AI Enabled Very Small Aperture Terminal Modem

A Very Small Aperture Terminal (VSAT) is a satellite communication technology that enables two-way data transmission between a ground station and a satellite. It is used for data, voice, and video communications, particularly in remote or underserved areas lacking adequate terrestrial infrastructure, ensuring reliable and secure communication for both personal and commercial needs.
Optimizing the satellite link in hubless full mesh VSAT technology is critical for achieving bandwidth efficiency, reducing latency, enhancing reliability, supporting scalability, and maintaining high Quality of Service. However, the resource constraints of hubless full mesh modems, due to the absence of a central network manager, pose significant challenges. Existing rule-based approaches limit the potential for smarter link management.
Artificial Intelligence (AI) has the potential to enhance the development of new modems supporting this technology. AI can optimize the features such as bandwidth and power allocation, adaptive coding and modulation, enable predictive maintenance, manage traffic and QoS. These advancements can result in more efficient, reliable, and cost-effective satellite communication networks.
This project will conduct a comprehensive study on suitable AI models for integration into embedded systems. The project will involve identifying appropriate AI models, optimizing these models for embedded systems, and developing the accelerator architecture for them.

Faculty Supervisor:

Otmane Ait Mohamed;Sébastien Le Beux;Sébastien Le Beux;Otmane Ait Mohamed

Student:

Partner:

PolarSat

Discipline:

Engineering

Sector:

Manufacturing

University:

Concordia University

Program: