LoRaWAN is emerging as an attractive network infrastructure for ultra low power Internet of Thing devices. Albeit the technology itself is quite mature and specified, how to effectively allocate wireless resources so as to support a large amount of devices in a same terrestrial area is an open challenge. This paper contributes by proposing two algorithms (of incremental complexity) which are shown to outperform the basic Adaptive Rate Strategy (ADR) so far considered. A first approach, named EXPLoRa-SF, shows the benefits of a simple strategy which does not limit to use (as per ADR) distance/RSSI measurements, but also selects Spreading Factors (SF) based on the total number of connected devices. The advantages attained with EXPLoRa-SF further lead us to propose a more sophisticated algorithm, named EXPLoRa-AT, which employs an innovative “ordered waterfilling” approach which attempts to allocate the spreading factors so as to equalize the Time on Air of the packets transmitted by the system's end devices in each spreading factor's group. Simulation results show that the proposed algorithms significantly outperform the basic ADR strategy, and particularly the EXPLoRa-AT algorithm appears very robust to different operating conditions and consistently guarantees high bit rates in the case of high traffic loads.

Sensor networks exhibit a unique funneling effect which is a product of the distinctive many-to-one, hop-by-hop traffic pattern found in sensor networks, and results in a significant increase in transit traffic intensity, collision, congestion, packet loss, and energy drain as events move closer toward the sink. While network (e.g., congestion control) and application techniques (e.g., aggregation) can help counter this problem they cannot fully alleviate it. We take a different but complementary approach to solving this problem than found in the literature and present the design, implementation, and evaluation of a localized, sink-oriented, funneling-MAC capable of mitigating the funneling effect and boosting application fidelity in sensor networks. The funneling-MAC is based on a CSMA/CA being implemented network-wide, with a localized TDMA algorithm overlaid in the funneling region (i.e., within a small number of hops from the sink). In this sense, the funneling-MAC represents a hybrid MAC approach but does not have the scalability problems associated with the network-wide deployment of TDMA. The funneling-MAC is 'sink-oriented' because the burden of managing the TDMA scheduling of sensor events in the funneling region falls on the sink node, and not on resource limited sensor nodes; and it is 'localized' because TDMA only operates locally in the funneling region close to the sink and not across the complete sensor field. We show through experimental results from a 45 mica-2 testbed that the funneling-MAC mitigates the funneling effect, improves throughput, loss, and energy efficiency, and importantly, significantly outperforms other representative protocols such as B-MAC, and more recent hybrid TDMA/CSMA MAC protocols such as Z-MAC.

5G and beyond mobile networks will support heterogeneous use cases at an unprecedented scale, thus demanding automated control and optimization of network functionalities customized to the needs of individual users. Such fine-grained control of the Radio Access Network (RAN) is not possible with the current cellular architecture. To fill this gap, the Open RAN paradigm and its specification introduce an “open” architecture with abstractions that enable closed-loop control and provide data-driven, and intelligent optimization of the RAN at the user-level. This is obtained through custom RAN control applications (i.e., xApps) deployed on near-real-time RAN Intelligent Controller (near-RT RIC) at the edge of the network. Despite these premises, as of today the research community lacks a sandbox to build data-driven xApps, and create large-scale datasets for effective Artificial Intelligence (AI) training. In this paper, we address this by introducing ns-O-RAN , a software framework that integrates a real-world, production-grade near-RT RIC with a 3GPP-based simulated environment on ns-3, enabling at the same time the development of xApps, automated large-scale data collection and testing of Deep Reinforcement Learning (DRL)-driven control policies for the optimization at the user-level. In addition, we propose the first user-specific O-RAN Traffic Steering (TS) intelligent handover framework. It uses Random Ensemble Mixture (REM), a Conservative $Q$ -learning (CQL) algorithm, combined with a state-of-the-art Convolutional Neural Network (CNN) architecture, to optimally assign a serving base station to each user in the network. Our TS xApp, trained with more than 40 million data points collected by ns-O-RAN, runs on the near-RT RIC and controls the ns-O-RAN base stations. We evaluate the performance on a large-scale deployment with up to 126 users with 8 base stations, showing that the xApp-based handover improves throughput and spectral efficiency by an average of 50% over traditional handover heuristics, with less mobility overhead.

The topology of a cognitive radio Ad Hoc network is highly influenced by the behavior of both licensed (primary) and unlicensed (secondary) users. In fact, the network connectivity could be impaired by the activity of primary users. This aspect has a significant impact on the design of routing protocols. We design a routing scheme for cognitive radio Ad Hoc networks, named Gymkhana, which is aware of the degree of connectivity of possible paths towards the destination. Gymkhana routes the information across paths that avoid network zones that do not guarantee stable and high connectivity. To this aim we use a mathematical framework, based on the Laplacian spectrum of graphs, that allows a comprehensive evaluation of the different routing paths of the cognitive radio network. Laplacian matrixes are used to compute the connectivity of the different network paths. Gymkhana uses a distributed protocol to collect some key parameters related to candidate paths from an origin to a destination. The parameters are fed into a basic mathematical structure which is used to compute efficient routing paths. Besides the basic idea of Gymkhana, the use of Laplacian matrixes to derive a closed formula to measure the path connectivity is another contribution of ours.

The advancements of the Internet of Things and Low-Power Wide-Area Network technology will accelerate in the next future the adoption of smart meters in water distribution systems, enabling the collection of a huge amount of fine-grained data. How to turn massive smart meter data into actionable knowledge will be the key point to limit water wastage and promote efficient and sustainable distribution. Although the collection of data worldwide is currently limited, the potential future impact of exploiting data-driven and machine learning methods is increasingly recognized in research and industry, as shown by many scientific works published in recent years. In particular, the interest in deep learning for smart water distribution systems is increasing, motivated by the ability to learn intricate patterns from big data. This work aims to provide an overview of the current research and identify challenges for future directions by conducting an application-oriented survey. Specifically, by analysing data characteristics and operational targets, we propose a new taxonomy that helps structure properly the macro-areas of water management into infrastructure analysis, demand analysis and water quality monitoring. Existing methods are discussed for each application under these three stages. In addition, we also discuss potential research directions, such as federated learning, incremental learning, probabilistic modeling and explainability and address broad issues like data availability and implications for privacy.