Back to results

National and Kapodistrian University of Athens

Hardware architectures of aerospace data systems for on-board hyperspectral image data processing/compression

Abstract

dc:description

Earth remote sensing is a crucial technique for gathering information about the planet’s surface, water areas and atmosphere. Hyperspectral imaging records data in continuous spectral bands, allowing the detailed characterization of materials and physical features. These systems generate huge amounts of data (hundreds of GBs per day), necessitating the need for efficient compression to optimize bandwidth usage and support real-time applications.Modern space missions demand sophisticated in-flight data processing techniques. Processing platforms range from general-purpose CPUs to specialized Radiation-Hardened-By-Design (RHBD) processors and Field-Programmable Gate Arrays (FPGAs). FPGAshave proven exceptionally well suited for space applications due to their excellent energy efficiency and high computing capabilities, achieving data rates that can exceed 1 Gbps. Hyperspectral data compression requires specialized algorithms that exploit spectral and spatial correlation. The CCSDS 123.0 standard is a prediction-based algorithm developed specifically for the compression of multispectral and hyperspectral data. The CSDS123.0-B-2 version extended the capabilities of the standard by introducing near-lossless compression and a new Hybrid Entropy Coder. The first contribution of the thesis focuses on the design and implementation of the Hybrid Entropy Coder architecture, including all pixel orderings (BIP, BIL, BSQ) of the CCSDS standard, consisting of specialized submodules with data pipelining for 1 sample/cycle performance. The architecture implemented in Xilinx Kintex Ultrascale XQKU060 FPGA achieves stable performance of 305 MSamples/s (@16 bpppb), consuming 2.05% LUTs (5086), 0.89% DSP (17) and 0.08% BRAM (1) for the dimensions of the AVIRIS instrument. The second contribution concerns the variation of the CCSDS 123. 0-B-2 standard with external quantization and the implementation of a high-performance 6-core parallel architecture by segmenting the hyperspectral cube towards the direction of the horizontal axis. The modification of the standard with external quantization has been proven to offer similar quality and compression performance to the original standard, while the implementation of the architecture in a Xilinx Kintex Ultrascale XCKU040 FPGA achieves 285 MSamples/s (4.56 Gbps) @ 1.68W for one core and 1590 MSamples/s (25.44 Gbps) @ 6.12W for six cores. This work demonstrates that high compression efficiency can be achieved with simplified architectures that avoid complex algorithmic optimizations. The proposed solutions balances between performance, quality, and power consumption, constituting them suitable for modern and future Earth Observation space missions. The hardware architectures presented in this doctoral thesis has achieved significant technological maturation and adoption, evidenced by three key milestones. The introduced architecture of Hybrid Entropy Coder has been integrated into the ESA Lossless/LossyMultispectral & Hyperspectral Compression IP Core (4000136723/22/NL/CRS), which will be included into the Agency’s portfolio of operational IP cores. Also, the Hybrid Entropy Coder architecture is part of the Payload Processing unit for the Traceable Radiometry Underpinning Terrestrial- and Helio- Studies (TRUTHS), which is planned to launch at 2030. A laboratory validation of the CCSDS 123.0-B-2 compression IP Core architecture with two cores was successfully conducted for radiation-hardened Xilinx Kintex Ultrascale XQRKU060 FPGA on an Airbus D&S platform in at the company’s facilities in Elancourt, France, as part of the European H2020 Hi-SIDE project, also receiving the Innovation in Space Award 2022. Furthermore, a separate implementation on a Xilinx Zynq 7045 SoC FPGA has been developed for an in-orbit demonstration aboard the ERMIS mission’s CubeSat constellation, scheduled for launch in March 2026. This implementation has attained Technology Readiness Level (TRL) 8, certifying it as a flight-ready, high-tech space product developed in Greece.

Degree

thesis:*
Grantor dc:publisher
National and Kapodistrian University of Athens
Year dc:date
2025

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Chatzioannou, Panagiotis
  • Χατζηαντωνίου, Παναγιώτης

Subjects

dc:subject × 16

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/60690

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Chatzioannou, Panagiotis; Χατζηαντωνίου, Παναγιώτης. Hardware architectures of aerospace data systems for on-board hyperspectral image data processing/compression. National and Kapodistrian University of Athens, 2025. http://hdl.handle.net/10442/hedi/60690