Non-Orthogonal Multiple Access (NOMA) is a Resource Sharing Mechanism (RSM) initially studied for 5G cellular networks and brought back to the agenda for 6G networks. While NOMA’s benefit at the level of a single cell has been properly established, assessing its performance at the scale of a cellular network remains an open research problem. This is mainly due to the inter-dependencies between scheduling, power control and inter-cell interference. Some algorithms have been proposed to optimize resource allocation in a multi-cell network, but they require a perfect and unrealistic knowledge of the whole channel states. In this paper, we leverage Bayesian Optimization techniques to build a versatile evaluation framework, able to assess the performance of multi-cell networks implementing a large variety of RSMs under a minimal set of assumptions. Subsequently, we illustrate how this evaluation framework can be used to compare the performance of several well-known RSMs under various fairness requirements and beamforming efficiencies. Our results show that, among the RSMs studied on a simple multi-cell network simulation, NOMA combined with a full reuse policy consistently emerges as the one able to achieve the highest end-users achievable rates under fairness constraints.

Bayesian Optimization (BO) is typically used to optimize an unknown function $f$ that is noisy and costly to evaluate, by exploiting an acquisition function that must be maximized at each optimization step. Even if provably asymptotically optimal BO algorithms are efficient at optimizing low-dimensional functions, scaling them to high-dimensional spaces remains an open problem, often tackled by assuming an additive structure for $f$. By doing so, BO algorithms typically introduce additional restrictive assumptions on the additive structure that reduce their applicability domain. This paper contains two main contributions: (i) we relax the restrictive assumptions on the additive structure of $f$ without weakening the maximization guarantees of the acquisition function, and (ii) we address the over-exploration problem for decentralized BO algorithms. To these ends, we propose DuMBO, an asymptotically optimal decentralized BO algorithm that achieves very competitive performance against state-of-the-art BO algorithms, especially when the additive structure of $f$ comprises high-dimensional factors.

Bayesian Optimization (BO) has proven to be very successful at optimizing a static, noisy, costly-to-evaluate black-box function $f : \mathcal{S} \to \mathbb{R}$. However, optimizing a black-box which is also a function of time (*i.e.*, a *dynamic* function) $f : \mathcal{S} \times \mathcal{T} \to \mathbb{R}$ remains a challenge, since a dynamic Bayesian Optimization (DBO) algorithm has to keep track of the optimum over time. This changes the nature of the optimization problem in at least three aspects: (i) querying an arbitrary point in $\mathcal{S} \times \mathcal{T}$ is impossible, (ii) past observations become less and less relevant for keeping track of the optimum as time goes by and (iii) the DBO algorithm must have a high sampling frequency so it can collect enough relevant observations to keep track of the optimum through time. In this paper, we design a Wasserstein distance-based criterion able to quantify the relevancy of an observation with respect to future predictions. Then, we leverage this criterion to build W-DBO, a DBO algorithm able to remove irrelevant observations from its dataset on the fly, thus maintaining simultaneously a good predictive performance and a high sampling frequency, even in continuous-time optimization tasks with unknown horizon. Numerical experiments establish the superiority of W-DBO, which outperforms state-of-the-art methods by a comfortable margin.

WLANs, which have overtaken wired networks to become the primary means of connecting devices to the Internet, are prone to performance issues due to the scarcity of space in the radio spectrum. As a response, IEEE 802.11ax and subsequent amendments aim at increasing the spatial reuse of a radio channel by allowing the dynamic update of two key parameters in wireless transmission: the transmission power (TX_POWER) and the sensitivity threshold (OBSS_PD). In this paper, we present INSPIRE, a distributed online learning solution performing local Bayesian optimizations based on Gaussian processes to improve the spatial reuse in WLANs. INSPIRE makes no explicit assumptions about the topology of WLANs and favors altruistic behaviors of the access points, leading them to find adequate configurations of their TX_POWER and OBSS_PD parameters for the ''greater good" of the WLANs. We demonstrate the superiority of INSPIRE over other state-of-the-art strategies using the ns-3 simulator and two examples inspired by real-life deployments of dense WLANs. Our results show that, in only a few seconds, INSPIRE is able to drastically increase the quality of service of operational WLANs by improving their fairness and throughput.

The latest amendment 802.11ax to the IEEE 802.11 standard, better known by its commercial name Wi-Fi 6, includes a feature that aims at improving the spatial reuse of a channel: each device can adapt its Clear Channel Assessment sensitivity threshold and its transmission power. In this paper, we use the Multi-Armed Bandit (MAB) framework to propose a centralized solution to dynamically adapt these parameters. We propose a new approach based on a Gaussian mixture to sample new network configurations, a specific reward function that prevents starvations when maximized, as well as a method based on Thompson Sampling to select the best network configuration. We evaluate our solution using the network simulator ns-3 and different topologies. Simulation results confirm the large benefits that 802.11ax may bring to spatial reuse. They also demonstrate the efficiency of our solution in finding appropriate parameter configurations that significantly improve the quality of service of the networks.