The rise of 5G networks has been driven by the increasing deployment of Internet of Things (IoT) devices and the expansion of mobile and fixed broadband subscriptions. This has been coupled with a rise in network-related attacks, driven by the expanding attack surfaces. Machine learning (ML) has emerged as a promising solution for detecting security threats in 5G-enabled networks and environments due to its ability to handle the vast amount of data generated. Two ML model types have shown great promise, namely boosting-based and Generative Adversarial Network (GAN) based models. Accordingly, this work proposed a comparative analysis of boosting-based and GAN-based ML models for intrusion detection in softwarized 5G networks. Experimental results using the 5G-NIDD dataset show that both boosting-based and GAN-based models have a high detection capability, are not significantly impacted by feature selection, and have reduced training and prediction times.

Modern vehicles, including connected vehicles and autonomous vehicles, nowadays involve many electronic control units connected through intravehicle networks (IVNs) to implement various functionalities and perform actions. Modern vehicles are also connected to external networks through vehicle-to-everything technologies, enabling their communications with other vehicles, infrastructures, and smart devices. However, the improving functionality and connectivity of modern vehicles also increase their vulnerabilities to cyber-attacks targeting both intravehicle and external networks due to the large attack surfaces. To secure vehicular networks, many researchers have focused on developing intrusion detection systems (IDSs) that capitalize on machine learning methods to detect malicious cyber-attacks. In this article, the vulnerabilities of intravehicle and external networks are discussed, and a multitiered hybrid IDS that incorporates a signature-based IDS and an anomaly-based IDS is proposed to detect both known and unknown attacks on vehicular networks. Experimental results illustrate that the proposed system can detect various types of known attacks with 99.99% accuracy on the CAN-intrusion-dataset representing the IVN data and 99.88% accuracy on the CICIDS2017 data set illustrating the external vehicular network data. For the zero-day attack detection, the proposed system achieves high F1-scores of 0.963 and 0.800 on the above two data sets, respectively. The average processing time of each data packet on a vehicle-level machine is less than 0.6 ms, which shows the feasibility of implementing the proposed system in real-time vehicle systems. This emphasizes the effectiveness and efficiency of the proposed IDS.

Content delivery networks (CDNs) provide efficient content distribution over the Internet. CDNs improve the connectivity and efficiency of global communications, but their caching mechanisms may be breached by cyber-attackers. Among the security mechanisms, effective anomaly detection forms an important part of CDN security enhancement. In this work, we propose a multi-perspective unsupervised learning framework for anomaly detection in CDNs. In the proposed framework, a multi-perspective feature engineering approach, an optimized unsupervised anomaly detection model that utilizes an isolation forest and a Gaussian mixture model, and a multi-perspective validation method, are developed to detect abnormal behaviors in CDNs mainly from the client Internet Protocol (IP) and node perspectives, therefore to identify the denial of service (DoS) and cache pollution attack (CPA) patterns. Experimental results are presented based on the analytics of eight days of real-world CDN log data provided by a major CDN operator. Through experiments, the abnormal contents, compromised nodes, malicious IPs, as well as their corresponding attack types, are identified effectively by the proposed framework and validated by multiple cybersecurity experts. This shows the effectiveness of the proposed method when applied to real-world CDN data.

The current trend in end-user device’s advancements in computing and communication capabilities makes edge computing an attractive solution to pave the way for the coveted ultra-low latency services. The success of the edge computing networking paradigm depends on the proper orchestration of the edge servers. Several Edge applications and services are intolerant to latency, especially in 5G and beyond networks, such as intelligent video surveillance, E-health, Internet of Vehicles, and augmented reality applications. The edge devices underwent rapid growth in both capabilities and size to cope with the service demands. Orchestrating it on the cloud was a prominent trend during the past decade. However, the increasing number of edge devices poses a significant burden on the orchestration delay. In addition to the growth in edge devices, the high mobility of users renders traditional orchestration schemes impractical for contemporary edge networks. Proper segmentation of the edge space becomes necessary to adapt these schemes to address these challenges. In this paper, we introduce a segmentation technique employing lax clustering and segregated mobility-based clustering. We then apply latency mapping to these clusters. The proposed scheme’s main objective is to create subspaces (segments) that enable light and efficient edge orchestration by-reducing the processing time and the core cloud communication overhead. A bench-marking simulation is conducted with the results showing decreased mobility-related failures and reduced orchestration delay.

Vehicle-to-everything (V2X) communication and services have been garnering significant interest from different stakeholders as part of future intelligent transportation systems (ITSs). This is due to the many benefits they offer. However, many of these services have stringent performance requirements, particularly in terms of the delay/latency. Multi-access/mobile edge computing (MEC) has been proposed as a potential solution for such services by bringing them closer to vehicles. Yet, this introduces a new set of challenges such as where to place these V2X services, especially given the limit computation resources available at edge nodes. To that end, this work formulates the problem of optimal V2X service placement (OVSP) in a hybrid core/edge environment as a binary integer linear programming problem. To the best of our knowledge, no previous work considered the V2X service placement problem while taking into consideration the computational resource availability at the nodes. Moreover, a low-complexity greedy-based heuristic algorithm named “Greedy V2X Service Placement Algorithm” (G-VSPA) was developed to solve this problem. Simulation results show that the OVSP model successfully guarantees and maintains the QoS requirements of all the different V2X services. Additionally, it is observed that the proposed G-VSPA algorithm achieves close to optimal performance while having lower complexity.

Cyber-security garnered significant attention due to the increased dependency of individuals and organizations on the Internet and their concern about the security and privacy of their online activities. Several previous machine learning (ML)-based network intrusion detection systems (NIDSs) have been developed to protect against malicious online behavior. This paper proposes a novel multi-stage optimized ML-based NIDS framework that reduces computational complexity while maintaining its detection performance. This work studies the impact of oversampling techniques on the models’ training sample size and determines the minimal suitable training sample size. Furthermore, it compares between two feature selection techniques, information gain and correlation-based, and explores their effect on detection performance and time complexity. Moreover, different ML hyper-parameter optimization techniques are investigated to enhance the NIDS’s performance. The performance of the proposed framework is evaluated using two recent intrusion detection datasets, the CICIDS 2017 and the UNSW-NB 2015 datasets. Experimental results show that the proposed model significantly reduces the required training sample size (up to 74%) and feature set size (up to 50%). Moreover, the model performance is enhanced with hyper-parameter optimization with detection accuracies over 99% for both datasets, outperforming recent literature works by 1-2% higher accuracy and 1-2% lower false alarm rate.

Deploying V2X services has become a challenging task. This is mainly due to the fact that such services have strict latency requirements. To meet these requirements, one potential solution is adopting mobile edge computing (MEC). However, this presents new challenges including how to find a cost efficient placement that meets other requirements such as latency. In this work, the problem of cost-optimal V2X service placement (CO-VSP) in a distributed cloud/edge environment is formulated. Additionally, a cost-focused delay-aware V2X service placement (DA-VSP) heuristic algorithm is proposed. Simulation results show that both CO-VSP model and DA-VSP algorithm guarantee the QoS requirements of all such services and illustrates the trade-off between latency and deployment cost.

The use of autonomous vehicles (AVs) is a promising technology in Intelligent Transportation Systems (ITSs) to improve safety and driving efficiency. Vehicle-to-everything (V2X) technology enables communication among vehicles and other infrastructures. However, AVs and Internet of Vehicles (IoV) are vulnerable to different types of cyber-attacks such as denial of service, spoofing, and sniffing attacks. In this paper, an intelligent intrusion detection system (IDS) is proposed based on tree-structure machine learning models. The results from the implementation of the proposed intrusion detection system on standard data sets indicate that the system has the ability to identify various cyber-attacks in the AV networks. Furthermore, the proposed ensemble learning and feature selection approaches enable the proposed system to achieve high detection rate and low computational cost simultaneously.

Security has become of paramount importance in recent times, especially due to the advent of cloud computing and Internet of Things. With so many devices in the mix, users have the choice of working from anywhere they want. But it also raises the possibility of being able to multiply the impact of any attack by using all devices at hand. Another important aspect to consider is the prevention of access to sensitive data by unauthorized users using authorized machines. Software Defined Perimeter (SDP) provides one such solution. It aims to only allow traffic from authorized users and machines to a hidden resource. This paper discusses the SDP concept and analyzes its performance in the event of a Distributed Denial of Service (DDoS) attack under two different environments - one virtual and one real-world. The results indicate that SDP provides a resilient method for protection again DDoS attacks. While it requires slightly more time for connection setup, it is offset by its exceptional performance even under duress.

Domain Name System (DNS) is a crucial component of current IP-based networks as it is the standard mechanism for name to IP resolution. However, due to its lack of data integrity and origin authentication processes, it is vulnerable to a variety of attacks. One such attack is Typosquatting. Detecting this attack is particularly important as it can be a threat to corporate secrets and can be used to steal information or commit fraud. In this paper, a machine learning-based approach is proposed to tackle the typosquatting vulnerability. To that end, exploratory data analytics is first used to better understand the trends observed in eight domain name-based extracted features. Furthermore, a majority voting-based ensemble learning classifier built using five classification algorithms is proposed that can detect suspicious domains with high accuracy. Moreover, the observed trends are validated by studying the same features in an unlabeled dataset using K-means clustering algorithm and through applying the developed ensemble learning classifier. Results show that legitimate domains have a smaller domain name length and fewer unique characters. Moreover, the developed ensemble learning classifier performs better in terms of accuracy, precision, and F-score. Furthermore, it is shown that similar trends are observed when clustering is used. However, the number of domains identified as potentially suspicious is high. Hence, the ensemble learning classifier is applied with results showing that the number of domains identified as potentially suspicious is reduced by almost a factor of five while still maintaining the same trends in terms of features' statistics.

In recent years, there has been great interest in utilizing the unlicensed spectrum for mobile data traffic, as deployment of mobile systems are facing severe challenges due to licensed spectrum scarcity. Increasing the bandwidth is a possible solution to increasing the network capacity, however, the licensed spectrum is limited and can be very costly to obtain. Additionally, the lack of available licensed band limits the network capacity which may affect the user experience. Due to this limitation, the research focuses on the coexistence of the LTE and Wi-Fi in the unlicensed band. Currently, there is minimal to no sharing mechanism between the two technologies, which will cause significant interference for the users. This project focuses on using the time-domain virtualization, where the sharing mechanism is allocated in time slots rather than allocating frequency for each technology. The coexistence mechanism of the two technologies are discussed and evaluated using the simulation of a proposed scheduling algorithm.

Cloud computing has become a business reality that impacts technology users globally. It has become a cornerstone for emerging technologies and an enabler of the future Internet services. This has been coupled with emerging trend of adopting software-based network infrastructure. Paradigms such as Software-defined networks (SDNs) have gained more attention for large scale networks due to the flexibility and agility they offer to the network. Parallel to this emergence, the power consumption has rapidly increased. Hence, developing green and sustainable solutions has become a prime concern for the cloud providers. In this paper, a novel power aware algorithm titled “Green Distributed Cloud Services Provisioning” to provision the tenants' cloud services requests on an SDN-enabled cloud environment is presented. The role of the cloud and SDN controllers within the considered environment is illustrated to better highlight the interactions with the underlaying infrastructure. Furthermore, a brief review of well known cloud simulators is given. The conducted simulations show that the proposed algorithm significantly reduces the power consumption while maintaining high admission rates when compared to other work from the literature. Moreover, the average number of hops needed for a tenant to reach its requested VM was shown to remain stable with the increase in network load.

Wireless networks are facing various challenges that demand continuous and rapid improvement. Long-Term Evolution (LTE) is a preferred wireless technology because of its satisfactory performance. Owing to an exponential increase in demand and new potential applications, the core network of LTE, which is known as the Evolved Packet Core (EPC), is affected by a surge in signaling caused by a variety of control functions. The signaling overhead decreases the users' Quality of Experience (QoE). The current study attempts to improve the intelligence of location management techniques. As an extension of our previous study [1], a Self-Organizing Network (SON) that enables dynamic reconfiguration of cell-to-TAL/MME is introduced. Both centralized and distributed pooling schemes are tested in terms of signaling overhead and user power consumption. A decomposition model that reduces the original formulated problem to two sub-problems is proposed, each of which is solved optimally. In addition, a smart cell-to-TAL selection scheme is proposed to prioritize potential cells that might be visited by a user equipment (UE). Our method is shown to outperform several state-of-the-art methods presented in the literature. Finally, a heuristic algorithm is presented to obtain a less complex solution than the optimal one.

Energy-efficient communications have become one fundamental aspect for today's cutting-edge wireless technologies due to its valuable impact on the environment. In this paper, we augment our earlier study for the user equipment's (UE) energy efficiency (EE) in the long-term evolution (LTE) downlink by looking at real-time heterogeneous traffic QoS requirements. In particular, we utilize the previously proposed cloud radio access network (C-RAN) and ray tracing (RT)-based scheduling model to optimize both of the EE and the packet delay jitter for real-time applications with fixed packet delay budget subject to other traffic types requirements. Using the utility-based scheduling approach, we formulate the resource allocation problem as a weighted sum binary integer programming (BIP) problem. Due to the inherent complexity of the problem formulation which hinders finding its solution directly, four heuristic algorithms are proposed to solve the optimization problem. Numerical simulations are conducted on three different traffic types each belonging to one of the popular QoS classes; best-effort class, rate, and delay-constrained classes. The obtained results demonstrate a substantial improvement in the system's performance achieved by our proposed schemes compared to other existing schemes.

Wireless networks have faced increasing demand to cope with the exponential growth of data. Conventional architectures have hindered the evolution of network scalability. However, the introduction of cloud technology has brought tremendous flexible and scalable on demand resources. Thus, cloud radio access networks (C-RANs) have been introduced as a new trend in wireless technologies. Despite the novel advancements that C-RAN offers, remote radio head (RRH)-to-base band unit (BBU) resource allocation can cause significant downgrade in efficiency, particularly the allocation of computational resources in the BBU pool to densely deployed small cells. This causes an increase in power consumption and wasted resources. Consequently, an efficient resource allocation method is vital for achieving efficient resource consumption. In this paper, the optimal allocation of computational resources between RRHs and BBUs is modeled. This is dependent on having an optimal physical resource allocation for users to determine the required computational resources. For this purpose, an optimization problem that models the assignment of resources at these two levels is formulated. A decomposition model is adopted to solve the problem by formulating two binary integer programming subproblems; one for each level. Furthermore, two low complexity heuristic algorithms are developed to solve each subproblem. Results show that the computational resource requirements and the power consumption of BBUs and the physical machines decrease as the channel quality worsens. Moreover, the developed heuristic solution achieves a close to optimal performance while having a lower complexity. Finally, both models achieve high resource utilization, cementing the efficiency of the proposed solutions.

The term Energy Efficiency (EE) is turning out to be a radical characteristic for today's 4G networks - which provide energy-hungry wireless services - especially from the battery-limited devices' perspective. In addition to the EE, meeting firm levels for the quality-of-service (QoS) of those energy demanding applications is inevitable. In this paper we study the EE of the user equipment (UE) in the LTE downlink and the delay jitter as a fundamental QoS metric for various real-time applications. The study focuses mainly on the Voice over LTE (VoLTE) traffic as being a heavily used service. We provide a multiobjective optimization for both the EE and the delay jitter subject to fixed delay budget. To address the complexity of the proposed optimal scheduler, two different heuristic algorithms were developed. Numerical results demonstrate that our proposed schedulers achieve better trade-off for the EE versus the delay jitter compared to existing state-of-the-art schedulers.

To meet the increasing mobile data services demand, several solutions have been proposed such as wireless resource virtualization and device-to-device (D2D) communication. Virtualization allows for more efficient utilization of the spectrum, reduces expenditures, and can support higher peak rates. D2D communication can achieve higher data rates due to the proximity of devices while controlling the interference it causes to cellular communication. However, the increase in data rate multimedia demand has led to an increase in global energy consumption. Thus, it is crucial to employ more energy-aware schemes as this would provide both environmental and financial gains for service providers. In this paper, we extend our work in [1] by formulating the problem of power-aware wireless resource virtualization with D2D communication underlaying the LTE network. Since the problem is a mixed integer non-linear programming problem (MINLP), it is divided into four smaller linear programs, each of which is solved to optimality. Two lower complexity heuristic algorithms to solve the power allocation problems are introduced. Results show significant savings at both eNodeB and D2D devices.

Long term evolution (LTE) is a leading mobile technology that provides very high speeds, low latency, and better quality-of-service (QoS). However, because of the exponential growth in the number of mobile users, the variety of new handheld devices, and the incremental use of different applications, the core network experiences a significant signaling overhead. This demand requires the design of intelligent, optimized mobility management methods. The present work attempts to overcome signaling overhead expansion and to define the fundamental basis for designing a tracking area list (TAL). In this context, we differ from other studies by introducing a model that relates the tracking area list to the mobility management entity (MME) which enables more control and adds intelligence to the system. Two MME pooling schemes are investigated namely, centralized and distributed MME schemes. The proposed model is NP-hard; thus, the problem can be simplified with a few assumptions (which do not violate the constraints of the problem) to become a solvable linear problem (LP). Moreover, a low-complexity heuristic algorithm is developed by determining the percentage use of the lists/MME in each cell. The results show that the centralized scheme outperforms the distributed one. Also, the heuristic algorithm offers sub-optimal results when compared to the LP solution.

In this paper, as an extension to [1], we propose a prioritized multi-layer network coding scheme for collaborative packet recovery in hybrid (interweave and underlay) cellular cognitive radio networks. This scheme allows the uncoordinated collaboration between the collocated primary and cognitive radio base-stations in order to minimize their own as well as each other's packet recovery overheads, thus by improving their throughput. The proposed scheme ensures that each network's performance is not degraded by its help to the other network. Moreover, it guarantees that the primary network's interference threshold is not violated in the same and adjacent cells. Yet, the scheme allows the reduction of the recovery overhead in the collocated primary and cognitive radio networks. The reduction in the cognitive radio network is further amplified due to the perfect detection of spectrum holes which allows the cognitive radio base station to transmit at higher power without fear of violating the interference threshold of the primary network. For the secondary network, simulation results show reductions of 20% and 34% in the packet recovery overhead, compared to the non-collaborative scheme, for low and high probabilities of primary packet arrivals, respectively. For the primary network, this reduction was found to be 12%.

In order to accommodate mobile users' consumed power with the rapid increase of multimedia-rich mobile data, additional network capacities with optimized power allocation scheduling algorithms should be deployed. Motivated by the fundamental requirement of extending the mobile devices' battery utilization time per charge, this work formulates the optimized power allocation problem in a virtualized scheme considered in the third generation partnership project-long term evolution (3GPP-LTE) uplink (UL) systems. The proposed framework efficiently shares the evolved nodeB's dedicated physical radio resources blocks of service providers having different requirements under dynamic channel conditions. The objective is to minimize the total transmission energy for all users subject to exclusive and contiguous allocation, maximum transmission power, and rate constraints. Two algorithms are developed. A binary integer programming (BIP)-based algorithm is used to solve a simplified version of the problem. A heuristic algorithm is also presented that approaches the BIP-based algorithm's performance. Simulation results show that the proposed framework offers a remarkable transmission power reduction in the virtualized scenario as compared to the non-sharing one.