Abstract
Edge applications evolved into a variety of scenarios that include the acquisition and compression of immense amounts of images acquired in space remote environments such as satellites and drones, where characteristics such as power have to be properly balanced with constrained memory and parallel computational resources. The CCSDS-123 is a standard for lossless compression of multispectral and hyperspectral images used in on-board satellites and military drones. This work explores the performance and power of 3 families of low-power heterogeneous Nvidia GPU Jetson architectures, namely the 128-core Nano, the 256-core TX2 and the 512-core Xavier AGX by proposing a parallel solution to the CCSDS-123 compressor on embedded systems, reducing development effort, compared to the production of dedicated circuits, while maintaining low power. This solution parallelizes the predictor on the low-power GPU while the entropy encoders exploit the heterogeneous multiple CPU cores and the GPU concurrently. We report more than 4.4 GSamples/s for the predictor and up to 6.7 Gb/s for the complete system, requiring less than 11 W and providing an efficiency of 611 Mb/s/W.
Highlights
IntroductionNamely the 128-core Nano, the 256-core TX2 and the 512-core Xavier AGX by proposing a parallel solution to the Consultative Committee for Space Data Systems (CCSDS)-123 compressor on embedded systems, reducing development effort, compared to the production of dedicated circuits, while maintaining low power
This solution parallelizes the predictor on the low-power graphics processing unit (GPU) while the entropy encoders exploit the heterogeneous multiple CPU cores and the GPU concurrently
Many remote sensing systems that generate multispectral and hyperspectral images (MHIs) incorporate the need of compression algorithms such as the Consultative Committee for Space Data Systems (CCSDS)-123 [1]. These are the cases of satellites and military drones, which impose severe power restrictions, creating the need of low-power systems and architectures capable of processing onboard MHI compression at acceptable cost [2]
Summary
Namely the 128-core Nano, the 256-core TX2 and the 512-core Xavier AGX by proposing a parallel solution to the CCSDS-123 compressor on embedded systems, reducing development effort, compared to the production of dedicated circuits, while maintaining low power. This solution parallelizes the predictor on the low-power GPU while the entropy encoders exploit the heterogeneous multiple CPU cores and the GPU concurrently.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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
