Matteo Sfragara: Queue-based random-access protocols for wireless networks
Time: Mon 2023-05-22 15.15 - 16.15
Location: 3721 (Lindstedtsvägen 25)
Participating: Matteo Sfragara, Stockholm University
AbstractIn this talk, we discuss mathematical models that address fundamental challenges in wireless networks. We first introduce Carrier-Sense Multiple-Access (CSMA) protocols, distributed algorithms that involve randomness to prevent the devices to transmit simultaneously and their signals to interfere with each other. These random-access models can be viewed as interacting particle systems on graphs, where the interference between signals in the network is captured by a hard-core interaction model on an appropriate interference graph. We then describe how these models exhibit metastability: when the activation rates become large, the system tends to stabilize in configurations with the maximum number of active nodes, with extremely slow transitions between them. As a consequence, individual nodes may experience prolonged periods of starvation, resulting in long queues and delays. Understanding metastability properties and slow transitions is thus crucial to design mechanisms to counter these effects and improve the performance of the network.
In particular, we focus on random-access models where the activation rates depend on the queues at the nodes. Not much is known about these queue-dependent models, since most of the literature deals with models with fixed activation rates. The joint activity state together and the joint queue length process evolve as a time-homogeneous Markov process, whose stationary distribution is challenging to analyze. We discuss three different network topologies: we start with complete bipartite networks, we then generalize our results to arbitrary bipartite networks, and finally we explore dynamic bipartite networks in which the interference graph changes over time, which allows us to capture some effects of user mobility.
This talk is based on three papers written during my PhD under the supervision of Frank den Hollander (Leiden University), Sem Borst (Eindhoven University of Technology) and Francesca R. Nardi (University of Florence).