other processing elements into the processing chain. For example there might be the need to truncate data in front of the compression or the data storage is not required as the mission wishes online downlink only (GEO applications). Further, programmatic aspects can lead to a separation of the processing chain into different manufactures. A scaleable design is desired to easily adapt the processing chain to different numbers of spectral channels or missions requiring high redundant design. The modularity / scalability brings only a small drawback concerning mass & power of about 10-15%. But it leads to a cost effective design. However, for very dedicated missions one can still combine elements of the processing chain like compression, storage and channel coding. This combination would give an optimised mass & power budget. The system architecture for a modular scalable high speed processing chain is given in Figure 2. The Camera Unit itself is not part of the processing chain and shown for overview purposes. The Camera Unit contains the detectors and provides finally the A/D converted digital signal stream as input for the processing chain. Figure 2 show the processing chain handling a data input stream of 1.2Gbit/s with an output rate of 280Mbit/s which is currently reachable by analogue X-Band transmitter. The input data stream consists of 12 spectral channel which are multiplexed to 4 inputs each with 3 channels with a data rate of 300Mbit/s. Multiplexing is introduced to have a small light weight harness.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call