Gabriela Mystkowska

Mrs. Gabriela Mystkowska

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
6 August 2025 in IEEE Access
Verified
Journal Article
Open access
Gabriela MystkowskaGabriela Mystkowska
Gabriela Mystkowska
Matteo MonopoliMatteo Monopoli
Matteo Monopoli
Pietro NannipieriPietro Nannipieri
Pietro Nannipieri
LZLuca Zulberti
Luca Zulberti
DCDavid Merodio Codinachs
David Merodio Codinachs
LFLuca Fanucci
Luca Fanucci
Artificial Intelligence is an emerging technology in the space industry as it enables real-time data processing and autonomous navigation. We aim to facilitate mission development by providing an extensive state-of-the-art review. We analyze recent advancements in hardware architectures for space on-the-edge applications. We explore the unique challenges of space environments, including radiation hardening, flexibility, and real-time processing needs. We identify the possible hardware accelerator architectures and assess their applicability on-board spacecraft. We examine key developments in hardware accelerators, general-purpose processors, and neuromorphic chips, with an emphasis on adaptability for space applications. We compare the available chips in terms of performance capabilities, power consumption, technology node and radiation hardening. We showcase future directions, offering a comprehensive overview of cutting-edge solutions for AI-driven space missions.

28 August 2024
Verified
Conference Paper
Closed access
Gabriela MystkowskaGabriela Mystkowska
Gabriela Mystkowska
LZLuca Zulberti
Luca Zulberti
MMMatteo Monopoli
Matteo Monopoli
PNPietro Nannipieri
Pietro Nannipieri
LFLuca Fanucci
Luca Fanucci
The rapid development of Artificial Intelligence (AI) algorithms has created a need for a resource-optimised hardware accelerator. Among various platforms, Coarse-Grained Reconfigurable Array (CGRA) have gained importance as on-edge accelerators. They comprise of heterogeneous Processing Element (PE) matrix, which allows for high flexibility and parallelisation of calculations. They are mainly used for speeding up Data Flow Graph (DFG) execution. We aim to provide a general purpose, highly parameterised, and flexible architecture for AI on-edge data crunching. We propose a CGRA with a vector extension which allows for dynamically adjustable precision of calculation while maintaining a desired performance-power-area optimisation. It targets 4 bits integer (INT4) and 8 bits integer (INT8) quantization for fast and efficient Neural Network (NN) processing. In this paper, we examined hardware costs required to support the vector extension functionality. We synthesised the design on the 40nm Standard-Cell technology from TSMC. The obtained results show that the proposed extension attains on average 28.2% decrease in power consumption and 21.6% decrease in area compared to a reference design of the same computation power.

2 October 2024
Verified
Preprint
Open access
Gabriela MystkowskaGabriela Mystkowska
Gabriela Mystkowska
LZLuca Zulberti
Luca Zulberti
MMMatteo Monopoli
Matteo Monopoli
PNPietro Nannipieri
Pietro Nannipieri
LFLuca Fanucci
Luca Fanucci
The rapid development of Artificial Intelligence (AI) algorithms has created a need for a resource-optimised hardware accelerator. Among various platforms, Coarse-Grained Reconfigurable Array (CGRA) have gained importance as on-edge accelerators. They comprise of heterogeneous Processing Element (PE) matrix, which allows for high flexibility and parallelisation of calculations. They are mainly used for speeding up Data Flow Graph (DFG) execution. We aim to provide a general purpose, highly parameterised, and flexible architecture for AI on-edge data crunching. We propose a CGRA with a vector extension which allows for dynamically adjustable precision of calculation while maintaining a desired performance-power-area optimisation. It targets 4 bits integer (INT4) and 8 bits integer (INT8) quantization for fast and efficient Neural Network (NN) processing. In this paper, we examined hardware costs required to support the vector extension functionality. We synthesised the design on the 40nm Standard-Cell technology from TSMC. The obtained results show that the proposed extension attains on average 28.2% decrease in power consumption and 21.6% decrease in area compared to a reference design of the same computation power.