Novel networking paradigms based on cloudification and open interfaces are making mobile networks more agile to adapt and meet the requirements of emerging use cases. Based on an automated end-to-end cloud-native open-source deployment of a 5G mobile network with over-the-air capabilities, this demonstration features the inclusion of a cloud-native monitoring system to follow the radio access network (RAN) performance evolution. By just defining the number of gNBs, this monitoring system automatically configures independent data sources and dashboards representing the evolution of received information while such gNBs are progressively created on-demand in a distributed environment.

The architecture and management of nextgeneration mobile networks are integrating novel networking paradigms to support challenging scenarios. This paper presents the design, implementation, and evaluation of a fully cloud-native, end-to-end mobile network integrating O-RAN functionality. Built on top of open-source software, our proposed framework develops a set of cloud-native deployment artefacts-such as Helm Charts and Open Source MANO packages-also released as open-source. This framework supports highly disaggregated and distributed deployments of key network functions, including the$\mathbf{5 G}$core, gNodeB (with decoupled Centralised and Distributed Unit entities), near-RT RIC, and xApps, enabling agile, portable, and flexible instantiations. Experiments show that the full cloudnative mobile network can be instantiated in less than five minutes, achieving a throughput comparable to bare-metal setups. This work highlights significant improvements in deployment efficiency and adaptability, contributing towards automated and autonomous cloud-native Beyond 5G/6G networks.

Integrating edge computing with 5G/6G, bolstered by standards such as ETSI Multi-access Edge Computing (MEC), becomes indispensable, particularly in addressing end-to-end latency-sensitive vertical applications. Thus, it is imperative to delve into infrastructure and network considerations, including the relocation of server workloads and optimization of user data planes. Notably, 3GPP facilitates third-party vertical providers’ access to user equipment (UE) metrics and data plane capacities through the Network Exposure Function (NEF). This paper showcases a demonstration that focuses on the seamless coordination of cloud-native workload relocation and user plane function (UPF) selection within the 5G network. Here, the interaction between a vertical provider and the 5G core enables informed decisions aimed at ensuring minimal latency by triggering UPF selection and relocation of containerized servers to the cloud or the edge, illustrating the anticipated trend of increased automation within 6G networks.

Novel networking paradigms, such as virtualization, are transforming mobile networks to support challenging use cases and scenarios. This demonstration presents an on-demand, distributed, and fully cloud-native deployment of a 5G mobile network (including core and RAN) with over-the-air capabilities using open-source software. This deployment strategy provides the flexibility and dynamicity required to manage and orchestrate forthcoming B5G/6G networks, thereby empowering mobile networks to adapt and scale effectively.

Effective and real-time management of data planes becomes of paramount importance to support the ever-increasing challenging use cases and scenarios foreseen in 5G and beyond. In this demonstration, we present a comprehensive showcase of a cloud-native open-source 5G core deployment, realised as a disaggregated non-public network (NPN) spanning multiple locations. Ranging from cloud to edge points of presence, the demonstration establishes an end-to-end experimental platform with over-the-air transmission capabilities, specifically highlighting the on-demand creation and deletion of User-Plane Functions (UPFs). This dynamic deployment approach harnesses the advantages offered by edge locations, empowering the mobile network to adapt and scale as per its specific requirements. Furthermore, our demo elucidates how to use Open Source MANO (OSM) for easing the management operations of network administrators.

The H2020 Distributed management of Network Slices in beyond 5G(MonB5G) project aims to provide zero-touch management and orchestration to support network slicing at scale to reduce the management burden on mobile operators by leveraging distribution of operations along with advanced data-driven Artificial Intelligence (AI)-based mechanisms. However, while this approach shows promise and large companies with abundant data and ML expertise are developing powerful MLdriven services, a critical aspect that remains to be analyzed is its business case. The vast majority of potentially valuable ML services, such as predictive maintenance, Quality of Service (QoS) optimization, network security enhancements, remain stuck at the idea or prototype stage. This paper delves into an analysis of how the MonB5G solutions in particular the tuples (Monitoring System (MS), Analytics Engine (AE), Decision Engine (DE) and Actuator (ACT) could be applied within the network management and orchestration market while investigating various business models and value chains. Numerical results based on experimental data have also been performed to evaluate the OpEX (Operational Expenditure) benefits associated with different network management techniques, for centralized and distributed systems.

This paper demonstrates an implementation of Federated Learning (FL) for streaming applications using cloud-native technology. Compared to a centralized management, by adopting a decentralized approach, the FL method improves convergence time, reduces communication overhead, and increases network energy efficiency. The cloud-native FL architecture presented comprises three sites, each with its own Kubernetes (K8s) cluster. The edge sites run FL Analytical Engines (AEs)/clients for local training and updates, and the central site runs the aggregation server for FL training. Some other relevant workloads deployed at the clusters are the video streaming server, the orchestrator, and monitoring components. As for the RAN, we showcase a multi-gNB setup from which we obtain monitoring data via custom sampling functions. Following the description of the testbed infrastructure and setup, this demonstration presents the real-time visualization of network parameters during FL training, and the enhancement of video streaming through proactive Central Processing Unit (CPU) scaling, made possible by the resource forecasting.

In this paper, we present WACA, the first system to simultaneously measure all 24 Wi-Fi channels that allow channel bonding at 5 GHz with microsecond scale granularity. With WACA, we perform a first-of-its-kind measurement study in areas including urban hotspots, residential neighborhoods, universities, and even a game in Futbol Club Barcelona’s Camp Nou, a sold-out stadium with 98,000 fans and 12,000 simultaneous Wi-Fi connections. We study channel bonding in this environment, and our experimental findings reveal the underpinning factors controlling throughput gain, including channel bonding policy and spectrum occupancy statistics. We then show the significance of the gathered dataset for finding insights, which would not be possible otherwise, given that simple channel occupancy models severely underestimate the available gains. Likewise, we characterize the risks of channel bonding due to other BSS’s, including their missed transmission opportunities and potential collisions due to imperfect sensing of bonded transmissions. We explore 802.11ax which imposes constraints on bonded channels to avoid fragmentation and defines different modes that can trade implementation complexity for throughput. Lastly, we show that the stadium, while seemingly too highly occupied for channel bonding gains, has transient gaps yielding impressive gains.

In this paper, we discuss the effects on throughput and fairness of dynamic channel bonding (DCB) in spatially distributed high-density wireless local area networks (WLANs). First, we present an analytical framework based on continuous-time Markov networks (CTMNs) for depicting the behavior of different DCB policies in spatially distributed scenarios, where nodes are not required to be within the carrier sense range of each other. Then, we assess the performance of DCB in high-density IEEE 802.11ac/ax WLANs by means of simulations. We show that there may be critical interrelations among nodes in the spatial domain-even if they are located outside the carrier sense range of each other-in a chain reaction manner. Results also reveal that, while always selecting the widest available channel normally maximizes the individual long-term throughput, it often generates unfair situations where other WLANs starve. Moreover, we show that there are scenarios where DCB with stochastic channel width selection improves the latter approach both in terms of individual throughput and fairness. It follows that there is not a unique optimal DCB policy for every case. Instead, smarter bandwidth adaptation is required in the challenging scenarios of next-generation WLANs.

Next generation of wireless local area networks (WLANs) will operate in dense, chaotic and highly dynamic scenarios that in a significant number of cases may result in a low user experience due to uncontrolled high interference levels. Flexible network architectures, such as the software-defined networking (SDN) paradigm, will provide WLANs with new capabilities to deal with users' demands, while achieving greater levels of efficiency and flexibility in those complex scenarios. On top of SDN, the use of machine learning (ML) techniques may improve network resource usage and management by identifying feasible configurations through learning. ML techniques can drive WLANs to reach optimal working points by means of parameter adjustment, in order to cope with different network requirements and policies, as well as with the dynamic conditions. In this paper we overview the work done in SDN for WLANs, as well as the pioneering works considering ML for WLAN optimization. Finally, in order to demonstrate the potential of ML techniques in combination with SDN to improve the network operation, we evaluate different use cases for intelligent-based spatial reuse and dynamic channel bonding operation in WLANs using Multi-Armed Bandits.

Despite being a relatively new communication technology, Low-Power Wide Area Networks (LPWANs) have shown their suitability to empower a major part of Internet of Things applications. Nonetheless, most LPWAN solutions are built on star topology (or single-hop) networks, often causing lifetime shortening in stations located far from the gateway. In this respect, recent studies show that multi-hop routing for uplink communications can reduce LPWANs' energy consumption significantly. However, it is a troublesome task to identify such energetically optimal routing through trial-and-error brute-force approaches because of time and, especially, energy consumption constraints. In this work we show the benefits of facing this exploration/exploitation problem by running centralized variations of the multi-arm bandit's e-greedy, a well-known online decision-making method that combines best known action selection and knowledge expansion. Important energy savings are achieved when proper randomness parameters are set, which are often improved when conveniently applying similarity, a concept introduced in this work that allows harnessing the gathered knowledge by sporadically selecting unexplored routing combinations akin to the best known one.