Game-Theory Based Policies For Flexible Spectrum Usage in IMT-Advanced Systems
Authors
Borgat, Erwann ; Palma, Valentina
Term
10. term
Publication year
2008
Pages
69
Abstract
Trådløse systemer af næste generation, som International Mobile Telecommunications-Advanced (IMT-A), sigter mod meget høje datahastigheder og kræver derfor brede frekvensbånd omkring 100 MHz. Radiofrekvensspektret er imidlertid begrænset og bruges ofte ikke optimalt, så det er ikke muligt at tildele 100 MHz til hver operatør i det samme område. Derfor er der behov for nye adgangsprotokoller til spektrum, der gør det muligt for flere operatører at dele frekvenser retfærdigt og effektivt. Afhandlingen undersøger en spilteoretisk løsning. Vi foreslår en dynamisk spilmodel, der anvender regret learning, hvor hver operatør tilpasser sin strategi baseret på tidligere resultater, til at håndtere spektrumdeling. Realistiske simuleringer viser, at operatører kan dele spektrum på en retfærdig og effektiv måde.
Next-generation wireless systems, such as International Mobile Telecommunications-Advanced (IMT-A), aim for very high data rates and therefore need wide frequency bands of around 100 MHz. However, radio spectrum is limited and often used inefficiently, so allocating 100 MHz to every operator in the same area is not feasible. This creates a need for new spectrum access protocols that let multiple operators share frequencies in a fair and efficient way. This thesis examines a game-theoretic approach. We propose a dynamic game model that uses regret learning, where each operator adapts its strategy based on past performance, to address the spectrum-sharing problem. Realistic simulations show that operators can share spectrum fairly and efficiently under this approach.
[This abstract was generated with the help of AI]
Keywords
Documents