5G will serve heterogeneous demands in terms of data-rate, reliability, latency, and efficiency. Mobile operators shall be able to serve all of these requirements using shared network infrastructure's resources. To this end, we propose in this paper a framework for resource orchestration for 5G network slices implementing four Quality of Service pillars. Starting from traffic classification, demands are marked so that they are best served by dedicated logical virtual networks called Network Slices (NSs). To optimally serve multiple NSs over the same physical network, we then implement a new dynamic slicing approach of network resources exploiting Machine Learning (ML). Indeed, as demands change dynamically, a mere recursive optimization leading to progressive convergence towards an optimum slice is not sufficient. Consequently, we need an initial well-informed slicing decision of physical resources from a total available resource pool. Moreover, we formalize both admission control and slice scheduler modules as Knapsack problems. Using our 5G experimental prototype based on OpenAirInterface (OAI), we generate a realistic dataset for evaluating ML based approaches as well as two baselines solutions (i.e. static slicing and uninformed random slicing-decisions). Simulation results show that using regression trees as an ML based approach for both classification and prediction, outperform other alternative solutions in terms of prediction accuracy and throughput.

Slicing is an emergent technology for 5G. It decomposes a single Radio Access Network (RAN) into multiple virtual networks "slices" to meet demands in term of throughput, mobility, latency, reliability, etc. Slicing needs real-time reconfigurations to keep current with demands' dynamics. This results in an increased cost of Operation Expenditures (OPEX). We approached this challenge as an optimization problem of infrastructure's resources. We virtualized and pooled Baseband Units (BBUs) resources on cloud. Dynamic allocation and interconnection with Remote Radio Heads (RRHs) are made possible by leveraging the advents of Network Function Virtualization (NFV) and Software-defined Networking (SDN). We implemented Distributed Base Station (DBS) using open software platform along to a public service orchestration tool for clouds. Our contribution is integrating service selection and deployment with real-time monitoring that allowed auto-control of resources by looping resources' lifecycle. In our experiments, we deployed several slices and we tested two scenarios. First scenario addressed slices' auto-scaling (Infrastructure Scale-Out/Scale-In) when free resources are available in the pool. Second scenario simulated slices' breathing (orchestration of resources) when the pool of resources is exhausted. In first scenario, results show that leveraging cloud elasticity by auto-scaling resources saves costs by providing exactly "what-is-needed" "when-it-isneeded" in term of cloud computing. In second scenario, results show that slices' breathing maximizes the usability by employing our "inhale-and-exhale" heuristic. It is about reusing resources from under-loaded slices in favor of overloaded ones with seamless effect on the user-experience.

With the emergence of virtualization and software automation for mobile networks, network slicing is enabling operators to dynamically provision network resources tuned to suit heterogeneous service requirements. This article investigates the architectures of the Fifth Generation (5G) of mobile networks experimental prototypes with a focus on network slicing. We present some existing 5G prototypes and identify their gaps. We, then, propose an architecture and a design of a 5G microservice based prototype. Such prototype has an ability to autoconfigure radio resources for network slices using machine learning (ML) powered decisions based on real-time acquired performance metrics. Finally, we discuss some use cases on top of this prototype and their related results before concluding.

No abstract

Multiple services with heterogeneous characteristics are anticipated with fifth generation of mobile communications (5G). Each requires an infrastructure tuned to serve requirements. To maintain affordability, operators will provide these services over a common infrastructure. We talk about provisioning Network Slices that are tailored to demands. A slice is a logical network built using Virtualized Network Functions (VNFs) that are enabled by leveraging Network Function Virtualization (NFV) and Software-Defined Networking (SDN). Orchestration is key for saving costs when tailoring infrastructures due to scarcity of resources. It enables operators to supply infrastructure with resources, meeting exactly with demands. In this demo, we present an on-the-fly provisioned cluster used to provision 5G Core Network (CN) as a Service (CNaaS). Our platform is designed for high-availability with self-healing feature. It uses Docker Swarm as orchestrator and Open Air Interface (OAI) for virtualizing the CN. We demonstrate resiliency, and the effortless on-demand scaling or orchestrated auto-scaling using bin-packing or spreading strategies for achieving elasticity.

No abstract

No abstract

Cloud-Radio Access Network (C-RAN) is an attractive solution to Mobile Network Operators. Firstly, C-RAN leverages the effect of pooling multiple Baseband Units (BBUs) to offer centralized processing resources while hosting them on cloud. This results in multiple benefits ranging from statistical multiplexing gains, to energy efficiency. Secondly, C-RAN allows deploying Remote Radio Heads (RRHs) in proximity of end-users allowing exploiting Inter-Cell Interference Cancellation (ICIC) to maximize throughput by coordinating multiple RRHs. In this context, we propose, in this paper, a new throughput-aware RRHs clustering method for C-RAN that maximizes the throughput for end-users, while meeting multiple constrained resources on BBUs. Our approach consists of two stages: First, individual throughput value and requirements of each RRH are calculated taking into account the Signal-to-Interference-plus-Noise Ratio (SINR) values and the distance between RRHs and users. Then, they are included into a k-dimensional Multiple-Choice Knapsack Problem (k-MCKP) subject to several constraints in terms of required resources in order to form RRHs clusters that maximize the global throughput. Simulation results demonstrate the good performance of our proposal in terms of end-users throughput, spectral efficiency and execution time, when compared with the optimal solution and the basic strategy using no-clustering scenarios.