Reinforcement Learning for Real-time Wireless Scheduling and Edge Caching: Theory and Algorithm Design
Bidragets beskrivning
Machine Learning methods are used to solve radio resource management problems. In wireless communication networks, resource allocation decisions are often based on the solving optimization problems assuming a static network state and complete information about the current system state. The situation changes, if the objective of network management becomes to proactively take into account changes in network state when making resource allocation decisions. In this project, we address real-time scheduling, as well as edge and user caching decisions in situations with dynamically changing network state. The problems will be formulated as Markov Decision Processes, and reinforcement learning methods will be used to find solutions.
Visa merStartår
2022
Slutår
2024
Beviljade finansiering
Finansiär
Finlands Akademi
Typ av finansiering
Kahdenväliset yhteistyösopimukset, yhteishaut
Beslutfattare
Forskningsrådet för naturvetenskap och teknik
16.12.2021
16.12.2021
Övriga uppgifter
Finansieringsbeslutets nummer
345109
Vetenskapsområden
El-, automations- och telekommunikationsteknik, elektronik
Forskningsområden
Tietoliikennetekniikka
Identifierade teman
computer science, information science, algorithms