Edge computing for space applications: Field programmable gate array-based implementation of multiscale probability...

Area of Science:

• Space Science and Engineering
• Computer Engineering
• Digital Signal Processing

Background:

• Space applications face data processing challenges due to limited bandwidth and high-resolution instruments.
• Significant on-board data loss occurs despite existing mitigation strategies.
• Efficient on-board data analysis is crucial for maximizing scientific return from space missions.

Purpose of the Study:

• To present a Field Programmable Gate Array (FPGA)-based architecture for on-board nonlinear data analysis and probability distribution function computation.
• To develop optimized FPGA implementations focusing on area and speed for efficient resource utilization.
• To reduce power consumption compared to traditional software-based data processing methods.

Main Methods:

• Designed and implemented an FPGA-based architecture for nonlinear data analysis.
• Developed two variants of the solution optimized for area and speed.
• Evaluated power consumption against classical software implementations.
• Tested the architecture with both synthetic and real-world space data.

Main Results:

• Achieved on-board nonlinear analysis and probability distribution function computation.
• Demonstrated efficient FPGA resource utilization through area and speed optimizations.
• Reduced power consumption by at least two orders of magnitude compared to software methods.
• Obtained excellent results with both synthetic and real data.

Conclusions:

• The proposed FPGA architecture offers an effective solution for on-board data processing in space applications.
• Optimized implementations ensure efficient use of limited FPGA resources on spacecraft.
• The significant power reduction and high performance pave the way for space-grade FPGA applications.