Next-generation radio access networks such as the Open Radio Access Network (O-RAN) have alleviated many of the 5G demand and management challenges. However, 0-RAN's intelligence, openness, and virtualization have signifi-cantly increased the attack surface of RAN s. This is specifically dangerous for critical 5G use cases such as Ultra-Reliable and Low-latency Communications (URLLC) due to its strict latency and reliability constraints. In this work, we focus on enhancing the security of the data streams and the security of the ML training and inference hosts in O-RAN URLLC deployments by introducing additional network security functions to 0- RAN's service function chains. Our goal is to maximize the amount of traffic examined by the security functions while adhering to 0- RAN's operational and functional constraints and upholding network defense patterns. Two security function types, encryption Virtualized Network Functions (VNFs) and intrusion detection system VNFs are chosen to achieve this objective. Encryption VNFs provide an additional layer of encryption for data traffic, while IDS VNFs protect the ML training and inference hosts of our solution. To solve this complex task, an advantage actor-critic deep reinforcement learning agent is developed, which actively allows adaptation to dynamic traffic. We demonstrate that our solution is capable of increasing the number of security functions in URLLC deployments allowing increased data protection and securing its own training and inference hosts.

Ultra-Reliable Low Latency Communications (URLLC) is a critical use case in 5G and B5G networks enabling applications such as Augmented Reality (AR)-assisted surgery, vehicle-to-everything communications, and smart grids to consistently deliver the promised Quality of Service to the end-users. The intelligence of the 5G core has made such applications possible, and the O-Radio Access Network (O-RAN) has extended this intelligence to Radio Access Networks (RANs) through its openness, cloudification, and ability to host machine learning models at every layer. However, the cloudification of O-RAN introduces challenges, such as securing availability and ensuring latency for URLLC. In this work, we propose an Availability- and Latency-Aware O-RAN Virtual Network Function (VNF) Protection (ALAP) solution. ALAP offers a shared VNF protection scheme based on deep Q-learning, efficiently providing this protection while minimizing the number of VNF backup components compared to dedicated protection schemes. Our solution protects against resource blockages and alleviates operational costs for network service providers. In addition to these objectives, ALAP ensures that the network meets URLLC’s strict availability and end-to-end latency constraints. ALAP has shown promising results in how quickly it can learn to optimize these objectives and in its capability to achieve its goals on large-scale O-RAN deployments.

The goal of Next-Generation Networks is to improve upon the current networking paradigm, especially in providing higher data rates, near-real-time latencies, and near-perfect quality of service. However, existing radio access network (RAN) architectures lack sufficient flexibility and intelligence to meet those demands. Open RAN (O-RAN) is a promising paradigm for building a virtualized and intelligent RAN architecture. This paper presents a Machine Learning (ML)-based Traffic Steering (TS) scheme to predict network congestion and then proactively steer O-RAN traffic to avoid it and reduce the expected queuing delay. To achieve this, we propose an optimized setup focusing on safeguarding both latency and reliability to serve URLLC applications. The proposed solution consists of a two-tiered ML strategy based on Naive Bayes Classifier and deep$Q- \mathbf{learning}$. Our solution is evaluated against traditional reactive TS approaches that are offered as xApps in O-RAN and shows an average of 15.81 percent decrease in queuing delay across all deployed SFCs.

The Open Radio Access Network (O-RAN) architecture brings openness, intelligence, and virtualization to RANs, allowing multi-vendor existence, achieving economics of scale, and enabling intelligent management and orchestration. O-RAN components such as near real-time RAN Intelligent Controllers (RICs), O-RAN Central Units (O-CUs), and O-RAN Distributed Unit (O-DUs) can be considered as virtual network functions hosted on the O-Cloud. This virtualization allows network service providers to disaggregate O-RAN functions from their hard-ware, enabling dynamic instantiation of services and reducing their capital and operating costs. However, with openness and virtualization, availability guarantees become more difficult to maintain as the network is now prone to both software and hardware failures. In this paper, we investigate a decomposition model for the design of reliable 0-RAN deployment under a dedicated virtual network function (VNF)-protection scheme. The proposed model maximizes the network's yearly availability by providing a placement decision for all 0-RAN VNFs and their backup instances. The model is solved by a column generation algorithm making it a scalable algorithm for large-scale 0-RAN deployments. Extensive computational results show that the algorithm can produce ε-optimal solutions with negligible ε (less than 0.1%) in reasonable computational times. These results significantly enlarge the exact solutions of the state-of-the-art algorithms for this problem.

Due to the huge surge in the traffic of IoT devices and applications, mobile networks require a new paradigm shift to handle such demand roll out. With the 5G economics, those networks should provide virtualized multi-vendor and intelligent systems that can scale and efficiently optimize the investment of the underlying infrastructure. Therefore, the market stakeholders have proposed the Open Radio Access Network (O-RAN) as one of the solutions to improve the network performance, agility, and time-to-market of new applications. O-RAN harnesses the power of artificial intelligence, cloud computing, and new network technologies (NFV and SDN) to allow operators to manage their infrastructure in a cost-efficient manner. Therefore, it is necessary to address the O-RAN performance and availability challenges autonomously while maintaining the quality of service. In this work, we propose an optimized deployment strategy for the virtualized O-RAN units in the O-Cloud to minimize the network's outage while complying with the performance and operational requirements. The model's evaluation provides an optimal deployment strategy that maximizes the network's overall availability and adheres to the O-RAN-specific requirements.

The shift towards a completely virtualized networking environment is triggered by the emergence of software defined networking and network function virtualization (NFV). Network service providers have unlocked immense capabilities by these technologies, which have enabled them to dynamically adapt to user needs by deploying their network services in real-time through generating Service Function Chain (SFCs). However, NFV still faces challenges that hinder its full potentials, including availability guarantees, network security, and other performance requirements. For this reason, the deployment of NFV applications remains critical as it should meet different service level agreements while insuring the security of the virtualized functions. In this paper, we tackle the challenge of securing these SFCs by introducing virtual security functions (VSFs) into the latency-aware deployment of NFV applications. This work insures the optimal placement of the SFC components including the security functions while considering the performance constraints and the VSFs' operational rules such as, functions' alliance, proximity, and anti-affinity. This paper develops a mixed integer linear programming model to optimally place all the requested SFCs while satisfying the above constraints and minimizing the latency of every SFC and the intercommunication delay between the SFC components. The simulations are evaluated against a greedy algorithm on the virtualized Evolved Packet Core use case and have shown promising results in maintaining the security rules while achieving minimum delays.