1 / 30

Dual Core System-on-Chip Design to Support Inter-Satellite Communications

Dual Core System-on-Chip Design to Support Inter-Satellite Communications. EEL 6935 - Embedded Systems Dept. of Electrical and Computer Engineering University of Florida. Liza Rodriguez Aurelio Morales. Outline Introduction Picosatellite Demostrator Design Dual Core Processor Design

iniko
Télécharger la présentation

Dual Core System-on-Chip Design to Support Inter-Satellite Communications

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Dual Core System-on-Chip Design to Support Inter-Satellite Communications EEL 6935 - Embedded Systems Dept. of Electrical and Computer Engineering University of Florida Liza Rodriguez Aurelio Morales

  2. Outline • Introduction • Picosatellite Demostrator Design • Dual Core Processor Design • Dual Core Processor Implementation • Network Topology Reconfiguration • Conclusions

  3. Outline • Introduction • Picosatellite Demonstrator Design • Dual Core Processor Design • Dual Core Processor Implementation • Network Topology Reconfiguration • Conclusions

  4. Introduction Distributed Satellite System (DSS) • Satellites that provide multi-point sensing • Low cost, redundancy, flexibility • Types of DSS: • Formation Flying: strict formation • Clustering Mission: satellites are loosely coupled around each other • Virtual Satellite Mission (fractioned mission): a satellite has its subsystems divided onto multiple crafts (computing, imaging, etc.)

  5. Introduction DSS in Low Earth Orbit (LEO) • Small satellites deployed at the same time in multiple orbits • Use for disaster monitoring prevention • Ad-hoc network for multipoint sensing like WSN • Challenges: • Attitude and orbit control, intersatellite links, on-board computing • Deal with perturbations: Earth’s geophysical forces, solar radiation • Network connectivity and topology over time

  6. Introduction Distributed Computing Requirements • Node Level • At Individual satellite level • Store and forward data using the network: • High priority apps using Client/Server. Payload data through the network such imaging data • Low priority apps using Peer-to-Peer telemetry. Location and velocity changes, “byte” size payload data (GPS) • Network Level • Applied to multiple satellites • Provide adaptable and redundant ground-link communication schemes, main “sink” to ground • React proactively and reactively to their environment

  7. Motivation • Meet requirements for processing and network capabilities in “cluster” of satellites in the presence of space disturbances Proposal • Dual core System-on-Chip design using a general purpose soft-core processor and a specific core for real-time applications, such as agents Introduction

  8. Agenda • Introduction • Picosatellite Demonstrator Design • Dual Core Processor Design • Dual Core Processor Implementation • Network Topology Reconfiguration • Conclusions

  9. Picosatellite Demonstrator Design Prototype • Use of embedded hardware technology • Standard picosatellite platform CubeSat • For fast prototype, COTS components/boards: • Flight OBC and satellite chassis from Pumpkin • Power module from Clyde-Space • SGR-05 GPS module from SSTL • MHX transceiver from Microhard Systems • PF5100 Virtex-4 FPGA FX60 Board for SoC • IEEE 802.11 PC/104 Board from Elcard

  10. CubeSat Platform with Flight Module, IEEE 802.11, FPGA and development boards Prototype

  11. Power Module SGR-05U – Space GPS Receiver Flight module and satellite Chassis MHX 900 MHz Transceiver PF5100 Board with Virtex-4 FPGA IEEE 802.11 Board Prototype

  12. 1999, CalPoly and Stanford University developed specs to help universities worldwide perform space exploration. • Very small satellite • Use COTS components • 10x10x10 cm structure • Weight of 1 Kg • Also in 2U and 3U sizes CubeSat

  13. Demonstrator Satellite Architecture • FPGA board, IEEE 802.11 board, camera as payloads. • Architecture controlled by the Flight On-Board Computer (FM430 OBC) • SoC to act as HW/SW mediator for: • Hard and soft resets • Sleep mode • SoC also used as interface between various buses Demonstrator Satellite Architecture

  14. Agenda • Introduction • Picosatellite Demonstrator Design • Dual Core Processor Design • Dual Core Processor Implementation • Network Topology Reconfiguration • Conclusions

  15. LEON3 Processor • Synthesisable VHDL model of 32-bit processor compliant with SPARC V8 architecture • Suitable for SoC designs JOP Processor • Java Optimized Processor • Enables real-time Java functionality • Smallest and fastest Java core Dual Core Processor Design

  16. Dual Core Processor Design LEON3 Processor

  17. Dual Core Processor Design FPGA SoC design LEON3 core and JOP core in a FPGA AMBA = Advanced Microcontroller Bus Architecture APB = Advanced Peripheral Bus

  18. Dual Core Processor Design Design Considerations • Memory sharing system between LEON3 and JOP for access to external RAM • Cache between cores must maintain coherency • Reconfiguration in cases of single event upsets (SEUs) or single event latch-ups (SELs)

  19. Memory Footprint Comparison • System must have low memory footprint, including OS and network stack • System must be real-time • CLDC and pjava are designed for devices with intermittent network connection, slow processors, limited memory (e.g. mobile phones, PDAs), making them ideal for JOP core CLDC = Connection Limited Device Configuration pjava = PersonalJava JADE= Java Agent DEvelopment Framework LEAP =Light Extensible Agent Platform CORBA = Common Object RequestBroker Architecture Dual Core Processor Design Multi-layer software design

  20. Dual Core Processor Design Hardware and software layer design LEON3 and JOP in FPGA: • Reduce memory footprint • Increase FPGA utilization • Enable Java apps, such as Agents, for real-time apps RTEMS = Real-Time Exceutive for Multiprocessor Systems

  21. System-on-Chip Block Diagram Dual Core Processor Design Detailed System-on-Chip design

  22. Outline • Introduction • Picosatellite Demonstrator Design • Dual Core Processor Design • Dual Core Processor Implementation • Network Topology Reconfiguration • Conclusions

  23. Dual Core Processor Implementation Timing Results • Max Frequency of 37.398 MHz • WCET found between: • LEON3 and AMBA memory controller • LEON3 and JOP AHB Master • JOP cache and JOP address bus • Speed optimization is needed to satisfy IEEE 802.11 MAC. Trade-off between area and speed

  24. Resource Utilization Memory Trade-off • On-chip or off-chip memory? • Speed and power requirements • On-chip: fast but increase power consumption and area • Power consumption of SoC design: 2.33W (1.76W in memory interfacing), using XPower from Xilinx Dual Core Processor Implementation

  25. Outline • Introduction • Picosatellite Demonstrator Design • Dual Core Processor Design • Dual Core Processor Implementation • Network Topology Reconfiguration • Conclusions

  26. HW & SW are discovered • Network topology can be reconfigured Network Topology Reconfiguration Procedure • Stage 1: Startup FPGA Bus System & LEON3 • LEON3 started, id and starting tasks discovered • Stage 2: Startup JOP & JADE-LEAP • Start Java application with argument passing to main host and services required • Stage 3: Network Topology Refresh • Initialize, check or change the network topology

  27. Outline • Introduction • Picosatellite Demonstrator Design • Dual Core Processor Design • Dual Core Processor Implementation • Network Topology Reconfiguration • Conclusions

  28. Conclusions • A COTS solution for picosatellite including a SoC design that meets CubeSat platform was introduced. • LEON3 IP and JOP IP cores were used to meet strict requirement of low memory footprint, Java functionality and real-time operation • An Java agent software was proposed to support inter-satellite communication based on IEEE 802.11 wireless connectivity

  29. References • http://ieeexplore.ieee.org/search/wrapper.jsp?arnumber=4584273 • http://www.cubesat.org/ • http://en.wikipedia.org/wiki/CubeSat • http://www.cubesatkit.com/index.html • http://www.derivation.com/products/pf5100.html • http://www.clyde-space.com/products/electrical_power_systems/cubesat_power • http://www.sstl.co.uk/assets/Downloads/SGR-05U%20v1_13.pdf • http://www.data-connect.com/Microhard_MHX-910.htm • http://www.gaisler.com/doc/leon3_product_sheet.pdf

  30. Questions?

More Related