Search

In this paper, we investigate the problem of wireless service provisioning through a rotary-wing UAV which can serve as an aerial base station (BS) to communicate with multiple ground terminals (GTs) in a boost demand area. Our objective is to optimize the UAV control for maximizing the UAV.s energy efficiency, in which both aerodynamic energy and communication energy are considered while ensuring the communication requirements for each GT and backhaul link between the UAV and the terrestrial BS. The mobility of the UAV and GTs lead to time-varying channel conditions that make the environment dynamic. We formulate a nonconvex optimization for controlling the UAV considering the practical angle-dependent Rician fading channels between the UAV and GTs, and between the UAV and the terrestrial BS. Traditional optimization approaches are not able to handle the dynamic environment and high complexity of the problem in real-time. We propose to use a deep reinforcement learning-based approach namely Deep Deterministic Policy Gradient (DDPG) to solve the formulated nonconvex problem of UAV control with continuous action space that takes into account the real-time of the environment including time-varying UAV-ground channel conditions, available onboard energy of the UAV, and the communication requirement of the GTs. However, the DDPG method may not achieve good performance in an unstable environment and will face a large number of hyperparameters. We extend our approach to use the Trust Region Policy Optimization (TRPO) method that can improve the performance of the UAV compared to the DDPG method in such a dynamic environment.

Network traffic prediction is substantial for network optimization and resource management. However, designing an efficient predictive model considering different traffic characteristics, including periodicity, nonlinearity, and nonstationarity, is challenging. Recently, ensemble learning is attracting much attention from researchers in the machine learning community. Although ensemble learning has proven exceptional performance in modelling the intricate problems, it may not be able to handle varying patterns and chaotic behaviour, which are typical properties of traffic data (and many other time-series problems). For this reason, ensemble methods show a limited prediction accuracy in network traffic modelling. We address this issue by proposing an ensemble of learners for time-series prediction that considers the accuracy of individual learners as well as diversity among their outcomes. Each learner contributes to the optimization process by finding the optimal accuracy-diversity balance in a segment of feature space. This divide-and-conquer approach avoids complicated objective functions with many local optimums while fitting the ensemble model on large datasets. Experimental results on the real traffic traces show our proposed method outperforms other state-of-the-art predictors with 12% on average in prediction accuracy for different datasets.

Today, increasing attention is being paid to Data Center (DC) traffic classification since these infrastructures have become the heart of a variety of time-sensitive and data-intensive service platforms. Classification provides the required tools for better understanding traffic patterns in order to ensure high Quality of Service (QoS) performances and solve scalability problems. Unfortunately, existing classification algorithms cannot deal efficiently with two critical challenges in DC traffic: inter-class imbalance and critical time constraints. In this paper, we propose a novel correlation-based algorithm following a cost-sensitive approach combined with a Bagged Random Forest (BRF) ensemble algorithm, to address the inter-class imbalance problem while meeting time requirements in a data center environment. In this strategy, a new method based on Reverse k-Nearest Neighbors (RkNN) is proposed to capture the rebalancing weights expressing inter-flow correlations, in order to perform an online classification approach. We demonstrate the efficiency of the algorithm by comparing its performance to several existing methods from data level, algorithm level, and cost-sensitive strategies on four real-world datasets. The results reveal that the proposed algorithm outperforms most approaches in the different datasets in terms of precision, recall, F1 measure, AUC and Kappa, as opposed to other algorithms that result in either high precision with low recall and low precision and high recall causing congestion or resource over provisioning.

Flying and ground-based cars require various services such as autonomous driving, remote pilot, infotainment, and remote diagnosis. Each service requires specific Quality of Service (QoS) and network features. Therefore, network slicing can be a solution to fulfill the requirements of various services. Some services, such as infotainment, may have similar requirements to serve flying and ground-based cars. Therefore, some slices can serve both kinds of cars. However, when network slice resource sharing is too aggressive, slices can not meet QoS requirements, where resource under-provisioning causes the violation of QoS, and resource over-provisioning causes resource under-utilization. We propose two closed loops for managing RAN slice resources for cars to address these challenges. First, we present an auction mechanism for allocating Resource Block (RB) to the tenants who provide services to the cars using slices. Second, we design one closed loop that maps slices and services of tenants to virtual Open Distributed Units (vO-DUs) and assigns RB to vO-DUs for management purposes. Third, we design another closed loop for intra-slices RB scheduling to serve cars. Fourth, we present a reward function that interconnects these two closed loops to satisfy the time-varying demands of cars at each slice while meeting QoS requirements in terms of delay. Finally, we design distributed deep reinforcement learning approach to maximize the formulated reward function. The simulation results show that our approach satisfies more than 90% vO-DUs resource constraints and network slice requirements.

As one of the main scenarios in 5G mobile networks, ultra-reliable low-latency communication (URLLC) can satisfy the stringent requirements of many emerging applications. To ensure end-to-end secure delivery of critical data, 5G URLLC needs an efficient hybrid access scheme for licensed and unlicensed spectrum in mmWave bands. This article introduces machine learning (ML) and fountain codes into mmWave hybrid access, and proposes an adaptive channel assignment method. The proactively predictive power of ML can reduce the transmission delay, and the rateless characteristic of fountain codes can ensure transmission reliability without retransmission. Finally, through a vehicle- to-everything use case, the proposed method is demonstrated to clearly ensure the URLLC transmission requirements for critical data.

Smart home gateways have to forward multi-sourced network traffic generated with different distributions and with different quality-of-service (QoS) requirements. The state-of-the-art QoS-aware scheduling methods consider only the conventional priority metrics based on the IP type of service (ToS) field to make a decision for bandwidth allocation. Such priority-based scheduling methods are not optimal to provide both QoS and quality of experience (QoE), since higher priority traffic may not require lower delay than lower priority traffic (for example, traffic generated from medical sensors has a higher priority than traffic from streaming devices, but the latter one requires lower maximum delay). To solve the gaps between QoS and QoE, we propose a new queuing model for QoS-level Pair traffic with mixed arrival distributions in the smart home network (QP-SH) to make dynamic QoS-aware scheduling decisions meeting delay requirements of all traffic while preserving their degrees of criticality. A new metric that combines the ToS field and the maximum number of packets that can be processed by the system' s service during the maximum required delay is defined. Our experiments show that the proposed solution increases 15% of packets that meet their priorities and 40% of packets that meet their maximum delays as well as 25% of the total number of packets in the system.

In a frequency division duplexing multiple-input multiple-output (FDD-MIMO) system, the user equipment (UE) send the downlink channel state information (CSI) to the base station for performance improvement. However, with the growing complexity of MIMO systems, this feedback becomes expensive and has a negative impact on the bandwidth. Although this problem has been largely studied in the literature, the noisy nature of the feedback channel is less considered. In this paper, we introduce PRVNet, a neural architecture based on variational autoencoders (VAE). VAE gained large attention in many fields (e.g., image processing, language models, or recommendation system). However, it received less attention in the communication domain generally and in CSI feedback problem specifically. We also introduce a different regularization parameter for the learning objective, which proved to be crucial for achieving competitive performance. In addition, we provide an efficient way to tune this parameter using KL-annealing. Empirically, we show that the proposed model significantly outperforms state-of-the-art, including two neural network approaches. The proposed model is also proved to be more robust against different levels of noise.

Internet of Things is becoming one of the main triggers in designing and deploying new services aiming at fulfilling the wide demand imposed by end-users. Usually, concrete solutions addressing the optimization of the wireless segment are found in the literature. However, it is much less frequent to find end-to-end solutions to be easily adopted by the corresponding stakeholders. It is in this context that FLEXNET brings an integrated solution, relying on cutting-edge technologies, dealing with a wide set of technical requirements imposed by the different applications and services. This work was supported by FLEXNET Project: "Flexible IoT Networks for Value Creators" (Celtic 2016/3), in the Eureka Celtic-Next Cluster.

Among the innovative approaches to reduce the greenhouse gas (GHG) emissions of data centres during their use phase, cloud computing systems relying on data centres located in different regions appear promising. Cloud computing technology enables real-time load migration to a data centre in the region where the GHG emissions per kWh are the lowest. In this paper, we propose a novel approach to minimize GHG emissions cloud computing relying on distributed data centres. Unlike previous optimization approaches, our method considers the marginal GHG emissions caused by load migrations inside the electric grid instead of only considering the average emissions of the electric grid's prior load migrations. Results show that load migrations make it possible to minimize marginal GHG emissions of the cloud computing service. Comparison with the usual approach using average emission factors reveals its inability to truly minimize GHG emissions of distributed data centres. There is also a potential conflict between current GHG emissions accounting methods and marginal GHG emissions minimization. This conflict may prevent the minimization of GHG emissions in multi-regional systems such as cloud computing systems and other smart systems such as smart buildings and smart-grids. While techniques to model marginal electricity mixes need to be improved, it has become critical to reconcile the use of marginal and average emissions factors in minimization of and accounting for GHG emissions.

Software in IoT devices needs to be improved regularly to adapt security issues and new user requirements. In advanced IoT networks, devices employ the component-based software architecture in which components can be updated at run-time, such devices can download software components from neighbors, enabling fast distribution of updates in the entire network. One of the most energy consuming operations in the update process is flash re-writing in which the order of re-writing components into the flash memory is decisive for energy consumption. In this paper, we propose a mechanism that schedules updates in an entire IoT network to minimize the energy consumption, while satisfying the deadline constraint for updating all the devices. We mathematically formulate the problem of energy efficient update scheduling as an optimization problem with a novel energy model of the update process, then propose an algorithm to approximate the optimal schedule for updating all devices in the network. We examine the proposed algorithm in three different network instances including a tree, a partial mesh and a full mesh topology. Simulation results illustrate that our algorithm can obtain a near optimum which is, in the best case, only 3.2% different from the minimum.

Smart home gateway has to process different types of network traffic generated from several devices in an optimal way to meet their QoS requirements. However, the fluctuation of network traffic distributions results in packets concurrency. Current QoS-aware scheduling methods in the smart home networks do not consider concurrent traffic in their scheduling solutions. This paper presents an analytic model for a QoS-aware scheduling optimization of concurrent smart home network traffic with mixed arrival distributions and using probabilistic queuing disciplines. We formulate a hybrid QoS-aware scheduling problem for concurrent traffics in smart home network, and propose an innovative queuing design based on the auction economic model of game theory to provide a fair multiple access over different communication channels/ports. Our experiments show the proposed solution achieves an improvement of 14% of packets that meet their required delay and 57% of delay for different number of concurrent flows in the system.

Summary The environmental impacts of data centers that provide information and communication technologies (ICTs) services are strongly related to electricity generation. With the increasing use of ICT, many data centers are expected to be built, causing more absolute impacts on the environment. Given that electricity distribution networks are very complex and dynamic systems, an environmental evaluation of future data centers is uncertain. This study proposes a new approach to investigate the consequences of future data center deployment in Canada and optimize this deployment based on the Energy 2020 technoeconomic model in combination with life cycle assessment methodology. The method determines specific electricity sources that will power the future Canadian data centers and computes related environmental impacts based on several indicators. In case-study scenarios, the largest deployment of data centers leads to the smallest impact per megawatt of data centers for all of the environmental indicators. It is found that an increase in power demand by data centers would lead to a reduction in electricity exports to the United States, driving the United States to generate more electricity to meet its energy demand. Given that electricity generation in the United States is more polluting than in Canada, the deployment of data centers in Canada is indirectly linked to an increase in overall environmental impacts. However, though an optimal solution should be found to mitigate global greenhouse gas emissions, it is not clear whether the environmental burden related to U.S. electricity generation should be attributed to the Canadian data centers.

Infrastructure as a Service (IaaS) is considered a prominent model for IP based service delivery. As grid and cloud computing have become a stringent demand for today's Internet services, IaaS is required for providing services, particularly "private cloud," regardless of physical infrastructure locations. However, enabling IaaS on traditional Internet Service Provider (ISP) network infrastructures is challenging because IaaS requires a high abstraction level of network architectures, protocols, and devices. Network control plane architecture plays therefore an essential role in this transition, particularly with respect to new requirements of scalability, reliability, and flexibility. In this article we review the evolutionary trend of network element control planes from monolithic to distributed architectures according to network growth, and then present a new virtualization oriented architecture that allows infrastructure providers and service providers to achieve service delivery independently and transparently to end users based on virtualized network control planes. As a result, current ISP infrastructures will be able to support new services, such as heavy resource consuming data center applications. We also show how to use network virtualization for providing cloud computing and data center services in a flexible manner on the nationwide CANARIE network infrastructure.

The falling trend in the revenue of traditional telephony services has attracted attention to new IP based services. The IP Multimedia System (IMS) is a key architecture which provides the necessary platform for delivery of new multimedia services. However, current implementations of IMS do not offer automatic scalability or elastisity for the growing number of customers. Although the cloud computing paradigm has shown many promising characteristics for web applications, it is still failing to meet the requirements for telecommunication applications. In this paper, we present some related cloud computing patterns and discuss their adaptations for implementation of IMS or other telecommunication systems.

The IP Multimedia Subsystem (IMS) defined by the 3GPP has been mainly developed and deployed by telephony vendors on vendor-specific hardware. Recent advances in Network Function Virtualisation (NFV) technology paved the way for virtualized hardware and telephony function elasticity. As such, Telecom vendors have started to embrace the cloud as a deployment platform, usually selecting a privileged virtualization platform. Operators would like to deploy telecom functionality on their already existing IT cloud platforms. Achieving such flexibility would require the telecom vendors to adopt a software architecture allowing deployment on many cloud platforms or even heterogeneous cloud platforms. We propose a distributed software architecture enabling the deployment of a single software version on multiple cloud platforms thus allowing for a solution-based deployment. We also present a prototype we developed to study the characteristics of this architecture.

In recent years, the exponential growth of Internet users with increased bandwidth requirements has led to the emergence of the next generation of IP routers. Distributed architecture is one of the promising trends providing petabit routers with a large switching capacity and high-speed interfaces. Distributed routers are designed with an optical switch fabric interconnecting line and control cards. Computing and memory resources are available on both control and line cards to perform routing and forwarding tasks. This new hardware architecture is not efficiently utilized by the traditional software models where a single control card is responsible for all routing and management operations. The routing table manager plays an extremely critical role by managing routing information and in particular, a forwarding information table. This article presents a distributed architecture set up around a distributed and scalable routing table manager. This architecture also comes provides improvements in robustness and resiliency. The proposed architecture is based on a sharing mechanism between control and line cards and is able to meet the scalability requirements for route computations, notifications, and advertisements. A comparative scalability evaluation is made between distributed and centralized architectures in terms of required memory and computing resources.

As one of the first worldwide initiatives provisioning ICT (Information and Communication Technologies) services entirely based on renewable energy such as solar, wind and hydroelectricity across Canada and around the world, the GreenStar Network (GSN) is developed to dynamically transport user services to be processed in data centers built in proximity to green energy sources, reducing GHG (Greenhouse Gas) emissions of ICT equipments. Regarding the current approach, which focuses mainly in reducing energy consumption at the micro-level through energy efficiency improvements, the overall energy consumption will eventually increase due to the growing demand from new services and users, resulting in an increase in GHG emissions. Based on the cooperation between Mantychore FP7 and the GSN, our approach is, therefore, much broader and more appropriate because it focuses on GHG emission reductions at the macro-level. Whilst energy efficiency techniques are still encouraged at low-end client equipments, the heaviest computing services are dedicated to virtual data centers powered completely by green energy from a large abundant reserve of natural resources, particularly in northern countries.

The routing engine is the essential part of a router. As a software component, the routing engine is used to control the router activities and to build the data forwarding table. Along with the hardware evolution, several software generations for routing engines have been experienced. In this paper, we first review the different routing engine architectures over the time and their main components. In some previous papers, we have investigated new directions for routing engine development for next generation routers, where specific distributed architectures have been proposed for several routing protocols. Taking into account the increasing traffic in the core networks, we next propose a framework integrating all specific distributed routing protocol architectures in order to significantly improve the scalability of the next generation routers. Bottlenecks are reduced, resulting in improving both the overall performance and the resiliency in the presence of faults. The scalability of the proposed distributed framework is estimated with respect to each routing protocol.

The information and communications technologies (ICT) sector is seeking to reduce the electricity consumption of data processing centres. Among the initiatives to improve energy efficiency is the shift to cloud computing technology. Thanks to very favourable geographical conditions, the Canadian energy mix is highly suited to the implementation of data centres, especially in light of the significant potential of renewable energy, which can help to curb greenhouse gas emissions. In the green sustainable Telco cloud (GSTC) project, an efficient cloud computing network would be set up to optimize renewable energy use based on several data centres. This study aimed to develop a temporally differentiated life cycle assessment (LCA) model, adapted to short-term predictions, to provide a regionalized inventory to model electricity generation. Purpose of this model is (i) to calculate more accurately the carbon emissions of ICT systems and (ii) to minimize the daily carbon emissions of the GSTC servers. This paper focuses mainly on the electricity generation modelling during the use phase in the context of the life cycle assessment methodology. Considering the time scale of the model, the difference between the annual fixed average and a shorter period may be highly relevant, in particular when assessing the green house gases (GHG) emissions of a process such as an ICT system, which mainly operates during peak load hours. The time dependent grid mix modelling makes it possible to manage the server load migrations between data centres on an hourly basis. Index Terms—Life cycle assessment, data centre, carbon footprint, dynamic power mix.