This paper introduces the first-ever open-source Multicast and Broadcast Services (MBS) implementation for 5G and beyond cellular networks in OpenAirInterface5G. It outlines the challenges and essential modifications required to incorporate MBS into the 5G architecture while ensuring compliance with the 3GPP specification. The study investigates the performance of the three Hybrid Automatic Repeat Request (HARQ) feedback options for MBS, including the No-ACK, ACK-NACK, and NACK-Only HARQ-ACK options. It evaluates the user plane performance of these HARQ feedback options under realistic network conditions, providing insights into their impact on reliability, throughput, spectral efficiency, and resource utilization for MBS in 5G and beyond networks. Specifically, it examines their effectiveness in scenarios where the User Equipment (UE) devices experience uniform or diverse network conditions, analyzing their overhead and the impact of the Modulation and Coding Scheme (MCS) on their overall performance. Our experiments highlight the substantial performance benefits of MBS over Unicast transmissions and the trade-off between reliability and performance introduced by the HARQ-ACK options. This work advances the practical implementation of MBS in 5G and provides valuable guidelines on optimal deployment strategies and the HARQ feedback option selection for 5G and beyond cellular networks.

This paper proposes a state-of-the-art framework for adapting Active Queue Management (AQM) in 5G and beyond cellular networks with disaggregated Radio Access Network (RAN) deployments. While existing AQM algorithms effectively mitigate bufferbloat in monolithic RAN deployments, their potential in disaggregated ones remains largely unexplored. This gap particularly relates to AQM algorithms relying on communication between layers distributed across distinct network entities to operate. Our research explores the current literature on AQM, identifies the gaps regarding disaggregated deployments, and introduces a comprehensive framework that employs Artificial Intelligence (AI) and Machine Learning (ML) within the RAN Intelligent Controller (RIC) for adapting AQM in such deployments. We evaluate our novel solution on a previously proposed AQM algorithm which requires cross-layer communication, using OpenAirInterface5G (OAI5G) to deploy a disaggregated RAN and a connected User Equipment (UE) that experiences realistic network conditions, including noise and mobility. Finally, we assess its accuracy through the Quality of Service (QoS) achieved for our disaggregated deployment on the NITOS testbed.

We study TDMA-scheduled Software Defined Wireless Multihop Networks (SDWMNs), whereby the data traffic and SDN control messages share the same network links and TDMA resources. Since the topology of WMNs dynamically changes, maintaining a responsive SDN plane is essential for meeting data traffic rate requirements. Placing more SDN controllers reduces communication delays at the SDN layer and increases its responsiveness. However, it demands more TDMA resources and reduces the available ones for data traffic. We analyze this trade-off between data traffic performance and SDN layer responsiveness by delving into two distinct resource allocation mechanisms in the WMN, the SDN controller placement and TDMA scheduling. We capture their interaction into an optimization problem formulation, which aims at maximizing the SDN-responsiveness subject to data traffic rate requirements, topology conditions, and the available TDMA resources. We propose a novel heuristic for the hard-to-solve problem that leverages the network state information gathered at the SDN layer. We find that our heuristic can increase the SDN-responsiveness by 44% when varying the rate reserved for rate-elastic data traffic within 40% of what is nominally requested. The heuristic is modular in accommodating different controller placement algorithms and robust to different alternative for the SDN software implementation.

Numerous well-known applications use the Raft consensus algorithm to maintain consistent replicas of their data on distributed nodes. Raft is based on a dynamically elected leader which is one of the distributed nodes, and its operations are unfortunately suspended during the election of the leader. Elections can be triggered by the failure of the current leader, in which case they are unavoidable, or by a network disconnect between the leader and another node, in which case a new inefficient leader will likely replace the previous one at the expense of additional system downtime. In this paper, Raft messages are monitored at every node, and Machine Learning is used to classify the aforementioned causes of each election. This data is used to increase the system's availability by decreasing the total number of elections that could be conducted in a given time unit. Three supervised classifiers were trained with messages generated in a real Raft-operated distributed system that was deployed on a testbed and where multiple events triggering elections were applied. All classifiers are nearly 97% accurate at classifying the causes of these elections, approaching even 100% in some cases.

The adoption of the Software Defined Networking (SDN) paradigm is aggressively expanding from traditional datacenter networks to 5G and IoT deployments due to its flexibility in network management and routing. Mobile Ad-Hoc Networks (MANETs), with their unique characteristics and requirements stemming from the mobility and volatile wireless medium conditions, have not yet realized the benefits of the SDN approach. Adoption has been hampered by technical challenges and the contradiction between the centralized concept of SDN and the distributed one of traditional MANET routing schemes. In this paper, we present our implementation of a fully-fledged SDN framework, which attempts to bridge the gap by combining the benefits of both worlds. Our developed prototype integrates with 802.11 Ad-Hoc networks and offers a) fault tolerance for the SDN controller by recovering from failures with the (re)-election of a new controller, b) dynamic network topology discovery, and c) a cross-layer approach in routing based on 802.11 MAC layer statistics that more accurately characterize link quality and capacity. We validated the applicability and performance benefits of our method by evaluating our proposed SDN architecture in a proof-of-concept scenario.

Raft is a state-of-the-art consensus algorithm for state replication over a distributed system of nodes. According to Raft, all state updates occurring anywhere in the system are forwarded to the leader, which is elected among the system nodes to collect and replicate these updates to all other nodes. Thus, the time required for the state replication, named as system response time, depends on the delays between the leader and all other nodes. After multiple node failures and leadership transitions, each node can be leader with a probability that affects the expected response time. The leadership probabilities, in turn, are affected by the random intervals that nodes are waiting, after detecting a leader failure and before competing for the successive leadership. The Raft designers suggest the ranges of these intervals to be equal for all nodes. However, this may result in increased expected response time. In this paper, mathematical models are presented for estimating the ranges resulting in the desired leadership probabilities. The presented theoretical results are also confirmed by testbed experimentation with an open-source and widely used Raft implementation.

Despite recent progress in orchestration of Virtual Network Functions (VNFs) and in multi-technology SDN connectivity, the automated provisioning of end-to-end network services composed of virtual functions deployed across distributed compute locations remains an open challenge. This problem is especially relevant to support the deployment of future 5G networks, comprising virtual access and core network functions connected through a potentially multi-domain transport network. In this paper we present and demonstrate the 5GOS, a lightweight end-to-end orchestration framework that enables the automated provisioning of virtual radio access network services. Using an experimental multi-domain testbed we demonstrate that the 5GOS can provision multi-domain virtual Wi-Fi and LTE services in less than three minutes.

This paper discusses the benefits in the operation of a Raft based SDN controller cluster, when the election of the cluster leader becomes more or less “fair. Raft is a leader based consensus algorithm, which is used by the most popular open-source SDN controllers for replicating the network state. It requires all state changes to be confirmed by the leader, thus the leader election is very crucial for the Raft performance. In case that the inter-controller communication delay is the same for all controller pairs, the election process is absolute fair, meaning that the leadership is shared equally among the controllers. In all other cases, some controllers become leaders more frequently in benefit or at a cost of the average time required for a network state update. In this paper, we model this time as a function of the leadership probabilities of the cluster controllers. We also model these probabilities as a function of the time that each controller is waiting after detecting the current leader failure and before starting its campaign. We configure different ranges for the controller waiting times, adjusting the leadership probabilities and decreasing the average response time. Our model is confirmed by testbed experimentation.

Ethernet Virtual Private Network is an emerging Data Center Interconnect technology, which relies on Multiprotocol BGP and an encapsulation protocol over IP, such as Virtual Extensible LAN. It enables the creation of Layer 2 overlays on top of provider IP networks, which can be interconnected through the symmetric or asymmetric Integrated Routing and Bridging extension. In this work, we present our testbed implementation of these technologies using OpenFlow and we compare the advantages and disadvantages of the symmetric and asymmetric solutions.

Software Defined Networking (SDN) decouples the control and data planes and moves the control logic to the SDN controllers. The traffic in the control plane is either related to the interconnection of the controllers or to the connectivity between the SDN switches and the controllers. The controller placement affects the total control traffic. Multiple distributed controllers increase the inter-controller traffic, while few concentrated controllers increase the controller to switch traffic. We model the problem of determining the optimal controller placement for minimum control traffic. We further discuss the complexity of the optimization problem, and devise a heuristic algorithm for its solution. Our simulations and test bed experimentation reveal close to optimal performance of the presented heuristic algorithm.

Software-defined networking is at the root of future 5G network design. Among others, it allows for automated network reconfiguration and network slicing support. In this paper we present an implementation of a hierarchical control plane in the transport network architecture envisioned by the 5G-PICTURE project. We analyze the implementation of the slicing mechanism at the network edge and the end-to-end path establishment procedure, which involves interactions within a hierarchy of controllers. We also evaluate the latency performance of our control plane solution, by means of testbed experimentation.

In recent years, the rise of Network Function Virtualization (NFV) makes the Network Service (NS) deployment much agile and flexible. The proprietary and custom-made hardware is replaced by a virtual and software-based infrastructure, that is easily exploited in a common way for the NS deployment. One of the most challenging problems in this environment is deploying and organizing in a large-scale and multi-domain infrastructure, which contains geographically distributed but interconnected data-centers. The proposed solution focuses on the extra networking operations required in a NFV infrastructure, managed by Open Source MANO, OpenStack and OpenDaylight. We develop software proxies that collaborate with the aforementioned tools and enhance their functionality. Finally, we implement and evaluate the proposed architecture, using the NITOS experimentation testbed.

5G networks bring increased flexibility for the operator at different levels. On one side, RAN disaggregation based on the Cloud-RAN concept allows the instantiation of base stations in an area based on demand, whereas on the other side NFV-MANO orchestration brings effortless delivery of services deployed over distributed infrastructure. In this paper, we demonstrate a disaggregated heterogeneous CloudRAN, consisting of cellular and WiFi infrastructure, deployed through the Open Source MANO orchestrator in the distributed infrastructure of a testbed. Through a pair of mmWave and WiFi redundant links, we backhaul the RAN and in case of a broken link, seamlessly switch technologies without affecting the traffic delivered to the end-user.

In recent years, the rise of Network Function Virtualization (NFV) makes the Network Service (NS) deployment much agile and flexible. The proprietary and custom-made hardware is replaced by a virtual and software-based infrastructure, that is easily exploited in a common way for the NS deployment. One of the most challenging problems in this environment is deploying and organizing in a large-scale and multi-domain infrastructure, which contains geographically distributed but interconnected data-centers. The proposed solution focuses on the extra networking operations required in a NFV infrastructure, managed by Open Source MANO, OpenStack and OpenDaylight. We develop software proxies that collaborate with the aforementioned tools and enhance their functionality. Finally, we implement and evaluate the proposed architecture, using the NITOS experimentation testbed.

This paper proposes distributed load shedding policies for regulating excessive network load. Data packets are inserted into the network to be delivered to intended destinations. The intermediate network nodes may decide to forward or shed some packets depending on temporally available resources. It is possible for some packets to traverse several nodes in the network until they are finally dropped before reaching the destination, which exacerbates energy consumption. We define a multi-objective optimization problem where we aim to minimize the used energy subject to providing maximum sum throughput. For the case of single-path unicast sessions, we show that Energy-efficient Distributed Load Shedding (E-DLS), a simple shedding mechanism combined with pushback routing, solves this load di shedding optimization. We implement E-DLS in a testbed and use the experiments to select policy parameter values that strike a good balance between energy and delay performance. We then propose a heuristic extension of E-DLS for multirate multicast routing, and showcase via testbed experiments its optimal performance.

The wide proliferation of IEEE 802.11 compatible devices and the provisioning of costless Internet connectivity in most cases, have created fertile ground for investigating seamless client mobility and handoff management from a cellular technology to any wireless access point available. Although handoffs and client mobility are currently addressed by the IEEE 802.21 standard along with mobility management protocols such as Mobile IPv6, yet no remarkable efforts exist for the wide deployment of such solutions. Moreover, the adoption of such architectures requires considerable changes in the mobile node's networking stack. In this work, we propose a Software Defined Networking technology inspired scheme for managing client mobility among heterogeneous wireless networks, by adopting changes only on the network edges. Our solution is compatible with the existing IPv4 and IPv6 addressing solutions. By employing the OpenFlow technology on the border of our network with the Internet, we manage to keep both ends of the network aware of any topology changes, and thus preserve any already established connections, resulting in a seamless handoff process. We evaluate our technique in a real network setup, by employing WiFi and LTE technologies and benchmark it using higher layer protocols with multi-homing features, namely Stream Control Transmission Protocol and Multipath TCP.

We present a novel control scheme that dynamically optimizes multirate multicast. By computing the differential backlog at every node, our scheme adaptively allocates transmission rates per session/user pair in order to maximize throughput. An important feature of the proposed scheme is that it does not require source cooperation or centralized calculations. This methodology leads to efficient and distributed algorithms that scale gracefully and can be embraced by low-cost wireless devices. Additionally, it is shown that maximization of sum utility is possible by the addition of a virtual queue at each destination node of the multicast groups. The virtual queue captures the desire of the individual user and helps in making the correct resource allocation to optimize total utility. Under the operation of the proposed schemes backlog sizes are deterministically bounded, which provides delay guarantees on delivered packets. To illustrate its practicality, we present a prototype implementation in the NITOS wireless testbed. The experimental results verify that the proposed schemes achieve maximum performance while maintaining low complexity.

In this work, we present how programmable data-plane technology (software routers) offers an easy-to-apply mechanism to create virtual wireless networks and support buffering and scheduling decisions. Furthermore, we present a feedback-based buffering mechanism that is able to provide throughput ratio guarantees per virtual network, without requiring any modifications in the 802.11 driver and without relying on statistical knowledge of the workload per virtual network or knowledge regarding the channel conditions. We implement the proposed mechanism in a software router in a 802.11 Access Point and we evaluate its performance in a wireless testbed environment. The methodology and the mechanics developed are generic and with some modifications can be applied to differentiating services for other types of guarantees like delay.

Multirate multicast improves user quality but complicates network optimization. This paper introduces a novel control scheme to dynamically optimize multirate multicast. We present MMT, an adaptive policy which combines differential backlog scheduling and intelligent packet dropping, both based on local information. MMT is shown to maximize network throughput by adapting to changing conditions such as channel quality, network congestion, and device capabilities. Then, we study the problem of per-receiver network utility maximization. To maximize sum utility we propose the MMU policy, an extension of MMT with receiver-end flow control. Under the operation of both policies backlog sizes are deterministically bounded, which provides delay guarantees on delivered packets. An important feature of the proposed scheme is that it does not require source cooperation or centralized calculations. To illustrate its practicality, we present a prototype implementation in the NITOS wireless testbed. Experimental results verify the optimality of the scheme and its low complexity.

In this paper, we propose and characterize the performance of a novel video-aware Opportunistic Routing (OR) algorithm for multicast, using a one-hop and two-hop forwarding scheme in 802.11 mesh networks. We believe that the OR approach exploits the inherent broadcast nature of the wireless medium and adapts very well in the lossy wireless environment. Inferring from the above, we extend a state of the art routing algorithm, namely MORE, that leverages on this approach and offers multicast support, but is not efficiently applicable in video streaming. By employing our scheme, we enable support for video streaming application with hard time-constraints. In addition, we focus on the orchestration of the packet transmissions and the prioritization of the video traffic towards improving the video-perception quality of the end users. In order to evaluate the proposed scheme, we conducted experiments in a realistic medium-scale wireless testbed. Our results show that the proposed scheme increases the average video-perception quality, measured in PSNR, by up to 270% in some cases or up to 175% in average, compared to the MORE algorithm.

In this demo paper, we demonstrate and evaluate a novel Opportunistic Routing (OR) protocol for video multicast, namely Video-aware Multicast Opportunistic Routing (ViMOR), over 802.11 two-hop mesh networks. OR exploits the broadcast nature of the wireless medium and offers spatial diversity among the receivers. ViMOR extends MORE, a state of the art OR algorithm, by orchestrating packet transmissions and prioritizing video traffic, in order to conform with video streaming requirements. For the demonstration and evaluation of the proposed scheme, we proceeded with the development of the implementation on NITOS wireless testbed. Results showed a significant increase in average video-perception quality, compared to MORE protocol.

Experimentation in testbeds is gaining increasing ground as a necessary validation step for every theoretical study in communication networks. However, the first-come-first-served policy employed today by most testbeds does not ensure the fair and efficient utilization of their resources, which often lie idle. Ideally, every testbed should be utilized as much as possible and serve the most important requests. In this paper we introduce a novel resource scheduling mechanism for the wireless testbed NITOS. The proposed scheme is based on VCG auctions and includes an allocation and a pricing rule which induce the users to judiciously submit experiment requests. We prove theoretically and demonstrate numerically that this scheme ensures that the testbed resources (nodes and channels) are assigned to the users with the highest needs. Our mechanism can be incorporated in the next generation resource management systems for NITOS and similar testbeds.

In this paper, we present an implementation design of a TDMA protocol for the canonical diamond-topology network containing a source, two relays and a destination (single unicast session). Getting inspired by the established Lyapunov-methodology, we propose an online strategy for the relay selection/scheduling problem. In contrast to existing works, we implement this strategy inside the proposed TDMA protocol in order to operate over a CSMA enabled Wi-Fi infrastructure-less network. We elaborate a network controller within the TDMA frame to solve a global optimization problem at each time slot in a centralized manner. In our formulation, we consider the class of scheduling policies that select concurrently a non-interfering subset of links. Our architecture is tailored to achieve the objectives of stabilizing the network and either maximizing throughput or minimizing the total power consumption. Our scheme has been implemented and tested thoroughly through experimentation in the NITOS wireless testbed by exploiting Wi-Fi technology features. The results revealed significant increase in networking efficiency for throughput maximization.

Due to the unreliable nature of the wireless medium, provisioning of quality of service (QoS) in wireless LANs is far more complicated than in wired networks. In order to address this challenge, IEEE 802.11e defines a framework for QoS support where packets are prioritized based on their traffic characteristics. In this paper, we propose two new QoS support schemes. One is based on a ldquouser centricrdquo approach and the other on a ldquopacket content basedrdquo approach. The new mechanisms, in addition to the traffic itself, take into consideration the identification of the station that generates the traffic or the content of the traffic. Therefore, they use a second prioritization level on top of the one that is implemented in IEEE 802.11e. In the ldquouser centricrdquo approach, the mechanism defines groups of stations based on their MAC addresses and assigns different priorities to different groups. Under this classification, stations are served based on the prioritization of the group they belong. Among stations with same priority, traffic is scheduled based on the priorities given by 802.11e. On the other hand, in the case of the ldquopacket content basedrdquo approach, the mechanism defines groups of words or phrases with their respective priority. A packet that includes words of a specific group is scheduled based on the priority that the particular content defines. The new schemes are simple yet efficient, since they are adapted to the realistic needs of todaypsilas WiFi networks. In order to evaluate the performance of these proposed schemes, we implement them using open source drivers in a Linux platform. We run experiments in a medium-size testbed. Experimentation results clearly demonstrate the performance superiority of the new schemes, as compared to the legacy IEEE 802.11e.

Due to the unreliable nature of the wireless medium, provisioning of the quality of service (QoS) in wireless LANs is by far more complicated than in wired networks. In this demo we show the efficiency of a new QoS support scheme that is based on an user centric approach. The new mechanism takes under consideration the identification of the station that generates the traffic additionally to the traffic itself. Therefore, it uses a second prioritization level on the top of the one that is implemented in IEEE 802.11e. It first defines groups of stations based on their MAC addresses and it assigns different priorities to different groups. Under this classification, stations are served based on the prioritization of the group they belong. Among stations with same priority, traffic is scheduled based on the priorities given by 802.11e. The new scheme is implemented using open source drivers and 802.11g WiFi cards. In the demo, a video clip is streamed from a server to a client with high priority under heavy traffic load. The underlying wireless transport alternates between the classic IEEE 802.11e and the new prioritization scheme. The new scheme delivers a smooth and jitter-free user experience, while the video playout over IEEE 802.11e experiences noticeable jitter and frequent distortions. The demo clearly demonstrates the performance superiority of the new implemented scheme, as compared to the legacy IEEE 802.11e.