Service-aware Network Research Articles

Service-aware real-time slicing for virtualized beyond 5G networks

Edge Intelligence is expected to play a vital role in the evolution of 5G networks, empowering them with the capability to make real-time decisions regarding various allocations related to their management and service provisioning to end-users. This shift facilitates the transition from a network-aware approach, where applications are developed to manage network quality fluctuations, to a service-aware network that self-adjusts based on the hosted applications. In this paper, we design and implement a service-aware network managed from the network edge. We utilize and assess various Machine Learning models to classify cellular network traffic flows in the backhaul, aiming to predict their future impact on network load. Leveraging these predictions, the network can proactively and autonomously reallocate slices in the Radio Access Network via programmable APIs, ensuring the demands of the traffic-generating applications are met. The approach integrates innovative MLOps methodologies for distributed and online training, enabling continuous model refinement and adaptation to evolving network dynamics. Our framework was tested in a real-world environment with realistic traffic scenarios, and the results were evaluated in real-time, down to a granularity of 10ms. Our findings indicate that the network can swiftly adjust to traffic, providing users with slices tailored to their application needs. Notably, our experiments show that under the studied settings, the users experienced up to 4 times lower latency (jitter) and nearly 4 times higher throughput when interacting with various applications, compared to the standard non-AI/ML unit. Furthermore, our dynamic scheme significantly optimizes resource allocation, ensuring energy efficiency by avoiding over- and under-provisioning of resources.

Communication and Computation O-RAN Resource Slicing for URLLC Services Using Deep Reinforcement Learning

The evolution of the future beyond-5G/6G networks toward a service-aware network is based on network slicing technology. With network slicing, communication service providers seek to meet all the requirements imposed by the verticals, including ultra-reliable low-latency communication (URLLC) services. In addition, the open radio access network (O-RAN) architecture paves the way for flexible sharing of network resources by introducing more programmability into the RAN. RAN slicing is an essential part of end-to-end network slicing since it ensures efficient sharing of communication and computation resources. However, due to the stringent requirements of URLLC services and the dynamics of the RAN environment, RAN slicing is challenging. In this article, we propose a two-level RAN slicing approach based on the O-RAN architecture to allocate the communication and computation RAN resources among URLLC end devices. For each RAN slicing level, we model the resource slicing problem as a single-agent Markov decision process and design a deep reinforcement learning algorithm to solve it. Simulation results demonstrate the efficiency of the proposed approach in meeting the desired quality of service requirements.

Service-Aware Design Policy of End-to-End Network Slicing for 5G Use Cases

End-to-End network slicing is an emerging technology with huge potential to carry flexible resource provisioning for the next generation of mobile networks (5G). With the powerful SDN and NFV technologies, multiple and logical network slices of various services can be deployed to the same underlying infrastructure. Physical resources are flexibly configured for different use cases. Therefore, it is necessary to explore how to guide the implementation of network slices via a general design policy. In this paper, we propose a service-aware network slicing design policy for 5G use cases, not only to satisfy various service requirements for different applications but also to enhance resource efficiency with fluctuated traffic demands. Firstly, multiple design objectives of slices are presented to guarantee different services in diverse use cases. Then we formulate a deterministic optimization model of the network slicing design problem and further extend it into a robust version considering uncertain traffic demands. Besides, a heuristic algorithm is proposed to obtain a promising network slicing design policy with the low computational effort, which is inspired by the multiple-objective particle swarm optimization (MOPSO) algorithm. The effectiveness of the present formulations and algorithms is verified by a series of simulations.

Online algorithm for migration aware Virtualized Network Function placing and routing in dynamic 5G networks

The goal of Network Function Virtualization (NFV) is to build agile and service-aware networks by building a new paradigm of provisioning network services where physical network functions are deployed as Virtualized Network Functions (VNFs). Due to the advantages of NFV, 5G networks have adopted NFV as one of its key enabling technologies. One of the most important challenges for realizing NFV-enable 5G Networks is VNF placing and routing. However, most existing studies neglected flow rate fluctuation of service requests, which could cause service migration and fluctuation of queues (buffers) in the networks. In this paper, we study the cost-minimizing problem of VNF placing and routing in this dynamic environment while considering service migration and queue backlog stability. We first formulate the problem as a stochastic optimization problem. Then we propose an online algorithm, named MACRO, to solve the problem. In order to bound the worst-case delay encountered by requests we propose WEB-MACRO. MACRO and WEB-MACRO base on Lyapunov optimization and make decisions without knowing future information. The theoretical analysis suggests that MACRO and WEB-MACRO achieve an optimality gap of O(1V), where V is a tunable parameter that controls the tradeoff between cost and backlogs. In addition, the queue backlogs maintained by MACRO are bounded by O(V) and the queue backlogs maintained by WEB-MACRO are upper bounded which accordingly bounds the worst-case delay encountered by service requests. The experiment results suggest that MACRO and WEB-MACRO achieve queue stability and reduce the total cost by 36.49% when V=2 and 45.84% when V=6, compared with the state-of-the-art method.

Smart Service-Aware Wireless Mixed-Area Networks

Smart services such as smart home, smart grid, smart transportation, and so on, are enabled with advanced communications networks, data analysis and control. However, challenges exist in designing a supporting network with optimal resource management while achieving quality- of-experience for a smart service. In this article, we propose a unified SA-WMN architecture that can be applied to different smart services. The core of the SA-WMN architecture is a smart service-aware network controller. The mega network controller has a service adaptation module, network cognition module and resource provisioning module that can achieve cost-effective wireless-mixed area network deployment, and efficient network resource management of SA-WMN. The proposed SA-WMN architecture is expected to close the gap on wireless network design, improve energy efficiency and enhance spectrum efficiency and access, in a smart service- aware paradigm. It will pave the way for realizing next-generation wireless network based smart services.

Enabling hard service guarantees in Software-Defined Smart Grid infrastructures

Information and Communication Technology (ICT) infrastructures play a key role in the evolution from traditional power systems to Smart Grids. Increasingly fluctuating power flows, sparked by the transition towards sustainable energy generation, become a major issue for power grid stability. To deal with this challenge, future Smart Grids require precise monitoring and control, which in turn demand for reliable, real-time capable and cost-efficient communications. For this purpose, we propose applying Software-Defined Networking (SDN) to handle the manifold requirements of Smart Grid communications. To achieve reliability, our approach encompasses fast recovery after failures in the communication network and dynamic service-aware network (re-)configuration. Network Calculus (NC) logic is embedded into our SDN controller for meeting latency requirements imposed by the standard IEC 61850 of the International Electrotechnical Commission (IEC). Thus, routing provides delay-optimal paths under consideration of existing cross traffic. Also, continuous latency bound compliance is ensured by combining NC delay supervision with means of flexible reconfiguration. For evaluation we consider the well-known Nordic 32 test system, on which we map a corresponding communication network in both experiment and emulation. The described functionalities are validated, employing realistic IEC 61850 transmissions and distributed control traffic. Our results show that hard service guarantees can be ensured with the help of the proposed SDN solution. On this basis, we derive extremely time critical services, which must not be subjected to flexible reconfiguration.

Design and Implementation of the Service-Aware Traffic Engineering (SATE) in the LISP Software- DefinedWireless Network (LISP-SDWN)

Software Defined Wireless Networks (SDWN) have been considered to have a feasible architecture that enables the fast deployment of new services and solutions in response to the explosion in the number of users and network traffic. Currently, the telecommunications sector is ensuring flexibility in network management and configuration. However, fluctuations in traffic are still beyond the control of SDWN providers. This paper suggests ways to manage fluctuations of traffic with service type. We propose the design of a service-aware network management service that achieves the maximum network utilization among heterogeneous Radio Access Networks (RANs) as a form of Traffic Engineering (TE). In this paper, we implement and test the Service-Aware Traffic Engineering (SATE) that distributes the network traffic to RANs according to the service type of traffic in the network layer. The traffic shift latency (e.g., from an LTE RAN to a Wi-Fi RAN) is considered as a performance metric that does not affect the end-to-end latency of some network applications (e.g., VoIP), and it is 3.51ms from our testbed. Therefore, it might not affect the end-to-end latency of the VoIP application in the telecommunications. SATE is implemented using OpenDaylight (ODL) and Ingress/Egress Tunneling Routers (xTRs) running on Vector Packet Processing (VPP).

Software Defined Wireless Networks: A Survey of Issues and Solutions

Wireless networks such as mobile networks, with their inflexible and expensive network infrastructure, are facing various challenges in efficiently handling the exponentially growing traffic demands of users. Hence, mobile network operators (MNOs) are looking forward to novel networking paradigms which could simplify the task of network management and control and allow faster deployment of newer solutions on top of existing hardware by software updates. Software defined networking (SDN) is a revolutionary technology which makes networks more agile and flexible by separation of data plane and control plane tasks. SDN is playing a key role in deploying mobile network services on network function virtualization (NFV) infrastructure for elastic and flexible deployment of core network services e.g., virtualization of evolved packet core in LTE to efficiently handle huge control signal overhead and traffic demands from machine-to-machine and internet of things devices. Besides, NFV and SDN are offering scalable, flexible and adaptable network service chaining platforms as a replacement for inflexible middleboxes. Mobile edge computing (MEC) is a new evolving platform which brings together IT services and cloud computing for offering end-users network aware services and solutions. In this survey, we enumerate numerous issues and challenges in designing SDN based wireless networks and review various SDN based seminal solutions for 4G/5G. Finally, this survey presents the role of SDN, NFV and MEC in designing 5G networks.

Software-Based Management for Ethernet Networks

An adequate network management is an essential part for each network permitting to keep the network performance at a suitable level. In this paper, we present our work focusing on an Ethernet network which is used as background interconnection technology for a smart home system in large buildings like tenements or an office block. In this case the network is larger and especially the applications are quite heterogeneous as compared to a private smart home system. Therefore, it is hard to supply a fair and quality of service aware network meeting all the demands. This is especially challenging in the case of an Ethernet network because this technology was intended to use it as a transparent connection system without the feature to support hard quality of service. We present how this problem could be solved by a middleware which best meets the special requirement of the mentioned use case. Finally, we discuss possible design improvements of this solution and present the recent Ethernet standards that could help solve this problem in an alternative way.

QoSPlan: A Measurement Based Quality of Service aware Network Planning Framework

In this article we present QoSPlan--a measurement based framework for preparing information relevant to Quality of Service (QoS)-aware IP network planning, which aims at reducing a core operational expenditure for the network operator. QoSPlan is designed to reduce the cost of deployment and maintenance of network monitoring systems. The process involves analysis of pre-existing accounting data to estimate a network-wide traffic matrix. Part of this estimation process relates to the generalization of QoS-related effective bandwidth coefficients taken from traffic analyzed on the network. We offer recommendations on how to appropriately realize QoSPlan to maximize its accuracy and effectiveness when applied to different network traffic scenarios. This is achieved through a thorough sensitivity analysis of the methods proposed using real traffic scenarios and indicative network topologies. We also provide an economic analysis of the deployment and maintenance costs associated with QoSPlan in comparison to a direct measurement approach, demonstrating cost savings of up to 60 % given different topology sizes.

Network-hosted avatar: User-terminal virtualization in the network

In a converged world, wireless users expect always-on connectivity in order to run several long-duration applications in parallel (e.g., streaming, messaging, peer-to-peer (P2P), or content publishing). This collides with the intrinsic limitations of wireless technologies: scarce and shared radio resources and the limited battery life of terminals. This letter introduces a new generation of service-aware network elements—network-hosted avatars—designed to help operators satisfy the expectations of their wireless customers. A network-hosted avatar is charged with representing a user's applications, data, and presence in a stable and permanent way. It hosts and executes user applications, independently of the terminal state. For example, a several hour file download may be completed while the terminal is switched off. In addition, it deals with radio conditions, ensuring optimal uplink/downlink transfer of content and information to the end user. The novelty of the approach resides in the selection and integration of state-of-the-art mechanisms best suited to help the operators. The resulting benefits are sessions without time constraints, reduced terminal power consumption, optimized usage of access network resources, ubiquitous access to personal data (anytime, anywhere), and permanent availability (i.e., intelligent presence and paging management). Preliminary performance evaluation indicates an ability to double the number of supported users and applications. Our next research steps will focus on a complete performance evaluation, on the means to add applications (plug-and-play) easily, and on the benefits of synergies among several collocated network-hosted avatars (farms).

Advances in network-supported media delivery in next-generation mobile systems

This article presents new concepts for network-supported media delivery in mobile networks. Automatic composition and merging of networks are central parts of these concepts. Media delivery is no longer an end-to-end service that only uses the network as an IP transport. Instead, these concepts create a service-aware network and provide customized delivery support through per-service overlay networks. They also integrate specialized processing nodes as part of the delivery topology, which include transcoders but also more complex processors, such as localized program insertions or personalized spam control. This article describes the underlying concepts and how these new network capabilities for media delivery services are requested, invoked, and managed

ProNet: A policy network environment

Network and business management convergence is seen as the next step towards harmonizing information technology resource allocation with target business goals. This paper proposes a management platform that starts at the specification of business targets in the form of policies and goes all the way to enforcing them in a next generation Quality of Service aware network. We specified both a policy language and a parser and also implemented procedures for mapping policies into underlying network mechanisms. Furthermore, case studies presented in this paper show performance results of network services controlled by such policies.