There is an urgent need to perform effective feature selection in distributed environments while preserving data privacy. In this paper, a new federated feature selection framework is developed to protect the privacy of input features held by multiple distributed clients, with applications in engineering systems where secure and efficient feature selection is critical in distributed environments. The proposed framework is based on federated learning and differential privacy techniques for distributed environments. The distributed clients send the noisy features’ values to the server preserving the privacy. The server then aggregates these noisy features’ values for further computations and feature selection. The performance of the proposed framework is evaluated against a centralized scenario where feature selection occurs centrally. Comparative analysis involves inputting the selected features into various machine learning models employing various evaluation metrics. The simulation results indicate comparable performance between the proposed federated approach and the centralized method. To further compare performance, a new method, ’Rank of Features’ is developed in this paper that evaluates similarity between features selected by the proposed framework and centralized method. The results of this analysis also demonstrate strong similarity between the two approaches. Further, privacy analyses are conducted in detail that include protection against reconstruction and membership inference attacks demonstrating robust preservation against data leakage and unauthorized inference of sensitive information.

Energy is a fundamental component of modern life, driving nearly all aspects of daily activities. As such, the inability to access energy when needed is a significant issue that requires innovative solutions. In this paper, we propose ED-DAO, a novel fully transparent and community-driven decentralized autonomous organization (DAO) designed to facilitate energy donations. We analyze the energy donation process by exploring various approaches and categorizing them based on both the source of donated energy and funding origins. We propose a novel Hybrid Energy Donation (HED) algorithm, which enables contributions from both external and internal donors. External donations are payments sourced from entities such as charities and organizations, where energy is sourced from the utility grid and prosumers. Internal donations, on the other hand, come from peer contributors with surplus energy. HED prioritizes donations in the following sequence: peer-sourced energy (P2D), utility-grid-sourced energy (UG2D), and direct energy donations by peers (P2PD). By merging these donation approaches, the HED algorithm increases the volume of donated energy, providing a more effective means to address energy poverty. Experiments were conducted on a dataset to evaluate the effectiveness of the proposed method. The results showed that HED increased the total donated energy by at least 0.43% (64 megawatts) compared to the other algorithms (UG2D, P2D, and P2PD).

Carbon footprint reduction can be achieved through various methods, including the adoption of renewable energy sources. The installation of such sources, like photovoltaic panels, while environmentally beneficial, is cost-prohibitive for many. Those lacking photovoltaic solutions typically resort to purchasing energy from utility grids that often rely on fossil fuels. Moreover, when users produce their own energy, they may generate excess that goes unused, leading to inefficiencies. To address these challenges, this paper proposes innovative blockchain-enabled energy-sharing algorithms that allow consumers -without financial means- to access energy through the use of their own energy storage units. We explore two sharing models: a centralized method and a peer- to- peer (P2P) one. Our analysis reveals that the P2P model is more effective, enhancing the sharing process significantly compared to the centralized method. We also demonstrate that, when contrasted with traditional battery-supported trading algorithm, the P2P sharing algorithm substantially reduces wasted energy and energy purchases from the grid by 73.6%, and 12.3% respectively. The proposed system utilizes smart contracts to decentralize its structure, address the single point of failure concern, improve overall system transparency, and facilitate peer-to-peer payments.

Federated learning (FL) is a promising approach for processing on-board data in vehicular networks due to its distributed nature and its ability to accurately and efficiently handle the large amount of sensed data. However, training and transmitting the model parameters during FL process can consume a significant amount of energy and time, which is not suitable for applications with strict real-time requirements. Moreover, the dynamicity of the vehicular network, as well as the varying capabilities of each vehicle, can impact the performance of the training process, bringing to the forefront the optimization of the participants selection and their resources. In this paper, we propose VOC-FL, a Vehicular-based Offloading and Clustering framework supported by FL. The proposed scheme bypasses communication bottlenecks by enabling groups of vehicles to train models simultaneously, with only the Cluster Head (CH) sending the aggregated results of each cluster to the roadside units for further processing. To form the clusters, we select a CH for each cluster based on multiple metrics, including stability, computational resources, bandwidth, and network topology. Moreover, the CH runs an offloading strategy that allows struggling nodes with limited computational resources to offload their tasks to other nodes with enough resources within the cluster, enabling efficient and effective use of resources.

With an innovative door opened for a new era of Machine Learning, Federated Learning (FL) is now revolutionizing Artificial Intelligence. It exploits both decentralized data and decentralized computation to preserve user privacy. Albeit its popularity and being the most widely used framework nowadays, FL becomes a sub-optimal solution when the convergence of the global model occurs at a slow pace, which exacerbates the communication bottlenecks. To address this challenge, we propose in this paper CISCO-FL, a Clustered FL with Intelligent Selection and Computation Offloading. First, we partition the clients into different groups, where sub-aggregations of the clients models are performed at each cluster before the global aggregation. Second, we study the computing resources of the clients, and we embed in the proposed approach an intelligent offloading model, where the clients with high computational resources can assist and optimize the model of those struggling with limited resources. As such, both communication cost and computation resources are reduced and optimized. Finally, thorough experimental results are presented to support our findings and validate our model.

Internet of Things (IoT), Digital Twin (DT), and Federated Learning (FL) are redefining the future vision of globalization. While IoT is about sensing data from physical devices, DTs reflect their digital representation and enable optimized decision-making by tightly integrating Artificial Intelligence (AI). Although swiftly growing, DTs are raising new challenges in privacy concerns, which are nowadays addressed by FL. However, the limited IoT resources, the communication overhead, and the lack of trust among clients are major obstacles that hinder the effectiveness of learning systems. In this paper, we design a new IoT-based architecture empowered by DT to improve the efficiencies of limited-resources devices. On top of this architecture, we leverage FL to construct the DT models. We further propose CISCO-FL, a Clustered FL with Intelligent Selection and Computation Offloading. Particularly, we study the computing resources of the clients and the quality of their models, and we embed in the proposed approach an intelligent offloading model, where the clients with high computational resources can assist and optimize the model of those struggling with limited resources. As such, both communication cost and computation resources are reduced and optimized. Finally, thorough experimental results are presented to support our findings and validate our model.

Intelligent Transportation Systems (ITS), mainly Autonomous Vehicles (AV's), are susceptible to security and safety problems that risk the users' lives. Sophisticated threats can damage the security of AV's communications and computational capabilities, slowing down their integration into our daily lives. Cyber-attacks are getting more complex, posing greater hurdles in identifying intrusions effectively. Failing to prevent the intrusions could tarnish the security services' reliability, including data confidentiality, authenticity, and reliability. IDS is an overall prediction paradigm for detecting malicious network traffic in the ITS. This article studies the role of machine or deep learning in Software Defined-Intrusion Detection System (SD-IDS) in ITS; discusses the mathematical analysis of existing deep learning models and evaluates their performances on the basis of the various metrics (i.e., accuracy, precision, recall, f-measure) to observe which model gives the best results for the existing state of art. The results show that improved Recurrent Neural Networks (RNN) is best suited for the detection of SD-IDS attacks in the data plane and control plane.

Abstract Growing intelligent cities is witnessing an increasing amount of local energy generation through renewable energy resources. Energy trade among the local energy generators (aka prosumers) and consumers can reduce the energy consumption cost and also reduce the dependency on conventional energy resources, not to mention the environmental, economic, and societal benefits. However, these local energy sources might not be enough to fulfill energy consumption demands. A hybrid approach, where consumers can buy energy from both prosumers (that generate energy) and also from prosumer of other locations, is essential. A centralized system can be used to manage this energy trading that faces several security issues and increase centralized development cost. In this paper, a hybrid energy trading system coupled with a smart contract named SynergyGrids has been proposed as a solution, that reduces the average cost of energy and load over the utility grids. To the best of our knowledge, this work is the first attempt to create a hybrid energy trading platform over the smart contract for energy demand prediction. An hourly energy data set has been utilized for testing and validation purposes. The trading system shows 17.8% decrease in energy cost for consumers and 76.4% decrease in load over utility grids when compared with its counterparts.

Advanced services leveraged for future smart cities have played a significant role in the advancement of 5G networks towards the 6G vision. Interactive immersive applications are an example of those enabled services. Such applications allow for the interaction between multiple users in a 3D environment created by virtual presentations of real objects and participants using various technologies such as Virtual Reality (VR), Augmented Reality (AR), Extended Reality (XR), Digital Twin (DT) and holography. These applications require advanced computing models which allow for the processing of massive gathered amounts of data. Motions, gestures and object modification should be captured, added to the virtual environment, and shared with all the participants. Relying only on the cloud to process this data can cause significant delays. Therefore, a hybrid cloud/edge architecturewith an intelligent resource orchestration mechanism, that is able to allocate the available capacities efficiently is necessary. In this paper, a blockchain and federated learning-enabled predicted edge-resource allocation (FLP-RA) algorithm is introduced to manage the allocation of computing resources in B5G networks. It allows for smart edge nodes to train their local data and share it with other nodes to create a global estimation of future network loads. As such, nodes are able to make accurate decisions to distribute the available resources to provide the lowest computing delay.

Advancements in communication technologies such as Beyond 5G have made it possible for on-demand services in vehicular networks. Novel advancements in the transportation area are needed to meet these feature requirements. Most of the communications in vehicular networks are based on names of content and thus caching inside the network makes Named Data Networking (NDN) a suitable option for content dissemination for connected vehicles. To achieve NDN maximum benefits, a new effective data forwarding scheme is needed as the size of the cache is limited to store all of the requested content, and cache replacement scheme to replace the data which is not required. In this paper, the challenge of in-network caching is addressed in terms of efficient cache utilization by considering the cache hit ratio and the challenge of vehicles' mobility. The volume of data cached during communication is called cache utilization. Many in-network caching strategies are proposed in the literature, but very few of the proposed schemes consider the efficient utilization of cache resources while considering the mobility of vehicles. Default cache replacement technique is Least Recently Used (LRU) whereas Leave Copy Everywhere (LCE), Probabilistic Caching (PC), Always Cache (AC), Leave Copy Down (LCD), Reactive Caching (RC), and Edge Caching (EC) are used as data forwarding schemes. In this paper, existing in-network caching strategies with important role in cache utilization such as data forwarding schemes, and cache replacement schemes, are implemented, evaluated and a new caching scheme is proposed which will address the issues faced by current schemes.

The vast distribution of low-power devices in IoT applications requires robust communication technologies that ensure high-performance level in terms of QoS, light-weight computation, and security. Advanced wireless technologies (i.e. 5G and 6G) are playing an increasing role in facilitating the deployment of IoT applications. To prolong the network lifetime, energy harvesting is an essential technology in wireless networks. Nevertheless, maintaining energy sustainability is difficult when considering high QoS requirements in IoT. Therefore, an energy management technique that ensures energy efficiency and meets QoS is needed. Energy efficiency in duty cycling solutions needs novel energy management techniques to address these challenges and achieve a trade-off between energy efficiency and delay. Predictive models (i.e., based on AI and ML techniques) represent useful tools that encapsulate the stochastic nature of harvested energy in duty cycle scheduling. The conventional predictive model relies on environmental parameters to estimate the harvested energy. Instead, Artificial Intelligence (AI) allows for recursive prediction models that rely on past behavior of harvested and consumed energy. This is useful to achieve better precision in energy estimation and extend the limit beyond predictive models directed solely for energy sources that exhibit periodic behavior. In this paper, we explore the usage of a ML model to enhance the performance of duty cycle scheduling. The aim is to improve the QoS performance of the proposed solution. To assess the performance of the proposed model, it was simulated using the INET framework of the OMNet++ simulation environment. The results are compared to an enhanced IEEE 802.15.4 MAC protocol from the literature. The results of the comparative study show clear superiority of the proposed AI-based protocol that testified to better use of energy estimation for better management of the duty cycling at the MAC sublayer.

Industrial investments into distributed energy resource technologies are increasing and playing a pivotal role in the global transactive energy, as part of a wider drive to provide a clean and stable source of energy. The management of prosumers, which consume and as well as generate energy, with heterogeneous energy sources is critical for sustainable and efficient energy trading procedures. This article proposes a blockchain-assisted adaptive model, namely SynergyChain, for improving the scalability and decentralization of the prosumer grouping mechanism in the context of peer-to-peer energy trading. Smart contracts are used for storing the transaction information and for the creation of the prosumer groups. SynergyChain integrates a reinforcement learning module to further improve the overall system performance and profitability by creating a self-adaptive grouping technique. The proposed SynergyChain is developed using Python and Solidity and has been tested using Ethereum test nets. The comprehensive analysis using the hourly energy consumption dataset shows a 39.7% improvement in the performance and scalability of the system as compared to the centralized systems. The evaluation results confirm that SynergyChain can reduce the request completion time along with an 18.3% improvement in the overall profitability of the system as compared to its counterparts.

Network Function Virtualization is known for its ability to reduce deployment costs and improve the flexibility and scalability of network functions. Due to processing capacity limitation, the infrastructure provider needs to instantiate one or more instances of a particular network function when the amount of traffic increases. Most of network functions are stateful, which means that they keep a state that may be frequently read or updated (e.g., statistics like number of packets or bytes per flow). As a result, the instances of the same virtual network function should constantly share the same state to prevent incorrect operation. In this context, a major challenge is how to efficiently ensure the consistency among instances while minimizing communication cost for synchronizing their state and ensuring the synchronization delay does not exceed a certain bound set by the operator.In this paper, we propose a technique to identify the optimal communication pattern between the instances of the same network function in order to minimize their synchronization cost. Moreover, we propose to use a special network function named Synchronization Function to ensure consistency among a set of instances and to minimize the synchronization cost. We first mathematically model the problem of finding the optimal synchronization pattern and the optimal placement and number of synchronization functions as an integer linear program that minimizes the synchronization cost and ensures a bounded synchronization delay. Last, we put forward three algorithms to cope with large-scale scenarios of the problem. Extensive simulations show that the proposed algorithms efficiently find near-optimal solutions with minimal computation time.

Software Defined Networking (SDN) has eased the management and control of networks through separation of the control and data planes. Software defined data centers (SDDC) automate the management of end systems which are physical machines and virtual machines. In data centers, although there is a vast work on minimizing power consumption of physical machines and virtual machine migration performance, energy efficiency of the network components is given little attention. In this paper, a software-based energy efficiency framework that jointly minimizes the power consumption of end systems and network components in SDDC is proposed. Moreover, a novel physical server utility interval based metric, namely Ratio for Energy Saving of Physical Machines (RESPM) which measures how energy efficient the physical servers with respect to virtual machines residing within is proposed. To jointly maximize network energy efficiency and RESPM values, an Integer Programming (IP) formulation has been introduced. Experiments conducted on real-world virtual migration traces show that the proposed framework jointly reduces the power consumption of end systems and network components. The system has shown an improvement of 9% in RESPM, 35% energy saving in Ratio of Energy Saving in SDN (RESDN), and more than 50% in links saving.

Connected and Autonomous Electric Vehicles (CAEVs) are becoming a feature of our roads in the imminent future. This disruptive technology is likely to enhance the way we get around the city in many ways by collecting accurate data in regards to the surrounding environment and events in a timely-manner. As such data that is time-sensitive where human life may be at risk requires reliable and on-time data delivery. In crowded dense environments, several issues can reduce the network performance due to the high density of objects such as skyscrapers and vehicles as well as the large number of exchanged data between connected vehicles and other objects on the road. Involving a swarm of autonomous Unmanned Aerial Vehicles (UAVs), namely drones, would enhance network connectivity, reduce CAEVs communication delay, facilitate CAEVs tasks distribution, and elevate provisioning services. In this paper, a routing scheme in an autonomous UAV-connected vehicles network is proposed that provides reduced communication delay. The proposed solution has been evaluated using simulations. The collected results show the feasibility and the advantage of the UAV-assisted connected vehicle network in meeting delay and energy consumption requirements.

Energy trading systems have revolutionized by taking advantage of energy users who produce surplusenergy. In the cyberphysical energy sharing systems, the participation of such consumers who can also sell their residuum energy for profit, namely prosumers, is critical for the sustainable and efficient energy sharing procedure and requires improved prosumer management. The idea of grouping the prosumers for better profits is a promising approach for prosumer management which is currently carried out in centralized manner; that face trust, security and scalability issues. Hence, a strong tool that can protect the prosumer privacy; log the changes for audit purposes and eventually improve the performance of the system is necessary. This paper proposes a blockchain-assisted approach using smart contracts for improved scalability and decentralization of the prosumer grouping mechanism in the context of P2P energy trading. The results show around 38.7% improvement in the performance and scalability of the system.

Mobile cellular users traveling in city buses are experiencing poor quality of signals due to the interference and the large number of mobile devices. To enhance the Quality-of-Service (QoS), deployment of small cell networks in city buses is a promising solution. The deployment of small cells in vehicular environment makes the resource allocation more challenging because of the dynamic interference relationships experienced by them. Therefore, resource allocation in vehicular environment within moving small cells (MSCs) needs to be handled carefully. In this study, we investigate the problem of resource allocation in city bus transit system with multiple routes. Then, we propose a Percentage Threshold Interference Graph (PTIG) based allocation of resources to MSCs in a network. City buses of multiple routes travel with variable speed and may share some of the same road segments which make it difficult to extract the exact interference patterns between them. Therefore, Long Short Term Memory (LSTM) neural networks are used to predict the city buses locations. The predicted locations of city buses are then used to generate PTIG by finding the dynamic interference relationship between MSCs. Graph coloring algorithm is used to allocate the resources to PTIG. Numerical results are presented to show the comparison of resource allocation using PTIG and Time Interval based Interference Graph (TIIG) in terms of resource block utilization and time complexity.

Network lifetime is the main issue of wireless sensor networks and IoT solutions in real world application. Sensors cannot have an infinite lifetime without battery recharge or replacement. Energy harvesting, from environmental energy sources, is a promising technology to provide sustainable powering for WSN. However, based on harvesting opportunities, nodes power may alternate between two states: a state with sufficient residual power and another with shortage in power. Hence, it is paramount to develop robust networking platforms that are energy-harvesting-aware and that support low-energy consumption and data integrity in a noisy, variable environment. In this paper, we present the EAMP-AIDC protocol, an energy aware MAC protocol for EH-WSN based on individual duty cycle optimization. It takes into consideration nodes' residual energy and application and data requirements in order to define individual dynamic duty cycles (Active and sleep periods) that allow to create a balanced load in term of cooperative data relaying tasks and in terms of energy consumption between the different participating nodes so as to ensure continuous network operation. The proposed protocol was evaluated using the network simulator OMNET++ and was compared to the standard IEEE 802.15.4 MAC. The results showed that EAMP-AIDC protocol outperformed the IEEE 802.15.4 standard in term of better energy consumption, increased survivability in energy savings and in guaranteeing continuous operations.