Collective Perception Messages (CPMs), defined by European Telecommunications Standards Institute (ETSI), enable vehicles and roadside infrastructure to exchange information about detected objects, enhancing situational awareness in cooperative environments. However, as the size of CPMs increases — particularly in dense traffic scenarios — the wireless channel can become saturated, leading to delays in transmission and reduced packet delivery ratios. This paper starts by assessing how the number of objects included per CPM impacts communication performance, highlighting the necessity for effective object selection strategies during periods of congestion. To address this issue, we propose a lightweight, real-time object prioritization algorithm based on deviations from the predicted path. Our method estimates each object’s expected state based on its last transmission, and prioritizes those whose current state deviates most from this prediction, as these are likely to be more informative. The evaluation uses a real-world dataset and demonstrates that our strategy significantly improves predictive accuracy by at least 7%. Moreover, the algorithm does not increase CPU or memory usage, demonstrating similar resource consumption compared to the method described in the Collective Perception Service (CPS) standard, making it well-suited for embedded platforms. These results confirm that Prediction–Deviation selection can enhance the efficiency and informativeness of CPMs, especially when the message size must be constrained due to network congestion.

The emergence of 6G networks introduces new challenges, including the planning of ultra-dense networks that can be optimized through Network Digital Twins (NDTs). In smart cities, NDTs are especially valuable, as they integrate network and mobility data to enable adaptive and intelligent management. This work proposes a NDT designed to analyze the impact of urban mobility on 5G networks. The NDT environment integrates real and simulated urban mobility with emulated 5G base stations, emulating real vehicles in the road accessing 5G network services while on the move. The proposed system leverages metric collection for beyond-5G networks with real-time and historical capabilities, to support the envisioned simulations. The system combines real-world data collection with large-scale network emulation using open-source projects, such as SUMO framework, UERANSIM, and Kubernetes to simulate multiple V2X connections between vehicles and 5G base stations. The proposed NDT bridges urban mobility and network behavior, providing a foundation for planning and emulating beyond-5G network management solutions in mobile and dynamic environments. Several scenarios have been tested, such as random, a marathon with road blocks, and lane closure. The results show a high aggregation of vehicles and data traffic in one of the 5G antennas when blocking roundabouts on fast roads in comparison to the random scenario; moreover, the closure of a double entry lane to the roundabout at the location of the 5G gNodeB translates to much lower peak utilization and a better balance of antenna resources during the running scenarios.

The growing scale of smart city sensing devices and infrastructure entails a wide variety of available sensing information that can provide valuable insights into user mobility and traffic congestion. This information can be used to optimize service delivery through the development of mobility-aware services. 5G systems and their associated technologies provide an ideal environment with capabilities to efficiently support edge computing and bring the processing and storage resources closer to the end users, which results in a latency and backhaul usage reduction. This article proposes the integration of edge computing in 5G operator network and a mobility/road-side infrastructure with edge orchestration to provide mobility-aware services to the end-users on demand. With this approach, a service instantiation can be translated into resource allocation both on the 5G platform through multi-slicing and the edge infrastructure. Resource management is then optimized for the users on the move by continuously allocating the necessary virtual network slices, processing, and storage resources in the appropriate locations for the user to consume its services while maintaining the appropriate QoS levels and optimized resource distribution in the edge platform. This approach is evaluated in a real mobile 5G network with emulated Radio Access Network (RAN) resources through two use cases based on infotainment and emergency services. The results show that the approach is efficient in using mobility, service requirements, and platform’s resources information to enable a proactive resource reservation both in the 5G base stations and edge computing nodes throughout the path traversed by the users. • Smart City sensing and mobility information can be used to optimize service delivery. • The integration of mobility, MEC and 5G enable the instantiation of emergency services. • Allocation of the network and computing resources, maintaining the required QoS levels. • Results successfully demonstrate proactive resource reservation in both 5G and MEC.

The growing need for healthier and more ecofriendly city transport has empowered the enforcement of electric and smart bikes, which integrate sensors and communication technologies to improve monitoring and safety. This paper presents a data management architecture designed to enhance the efficiency and adaptability of 2-wheeler monitoring and safety systems. The architecture integrates a data stream service and a persistence system, using fuzzy logic to dynamically prioritize and manage data transmission. The data stream service ensures the delivery of critical messages, such as diagnostics, malfunction and safety alerts, while the persistence system stores data batches and schedules their transmission based on factors such as priority, deadlines, and expiration status. Experimental evaluations show that fuzzy logic-based prioritization significantly reduces message waiting times, outperforming traditional approaches. These results demonstrate the effectiveness of the proposed solution in ensuring reliable and timely data management under varying conditions.

The advancement of smart city services requires innovative approaches to manage urban transportation and enhance tourist mobility through sensor-based IoT data models for efficient processing. This work proposes an integrated tourist recommendation system within a multimodal travel orchestration platform, named Multipres. Designed to support sustainable tourism, Multipres optimizes urban infrastructure, including buses, bicycles, and pedestrian paths, while reducing resource strain. The platform leverages AI-based recommendation engines, integrates IoT mobility sensors, bike-sharing stations and vehicles’ parking platforms, and external applications, such as Waze. Results demonstrate that Multipres provides recommendations according to the users’ needs with updated information through the GenAI integration, in a multimodal approach with real-time data from the city sensing and external platforms, both for mobility and traffic events, and taking into account environmentally indicators.

The growing need for healthier and more ecofriendly city transport has made electric and smart bikes more popular, but this also brings safety concerns to cyclists. This paper proposes an Advanced Driver Assistance System (ADAS) solution for smart bikes to improve cyclist safety, throughout a set of cooperative perception services that communicate using Vehicle to Everything (V2X) to a city network. The services presented collect physical data from the $\mathbf{2}$-wheelers and surrounding environment: detect rear obstacles using Light Detection and Ranging (LiDAR); use computer vision on the front to identify pedestrians and vehicles; and use inertial sensors to detect the road conditions. These services are designed with constraints on power consumption while meeting real-time detection requirements. Experiments in real scenarios show that: (1) the LiDARbased rear detection system achieved an F1-score of 0.98; (2) the road quality assessment system distinguished smooth roads from potholes with an F1-score of 0.75; and (3) the pedestrian detection service achieved $85 \%$ accuracy. This work provides innovative and practical solutions to improve road safety and collaborative perception in smart bicycles.

The rapid expansion of vehicular networks is expected to cause significant congestion in communication channels, potentially compromising the reliability of safety and traffic management systems. To mitigate congestion, the European Telecommunications Standards Institute (ETSI) developed the Decentralized Congestion Control (DCC) mechanism, which dynamically adjusts transmission parameters based on real-time channel conditions. However, real-time DCC data may not always be available due to hardware limitations. This paper proposes a new approach to estimate the Channel Busy Ratio (CBR) in ITS-G5 vehicular networks through passive channel observation. Using real hardware, we compare the proposed estimator with measurements from the wireless card, showing a maximum error below 5.6%. We also analyze the impact of standard vehicular messages and IP traffic on network congestion. The estimator offers accurate, non-intrusive CBR measurements, supporting congestion control where direct measurement is impractical.

Heterogeneous networks integrate diverse devices and technologies, creating significant challenges for efficient traffic forwarding and network management. Smart cities are an example of such networks, relying on applications ranging from transportation to public safety, and demanding adaptive communication infrastructures, terrestrial and non-terrestrial, to ensure seamless connectivity and performance. Simultaneously, Software Defined Networking (SDN) has emerged and proved itself as a valid new approach to network management, enabling intelligent and centralized control. This paper presents an approach, denominated Service-Aware Multi-path BAlancing (SAMBA), to service-aware traffic forwarding to manage heterogeneous, multi-technology, multi-path environments, overcoming the limitations of previous, more technology-specific methods. This approach is evaluated against Shortest Path Forwarding (SPF) across various network technologies, including Ethernet, wireless networks (IEEE 802.11p and private 5G), and Non-Terrestrial Networks. The results demonstrate significant improvements in service throughput and latency while enhancing load distribution and preventing network link saturation.

Augmented Reality (AR) and edge computing can be used to enhance Vulnerable Road User (VRU) safety, leveraging a greater degree of interaction with the user and presenting real-time warning notifications, a use case that demands low latency for critical warnings. However, a new Multi-Access Edge Computing (MEC) system that distributes computational needs across edge nodes within the infrastructure is needed to meet these low-latency requirements. This paper proposes a microservice architecture designed to process data from various sources, including awareness and perception messages from road users and smart city sensors, through multiple communications technologies such as ITS-GS and 5G. This work employs peer-to-peer decentralized communications to perform seamless inte-gration and real-time processing, ensuring timely and relevant warnings for VRUs. Real-world road tests conducted in the Aveiro Tech City Living Lab in Aveiro, Portugal, and in commercial V2X equipment in the Autonomous Driving Test Field Baden-Wurttemberg in Karlsruhe, Germany, showed a maximum end-to-end latency of 110 ms, showcasing the system's capabilities in real-life demanding use cases and the system's interoperability.

The operation of hybrid networks (wireless and wired-based) is subject to a number of significant challenges, particularly in relation to the frequent changes in topology and high mobility that are inherent to them. This is especially the case in Vehicle to Everything (V2X) networks. This paper proposes a Software Defined Network (SDN)-based monitoring platform, designed for improving network visibility and responsiveness in dynamic environments. The proposed approach leverages Cooperative Intelligent Transportation System (C-ITS) messages to enable real-time tracking of vehicular mobility and wireless links quality, which enhances the SDN controller's ability to adapt to rapidly changing network conditions, thereby providing crucial data for informed decision-making. This broader view of the network provides the foundation for future seamless communication devices. The system's efficiency in maintaining up-to-date network awareness, optimizing handover processes, and ensuring communication stability across hybrid networks has been demonstrated in outdoor testing. The results show that incorporating the monitoring and control approaches within SDN environments markedly enhances the management and reliability of dynamic wireless networks.

Multi-Access Edge Computing (MEC) brings computational resources closer to end-users, enabling applications to deliver lower latencies and higher Quality of Service (QoS). However, as users roam across multiple domains, maintaining uninterrupted service and low latencies becomes difficult, especially when individual MEC systems experience resource overload. To address these limitations, MEC Federation (MECF) enables seamless interoperability among distinct domains, or MEC systems, regardless of their underlying specifics by using a common reference point. Subsequently, MECF facilitates application migration and resource sharing. In this paper, we analyze existing works and standards on MECF and propose a complete end-to-end framework for managing MECF between operators. This framework supports application migration, inter-MEC resource sharing, and the orchestration of these actions within individual MEC systems. Our evaluations demonstrate that the proposed framework supports efficient, secure, and scalable MECF operations, supporting the deployment of diverse use cases in multi-operator environments.

This paper proposes an urban and mobility-based Digital Twin that provides the representation of an urban scenario with both real and simulated mobility data of vehicles, 2-wheelers and people. The platform integrates real-time data from the Aveiro Tech City Living Lab (ATCLL) with 2D and 3D visualizations using SUMO and CARLA, respectively, enabling detailed traffic analysis and management. Key functionalities include the synchronization of real-world sensor data with simulated environments, providing accurate and dynamic traffic visualizations. The platform also supports the change and blocking of intersections, roundabouts and lanes, being able to test scenarios when the road conditions change, anticipating the impact of those changes in the urban mobility. The platform results show that they can help decision makers to optimize the traffic flow and anticipate changes in the roads, providing information on the travel times, CO2 emissions and congestion in the roads.

Federated Learning (FL) is a promising approach for parameter normalisation in Machine Learning (ML) models, especially when data privacy and computing distribution are crucial. However, there are significant constraints in FL solutions, particularly concerning the handling of the mobility of participating nodes in the parameter aggregation processes, with a substantial impact on Vehicle to Everything (V2X) scenarios within the scope of smart cities. To address this challenge, we propose Mobile Federated Learning System (MobFedLS), a lightweight microservices-based framework capable of operating on various types of devices (mobile and non-mobile). MobFedLS features an interface to integrate ML models to cooperate in the FL process without intrusion between the parties. MobFedLS manages the entire federation process, from instantiating services on mobile nodes to the final parameter updates in the involved ML models and the release of resources used in all participating nodes. Additionally, MobFedLS handles node mobility and ensures the proper execution of federated processes, even with nodes entering and leaving at any stage of the aggregation process. To demonstrate the capabilities of MobFedLS, we use data collected through the city-scale infrastructure of Aveiro Tech City Living Lab (ATCLL), specifically the position of vehicles during their movement through the city. In the tests, we evaluate all phases of the aggregation process for mobile nodes. The results show that, even with intermittent connectivity to the city-infrastructure ATCLL, the MobFedLS system manages the node mobility and effectively handles node availability during the aggregation of ML model parameters.

Smart-city services are typically developed as closed systems within each city's vertical, communicating and interacting with cloud services while remaining isolated within each provider's domain. With the emergence of 5G private domains and the introduction of new M2M services focusing on autonomous systems, there is a shift from the cloud-based approach to a distributed edge computing paradigm, in acontinuumorchestration. However, an essential component is missing. Current orchestration tools, designed for cloud-based deployments, lack robust workload isolation, fail to meet timing constraints, and are not tailored to the resource-constrained nature of edge devices. Therefore, new orchestration methods are needed to support MEC environments. This paper addresses this gap. We propose an orchestration platform and its algorithms to facilitate the seamless orchestration of both cloud and edge-based services, encompassing both critical and non-critical services. This work extends the current Kubernetes orchestration platform to include a novel location-specific resource definition, a custom scheduler to accommodate real-time and legacy services, continuous service monitoring to detect sub-optimal states, and a refined load balancing mechanism that prioritizes the fastest response times.

Augmented Reality (AR) holds great potential for enhancing pedestrian’s urban experiences; however, its use in road traffic poses safety concerns due to potential distractions from interacting with AR interfaces. This paper investigates the effectiveness of AR applications for assisting pedestrians in crossing scenarios, against traditional crossing methods, by incorporating a collision warning system that uses an arrow to indicate the direction of a potential danger, and a virtual traffic light showing whether it is safe to cross. By leveraging Vehicle-to-Everything (V2X) communications within the living lab of Aveiro, Portugal, we conducted a user study to evaluate involved workloads, perceived safety and system usability in a realistic scenario. The findings from our study involving 20 participants reveal significant improvements in pedestrians’ perceived safety and a decrease in the perceived workload when using AR for pedestrian crossings, with both collision warning systems and virtual traffic lights demonstrating excellent usability.

The process of software updates for future autonomous vehicles on the road can be a critical process since they are constantly moving. Therefore, V2X services are susceptible to occasional connectivity interruptions, which can render traditional Over-the-Air (OTA) updates on the Software Defined Vehicle (SDV) concept impractical. By leveraging Mobile Edge Computing, using its Edge Nodes, with the assistance of the distributed protocol InterPlanetary File System (IPFS) and tools like Ansible, an OTA update system was designed to enable agile updates of V2X services under the Fifth Generation (5G) network and current Software Defined Network (SDN). The system enhances the network’s adaptability and minimizes service disruptions, reducing operational costs, and ensuring timely deployment of software patches and feature upgrades. This paper shows the proposed Vehicle to Everything (V2X) OTA updates solution and the tests performed in a real-world environment, the Aveiro Tech City Living Lab City (ATCLL) in Aveiro, Portugal, and shows how the vehicles gather the content blocks through the distributed edge road side units while the vehicles travel around in the different roads. We believe that such an approach can be set as an exemplary standard for modern Smart cities globally, in regard to the V2X service deployment.

In the last years there has been a strong effort in the development of Intelligent Transportation System (ITS)-based solutions, leading to an important change in the way that drivers and other road users become aware of the surroundings. The development of Cooperative-ITS, which utilises direct wireless short-range connections, is integrating cellular networks as well (4G and 5G), allowing the growing use of the road users smartphones to provide real-time information about Vulnerable Road Users (VRUs), like pedestrians and cyclists. Such increase is becoming a serious concern, since every VRU is likely to have one smartphone, which may lead to scalability and latency issues. This work presents an approach for a microservices-based application, targeting the always critical VRU safety use-case, in a multi-site scenario, using real road infrastructure Multi-Access Edge Computing (MEC) and mobile network provider cloud computing. The main novelty of this work is the multiple approaches on the deployment of the required microservices, and several scenarios that have been tested, and the investigation of the best approach to minimize the service-level latency of the safety application. The results show the potential of microservices distribution through the edge and cloud, with a strong impact on improving the efficiency of ITS. Depending on the services’ location, the latency of the VRU and vehicle’s notification is deeply affected, but using a federated scenario we are able to keep the VRU’s notification delay around the 200 ms, with better results being achieved if a closer mobile network provider cloud platform is used. The results present a noticeable advancement in the development of more scalable and operational solutions that work on improving ITS, with a focus on microservices and edge computing to minimize the delay of critical applications.

The growing use of Internet of Things devices and the increasing demand for high-speed, reliable, and secure wireless connectivity pose significant challenges for existing wireless networking solutions in modern smart homes. As such, there is an increasing urgency for the development of advanced and effective wireless technologies that can fulfill the new requirements for interconnected devices and services deployed in households. Tri-band Wi-Fi 6E equipment addresses these needs by reducing the network congestion and enhancing the performance across the 2.4 GHz, 5 GHz, and 6 GHz bands. However, improper band management can lead to frequent interference and network issues. As such, this work introduces a dynamic Wi-Fi link orchestration solution that follows a heuristic model. This approach aims to optimize the network layout and channel allocation based on device metrics, utilizing the EasyMesh specification for simplified network setup and management. The model was implemented and tested in a residential environment network, using EasyMesh and tri-band Wi-Fi 6E devices. Results show the effectiveness of the model in improving network capacity and adapting the links to the current traffic, outperforming the initial and base network configuration.

The demand for more secure, available, reliable, and fast networks emerges in a more interconnected society. In this context, 5G networks aim to transform how we communicate and interact. However, studies using 5G data are sparse since there are only a few number of publicly available 5G datasets (especially about commercial 5G network metrics with real users). In this work, we analyze the data of a commercial 5G deployment with real users, and propose forecasting techniques to help understand the trends and to manage 5G networks. We propose the creation of a metric to measure the traffic load. We forecast the metric using several machine learning models, and we choose LightGBM as the best approach. We observe that this approach obtains results with a good accuracy, and better than other machine learning approaches, but its performance decreases if the patterns contain unexpected events. Taking advantage of the lower accuracy in the performance, this is used to detect changes in the patterns and manage the network in real-time, supporting network resource elasticity by generating alarms and automating the scaling during these unpredictable fluctuations. Moreover, we introduce mobility data and integrate it with the previously traffic load metric, understanding its correlation and the prediction of 5G metrics through the use of the mobility data. We show again that LightGBM is the best model in predicting both types of 5G handovers, intra- and inter-gNB handovers, using the mobility information through Radars in the several roads, and lanes, near the 5G cells. • Development of a data exploration pipeline for 5G network metrics and mobility data. • 5G network data anonymization using PCA. • Traffic load metric research. • Machine Learning for network traffic prediction. • Correlation of traffic load and mobility data.

Over the years, WiFi became an essential technology. The success of the introduction of wireless devices with WiFi connectivity increased the demand for better WiFi networks. Such networks need better service and better coverage, either in mobile or residential environments. To solve this challenge, WiFi Alliance developed WiFi EasyMesh, a standard for WiFi networks that uses multiple access points that allow an easy setup and compatibility with WiFi certified devices. This work studies the performance of a mesh wireless network with different frequencies in use by the backhaul links (2.4 GHz and 5 GHz). The results can then be used to derive better backhaul steering algorithms to a better Quality-of-Service.

With the evolution of communication and sensing devices, in addition to connected and autonomous vehicles, real-time information about the status of the roads can become a reality. This paper presents a proposal of a Collective Perception Service (CPS), a system defined by the European Telecommunications Standards Institute (ETSI), which aims at enabling Intelligent Transport Systems (ITSs) to cooperate and exchange information about road users, obstacles on the road, and sensing information, through Collective Perception Messages (CPMs). The proposed system is able to join the information of the communication messages and the data from the different sensors, such as LiDAR, radars, and video cameras. The CPS is evaluated and tested in real-world scenarios in a connected city infrastructure. Results show that CPS allows Intelligent Transport System Stations (ITS-Ss) to cooperate and exchange information regarding perceived objects in a timely manner. A temporal analysis outlined the influence of the message size on the CPM dissemination process and the need for a mechanism that decides if specific information shall be sent in the CPM. Finally, the scalability tests showed the effect of network congestion, testing the system’s limits in the selected hardware.

Mobile networks have proved to be an attractive solution to support emergency services, due to their inherent mobility, which requires flexible and yet efficient communication. However, the management of resources in slices in a 5G network is still made static and with no mobility information and prediction. This paper proposes the integration of a 5G network platform and a city mobility network, which allows 5G network resources to be allocated dynamically and in advance according to the users mobility and their needs. This mobility and network integration is important for different types of services, from infotainment to emergency. The prediction of the mobility and location through the mobility network provides the knowledge for a 5G cellular network to dynamically reserve the required network slicing resources. This approach has been tested in real road scenarios and vehicles covered by 5G. The results show that the solution guarantees the required resources, reserved in a proactive and predictive approach, both to optimize the resources of one base station, and to optimize resources between base stations, offering the required quality to the services.

Vehicle-to-everything represent a major step in the evolution of Intelligent Transportation Systems (ITS), by allowing vehicles and the infrastructure to share information seamlessly. The new sensors introduced by the next generation of vehicles, that can be autonomous or partially assisted, generate huge volumes of data that, in some scenarios, may not be handled by the available resource-constrained devices. Therefore, edge and fog concepts have been applied to vehicular networks under Vehicle Edge Computing (VEC), to perform computational offloading, content caching, data management, flexible network management, security, and others. Technologies like virtualization and containerization are the cornerstones when addressing service deployment on edge and cloud environments. This work assesses different schemes of VEC service deployments to understand their use with the deterministic and reliable nature of some critical VEC services, such as an open-source implementation of the ETSI C-ITS protocol suite. Service prioritization and preemption are explored in order to achieve an upper-bound latency in native and containerized services deployment. The results show that the preemption leads to a slight increase in the average and median processing delay, but it also results in positive effects for real-time services; the prioritization of services is able to provide lower upper-bound latency, providing a performance without guarantees but with an expected delay.

The Federated Learning (FL) paradigm aims to provide performance advantages over centralized models, such as lower latency and communication overhead when doing most of the processing on the edge devices, better privacy as data does not travel over the network, easier handling in heterogeneous data sources and better scalability. However, the development of FL-based solutions is done through tools aimed for specialists as it always requires some programming. To cover this gap, we present an architecture for a lightweight container-based solution that offers a range of machine learning (ML) algorithms to build prediction engines for edge devices, which also includes the main options in algorithms/models for aggregation and refinement of models in the central server. The proposed framework allows the rapid build of containerized testbeds for the evaluation of ML and aggregation algorithms in the initial evaluation phase, and also later in the installation in real production infrastructures. We demonstrate the efficiency of our approach in estimating vehicle mobility in and out of the city of Aveiro, using real data collected by the communications and sensing infrastructure.

Over time, mobile communication networks have seen significant evolution. In the 5G generation, the rollout of the network requires network densification, due to the new data-hungry applications and an exponential number of new devices. The idea of a smart city, a technologically advanced metropolitan area that gathers data to enhance the general quality of life of its residents, also contributes to the densification topic. To provide content for this increased number of end-users, Content Delivery Networks (CDNs) appear as a promising solution: by distributing the content through multiple end-nodes that are near the user, they are capable to deliver content with low latency, over intermittent connections, leading to a reduction in the use of the network backhaul. This work focuses on the architecture for 5G networks with Multi-Access Edge Computing (MEC) to be used in smart cities. By using User Plane Function (UPF) selection, the Control User Plane Separation (CUPS) concept has enabled the disaggregation of the Core Network (CN). Thus, a CDN was deployed in the private 5G network using the InterPlanetary File System (IPFS) protocol, to evaluate the proposed approach. The demo presented in this paper shows the integration of IPFS intelligence with the CN, more specifically in the desirable UPF, and it also approaches the idea of an Application Function (AF).

Mobile communication networks evolved staggeringly throughout the years. Current evolution (5G) needs a denser network, due to new vertical-based data-hungry applications and the increase of UEs, leading to additional costs in cellular deployment. To solve this issue and optimize the deployment of 5G, one unified network could be devised and shared by multiple operators to deploy their own networks. With a city network planned, ways to take full advantage of a neutral hosting architecture begin to be executed, which serve various services, like Vehicle-to-Everything (V2X) and Internet-of-Things (IoT). This is where topics such as Multi-Access Edge Computing (MEC) arise, by maximizing the exploitation of the city infrastructure to reduce latency and remove the computational effort from cloud servers to the edge. This thesis focuses on developing a neutral architecture for 5G networks, with Multi-acess Edge Computing (MEC), which can be deployed in a city. In the concrete case of this thesis, this architecture will be deployed in the city of Aveiro, taking advantage of its multiple edge nodes. The disaggregation of the Core Network (CN) is a focal point of this thesis, creating a way to have a gateway closer to the edge, thus enabling MEC technologies to improve end-user and service experience. A disaggregated architecture was proposed, and extended to the city infrastructure, to accurately simulate the city environment and to attest to the city services. The test scenarios involved the comparison of a service instantiated in the cloud and in the edge, mobility scenarios with a handover through the same network to attest user plane relocation, different user plane selection, and lastly, different flow level priority assessment, to higher priority services in a full bandwidth occupation scenario. The obtained results show that the deployment of the user plane in the edge brings significant improvements, both in common user traffic and service-oriented traffic; moreover, this outlines the capabilities of this core solution as well as its limitations providing a foundation for future works.

WiGig technologies, such as IEEE 802.11ad and later IEEE 802.11ay, provide multi-gigabit short-range communication at 60 GHz for bandwidth-intensive applications. However, this band suffers from high propagation losses that can only be compensated using highly directional antennas, making millimeter-wave (mmWave) links susceptible to blockage and errors. This high sensitivity to blockage leads to unstable and unreliable connections, since proprietary IEEE 802.11ad mechanisms, such as beamforming training, have high overhead, and can only be triggered when performance degradation is already detected, which compromises QoS and QoE even more. This article proposes a proactive machine learning framework that uses real-life data acquired in an outdoor setting to improve the reliability and resilience of a blockage-prone WiGig-based network. In particular, we propose a link quality classifier, which can differentiate between normal, long-term blockage and short-term operation with a test F1-score of 97%. Moreover, we introduce a novel deep learning forecasting model that can accurately capture the interactions between past multi-layer observations under different environments to produce accurate forecasts for 16 KPIs.

This article presents the deployment and experimentation architecture of the Aveiro Tech City Living Lab (ATCLL) in Aveiro, Portugal. This platform comprises a large number of Internet of Things (IoT) devices with communication, sensing, and computing capabilities. The communication infrastructure, built on fiber and millimeter-wave (mmWave) links, integrates a communication network with radio terminals [WiFi, ITS-G5, cellular vehicular-to-everything, 5G and LoRa(WAN)], multiprotocol, spread throughout 44 connected points of access in the city. Additionally, public transportation has also been equipped with communication and sensing units. All these points combine and interconnect a set of sensors, such as mobility (radars, light detection and rangings (LiDARs), and video cameras) and environmental sensors. Combining edge computing and cloud management to deploy the services and manage the platform, and a data platform to gather and process the data, the living lab supports a wide range of services and applications: IoT, intelligent transport systems (ITSs) and assisted driving, environmental monitoring, emergency and safety, and among others. This article describes the architecture, implementation, and deployment to make the overall platform to work and integrate researchers and citizens. Moreover, it showcases some examples of the performance metrics achieved in the city infrastructure, the data that can be collected, visualized, and used to build services and applications to the cities, and, finally, different use cases in the mobility and safety scenarios.

The concept of a smart city comes with the need to support a data platform that can gather, process and export the data of millions of sensors, coming from different sources, with information in different formats, in a scalable approach for realtime and historical data visualization, processing, and actuation in the city. This paper proposes a Data Platform for the Aveiro Tech City Living Lab, to gather, process, visualize and actuate on mobility, environmental and network data. The architecture of the platform provides a real open platform that is accessible for third-parties to collect data and to experiment their own solutions, through a secure and open data platform at their disposal. The results with respect to the amount of data gathered and examples of data show how this platform can be used to develop new applications and use both real-time and historical data for future predictions and actuations in a smart city.