Effective congestion control is critical for wireless networks, where rapidly varying channel conditions and diverse traffic demands can severely degrade performance. Traditional congestion control algorithms rely on static heuristics that are often ill-suited for dynamic wireless environments. In this paper, we introduce Flecto, a Reinforcement Learning (RL)-based congestion control solution integrated into the QUIC protocol that, leveraging cross-layer metrics, including Signal-to-Noise Ratio, Block Error Rates, and Round-Trip Time measurements, can take decisions using a comprehensive view of network conditions. We implemented Flecto on a 5G testbed using OpenAirInterface and ETTUS USRP B210 radios, showing how it adapts transmission rates in real-time to maximize throughput and minimize latency while maintaining stability. Experimental results show that Flecto achieves an average throughput of 4539.5 KB/s approximately 6% higher both than Cubic (4267.2 KB/s) and New Reno (2674.1 KB/s) while reducing the average Round-Trip Time to 21.8 ms, significantly lower than Cubic’s 27.6 ms and New Reno’s 174.9 ms. These performance gains underscore the promise of integrating RL with cross-layer feedback for adaptive, efficient congestion control in next-generation wireless networks. Moreover, the modular design of Flecto facilitates its extension to other transport protocols and multi-user scheduling frameworks, paving the way for broader adoption in future wireless systems.

The escalating complexity and critical role of Operations and Maintenance Systems (OMS) in 5G networks require robust Disaster Recovery (DR) solutions to ensure uninterrupted service and minimal downtime. Disaster Recovery Systems (DRS) are crucial for network resilience, enabling seamless failover and recovery during disruptions. The switchover function is pivotal for maintaining 5G service continuity. This study addresses the limitations of traditional rule-based decision-making, which often employs simplistic binary switchover logic inadequate for 5G’s intricate demands. We propose the predictive OMS Switchover (pOM2S), a machine learning-driven approach that leverages historical computing and networking data to select the optimal OMS backup instance from multiple candidates. Evaluated in a 5G emulation testbed, LightGBM and CatBoost outperformed other models like Gradient Boosting, XGBoost, and Elastic Net Regression, achieving superior accuracy and efficiency. Compared to a baseline method relying solely on computational KPIs, pOM2S’s holistic approach selected a backup instance with a predicted switchover time of over twice as fast as the baseline’s. This demonstrates pOM2S’s effectiveness in enhancing DR efficiency and service continuity in 5 G networks.

The rapid evolution of 5G networks has introduced unprecedented challenges in maintaining service continuity, particularly in the eHealth mission-critical vertical. This paper presents the Proactive Disaster Recovery System (PDRS), a machine learning-driven disaster recovery system for 5G Operations and Maintenance Systems (OMS) in mission-critical eHealth verticals. Unlike traditional reactive approaches using binary switchover logic, PDRS enables proactive failover by: (1) continuously monitoring OMS status through KPIs, (2) predicting disasters via real-time analytics, and (3) selecting the optimal OMS backup instances from multi-redundant candidates using migration cost estimates. Proof-of-concept evaluation demonstrates PDRS’ superiority over baseline methods in maintaining service continuity, particularly for low-latency eHealth applications requiring 99.999% availability. Results highlight the necessity of predictive strategies for 5G network resilience in critical medical services.

One key challenge in Adaptive Video Streaming is the ever-changing edge network conditions at the last mile of access networks. The edge environment is particularly dynamic, influenced by user locations, fluctuation in resource demands and resource capabilities, in contrast to traditional Content Delivery Network (CDN) setups, where content routing decisions are relatively known. To address the dynamism of edge computing environments and to enable applications to better exploit edge-cloud computing resources, this article focuses on content steering technology, a recent addition to adaptive video protocols such as HLS and DASH. We present AVENUE as an architecture for Content Steering Services to orchestrate video delivery dynamically across the Edge-Cloud Continuum. This work designs the principles of the Content Steering Service to create a mechanism that involves two modules - monitoring and selector: The monitoring module captures real-time context metrics, and the selector module chooses an edge server according to the Select Server Algorithm. Our study addresses three steering algorithms with different performance profiles. Numerical results demonstrate that different configurations may yield varying network performance in terms of Quality of Experience (QoE), cache hits, and request load. Moreover, the appropriate selection of heuristics in the Selector module can also have a significant impact, depending on the metric being evaluated.

Rapid technological advancement has revolutionized the acquisition, processing, and storage of personal data, with notable data breaches from significant corporations emphasizing the value of data and the need for enhanced privacy protection. This has led to a global focus on individual privacy by enacting privacy-centric laws. The healthcare sector, known for its data sensitivity, presents distinct challenges necessitating stringent privacy protocols. The healthcare sector, known for its data sensitivity, presents distinct challenges necessitating stringent privacy protocols. Thus, there is a critical need for robust data privacy measures, including (pseudo)anonymization, to address this shift. This paper introduces a tailored multilevel (pseudo)anonymization architecture designed for healthcare data, capable of ensuring secure data handling and precise anonymization across various sources, even in a (pseudo)anonymized state. The proposed architecture was developed as a proof of concept and underwent thorough evaluation through a series of experiments. The outcomes are encouraging by showcasing effectiveness in achieving accurate anonymization, secure data linkage, and supporting re-identification when essential for individual security.

Industrial Internet-of-Things (IoT) massively deploys intelligent computing in industrial production and manufacturing environments seeking automation, reliability, and control. Machine Learning models provide intelligent decisions to drive manufacturing systems to the next level of productivity, efficiency, and safety. One of the critical challenges that must be faced is the deployment of Machine Learning models at the network edge to detect data anomalies caused by Industrial IoT hardware failures, since industrial IoT devices are prone to errors and failures. These anomalies can harm the industrial IoT system by producing false alarms, consuming network resources, and affecting productivity. Because of that, it is critical to rely on low latency and high precision detection systems to verify the data received from industrial IoT devices. In light of this, we assessed key performance indicators of five machine learning models running at edge computing, to provide in-depth discussions. The performance results were obtained from an oil refinery scenario using a real industrial IoT dataset. The performance was measured in terms of (a) Accuracy, (b) Precision, (c) Recall, (d) F1 score, (e) Training time, and (f) Response time.

The efficient design of SFC-enabled eHealth applications requires an accurate provision of the underlying infrastructure. This provision requires both computing and networking resources to meet stringent QoS requirements under any conditions of service demand. Cloud providers often offer automatic elasticity strategies based on monitoring specific metrics that lead to a waste of resources, time/energy consumption, and the problem of starvation with competing services. Our findings provide evidence that proactive-based elasticity overcomes these issues, when assisted by Machine Learning (ML) methods for predicting Internet traffic load. An optimal autoscaling algorithm depends on high precision and fast predictions to provide accurate results. Thus, this paper assesses ML algorithms to support SFC-enabled eHealth vertical applications. The experimental results suggest that the evaluated models achieved similar accuracy metrics, with an MLP architecture delivering the best performance in terms of time training and average prediction time.

The constant sharing of information among vehicles is of vital importance to provide different types of service in Intelligent Transportation Systems (ITS). Typically, ITS apply the sharing benefit to carrying out tasks such as extracting knowledge of vehicle traffic conditions and its distribution. The ITS that use this approach are able to perform the knowledge distribution, however, they lack of mechanisms to select the most appropriate vehicles to do so. It is common, in these systems, such tasks are performed by all vehicles. Consequently, it could easily cause a network overhead because of the highly redundant knowledge about the traffic that is being transmitted. With this in mind, we propose a system for information management and knowledge distribution named TRUSTed. The proposed system applies the egocentric betweenness measure to select the most relevant vehicle to carry out such tasks. Simulation results have shown that TRUSTed outperforms other systems found in the literature in several requirements.

Online video transmissions over wireless networks are rising in popularity and have already become part of our daily life. In the meantime, it is necessary to address a number of challenges ranging from the scarce resources, time-varying, and high error rates, to the fluctuating bandwidth, unveiling the need for an adaptive mechanism to ensure a good video transmission. Adaptive Forward Error Correction (FEC) techniques with Quality of Experience (QoE) assurance are appropriate to deliver QoE-aware video data to wireless users in dynamic and high error rates networks. This paper proposes an adaptive Video-aware FEC and Fuzzy Logic-based mechanism to shield realtime video transmissions against packet loss in wireless networks, improving both user experience and the usage of resources. The benefits and drawbacks of the proposed mechanism compared with exiting work are demonstrated through simulations and assessed with QoE metrics.

Video transmission over wireless networks has shown a great increase in recent years and it is becoming part of our daily life. Meanwhile, several difficulties can impair the success of the transmission, such as limited network resources, high error rates and fluctuating signal strength that may lead to variable bandwidth. Therefore there is the need for adaptive mechanisms that can provide a good video transmission. Adaptive Forward Error Correction (FEC) techniques which assure Quality of Experience (QoE) are a convenient means of delivering video data to wireless users in dynamic and error prone networks, while taking into account the content of the transmitted data. This paper proposes an adaptive content-aware and Random Neural Network (RNN) based mechanism to provide protection of real-time video streams against packet loss in wireless networks, improving user experience and optimising network resources. The benefits of the proposed mechanism are demonstrated through simulations and assessed with QoE metrics.

Wireless Multimedia Sensor Networks (WMSNs) play an important role in pervasive and ubiquitous systems. The multimedia content in such networks has the potential of enhancing the level of information collected, enlarging the range of coverage, and enabling multi-view support. For WMSN applications, the multi-tier network architecture has proven to be more beneficial than a single-tier in terms of energy-efficiency, scalability, functionality and reliability. In this context, a multimedia intrusion detection application appears as a promising application of multi-tier WMSNs, where the lower tier can detect the intruder using scalar sensors, and the higher tier camera nodes will be woken up to send real time video sequences from the detected area. The transmission of multimedia content requires a certain quality level from the user perspective, while energy consumption and network overhead should be minimized. Among the existing mechanisms for improving video transmissions, Forward Error Correction (FEC) can be regarded as a suitable solution to improve video quality level from the user point-of-view. In this work, we propose a Quality of Experience (QoE)-aware FEC mechanism for WMSNs, which creates redundant packets based on impact of the frame on the user experience. According to the simulation results, our proposed mechanism achieved similar video quality level compared with standard FEC, while reducing the transmission of redundant packets, which will bring many benefits in a resource-constrained system.