Grant: Discovery Grants Program - Individual
McMaster University
$40,000.00
Leveraging Hardware-Efficient Reconfigurable Architectures and Advanced Machine Learning Techniques to Revolutionize the Compute Stack of Neurotechnologies
FY 2024-25
Summary
Progress in neuronal sensing technologies has made remarkable advancements, enabling Brain-Computer Interfaces (BCIs) to act in real-time over ever-increasing neuronal populations, with neuron count doubling every seven years. Ultimately, all sensing technologies convert biosignals into the digital domain where sophisticated algorithms decipher these signals, enabling the alleviation of various neurological disorders. As advancements in sensing technologies accelerate, the downstream digital processing stack must also grow for the overall system to keep pace. The success of such systems hinges upon real-time processing, energy efficiency, and adaptability to user needs. Existing solutions are post-hoc, most of which are only appropriate for off-line analysis on a limited number of neurons, and often rely on commodity computing, resulting in limited portability and energy efficiency. For such systems to become practical, optimized algorithms and customized hardware are essential for real-time processing. While our previous solutions exhibit remarkable processing efficiency, they lack flexibility. It is imperative to develop adaptable BCIs that support multiple algorithms, enabling multiple therapies, while meeting the real-time and energy constraints. This necessitates developing purpose-built reconfigurable accelerators, leveraging application-specific features to create adaptable solutions and deliver orders of magnitude better performance and energy efficiency.
I intend to establish a pioneering research group with a long-term vision of contributing a computer-architect-targeted digital processing pipeline for efficient BCIs. Over the next five years, my research group will contribute scalable and highly efficient BCIs intended for integration with modern neurodevices by developing, designing, and demonstrating a novel class of hardware accelerators. These endeavors will be underpinned by the integration of cutting-edge Machine Learning (ML) algorithms and hardware. Our methodology will draw from the principles of neuromorphic computing and the utilization of asynchronous circuits to engineer ultra-low power designs, while reconfigurable implementations will be harnessed to craft adaptable solutions.
Program
Discovery Grants Program - Individual
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data
As published
Application id
RGPIN-2024-06875
Name nom
Abdelhadi, Ameer
Department département
Electrical and Computer Engineering
Organization id
27
Institution établissement
McMaster University
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
40000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1510
Committee name en
Electrical and Computer Engineering
Committee name fr
Génie électrique et informatique
Area of application code
1206
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Engineering
Area of application fr
Génie
Research subject code
2500
Research subject group en
Electrical and electronic engineering
Research subject group fr
Génie électrique et électronique
Research subject en
Electrical and electronic engineering
Research subject fr
Génie électrique et électronique
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Leveraging Hardware-Efficient Reconfigurable Architectures and Advanced Machine Learning Techniques to Revolutionize the Compute Stack of Neurotechnologies
Keywords
reconfigurable computing; field-programmable gate arrays (fpgas); hardware-efficient deep learning; computer architecture; custom-tailored application-specific hardware acceleration; energy-efficient computing hardware; high-performance computing hardware; embedded computing
Application summary
Progress in neuronal sensing technologies has made remarkable advancements, enabling Brain-Computer Interfaces (BCIs) to act in real-time over ever-increasing neuronal populations, with neuron count doubling every seven years. Ultimately, all sensing technologies convert biosignals into the digital domain where sophisticated algorithms decipher these signals, enabling the alleviation of various neurological disorders. As advancements in sensing technologies accelerate, the downstream digital processing stack must also grow for the overall system to keep pace. The success of such systems hinges upon real-time processing, energy efficiency, and adaptability to user needs. Existing solutions are post-hoc, most of which are only appropriate for off-line analysis on a limited number of neurons, and often rely on commodity computing, resulting in limited portability and energy efficiency. For such systems to become practical, optimized algorithms and customized hardware are essential for real-time processing. While our previous solutions exhibit remarkable processing efficiency, they lack flexibility. It is imperative to develop adaptable BCIs that support multiple algorithms, enabling multiple therapies, while meeting the real-time and energy constraints. This necessitates developing purpose-built reconfigurable accelerators, leveraging application-specific features to create adaptable solutions and deliver orders of magnitude better performance and energy efficiency.
I intend to establish a pioneering research group with a long-term vision of contributing a computer-architect-targeted digital processing pipeline for efficient BCIs. Over the next five years, my research group will contribute scalable and highly efficient BCIs intended for integration with modern neurodevices by developing, designing, and demonstrating a novel class of hardware accelerators. These endeavors will be underpinned by the integration of cutting-edge Machine Learning (ML) algorithms and hardware. Our methodology will draw from the principles of neuromorphic computing and the utilization of asynchronous circuits to engineer ultra-low power designs, while reconfigurable implementations will be harnessed to craft adaptable solutions.