Tags:high-performance embedded computing, hyperspectral imaging spectrometer, on-board Clouds screening, on-board data compression, on-board data processing, Solid State Data Recorder (SSDR) and System-on-Chip
Abstract:
System-on-a-chip (SoC) devices promise lighter, smaller, cheaper, more capable and more reliable space electronic systems. This presentation describes the focal plane interface electronics – digital (FPIE-D) Xilinx Zynq-based data acquisition, cloud-screening, compression, storage and downlink computing system developed by the Jet Propulsion Laboratory (JPL), Alpha Data, Correct Designs and Mercury Systems for imaging sprectometers such as the NASA Earth Surface Mineral Dust Source Investigation (EMIT). EMIT is an imaging spectrometer that acquires 1280 cross-track by 328 band images at 216 images/sec. Following launch (14 July 2022), EMIT has been installed outside the International Space Station (ISS) and is collecting data from science targets in arid dust source regions of the Earth. The FPIE-D Alpha Data hardware and components, including a Mercury Systems 440 GByte RH3440 Solid-State Data Recorder (SSDR), fit into a 280mm×170mm×40mm assembly. The FPIE-D peak power usage is 40 W. The computing element is a Xilinx Zynq Z7100 which includes a Kintex-7 FPGA and dual-core ARM Cortex-A9 Processor. The FPIE-D Zynq Processing System is responsible for running the flight software, which includes command & data handling, command & telemetry with ISS over 1553 and science data downlink over a 7.4 Mbps Ethernet interface to the ISS.The FPIE-D includes three processing elements implemented in the Zynq PL: (1) the Fast Lossless extended (FLEX) data compression block (a modified implementation of the CCSDS-123.0-B-2 recommended standard), which is providing 3.4:1 lossless compression and 21 MSamples/sec throughput; (2) co-adding capability of two successive images so that shorter exposures can be used, helping to avoid saturation of the Focal Plane Array during acquisition; and (3) cloud detection and screening so that cloudy images can be dropped prior to compression, saving SSDR space and downlink time.
High-Performance Embedded Space Computing System-on-Chip for Space Imaging Spectrometer