Search

Data-driven machine learning approaches have been proposed to facilitate wireless network optimization by learning latent knowledge from historical optimization instances. However, existing works use simplistic network representations that cannot properly encode the topological difference. They are often limited to fixed topology, and the performance is degraded because the learning target does not get sufficient information since the topological information is not well captured.To address this, we leverage the graphical neural network techniques and propose a two-stage topology-aware deep learning (TADL) framework, which trains a graph embedding unit and a link usage prediction module jointly to discover links likely to be used in optimal scheduling. By properly encoding the network structure, it makes input data with varying topology possible, and also provides more informative clues for the learning target.Important techniques are developed to ensure learning efficiency. The performance is evaluated on canonical multi-hop flow problems with diverse network structures, sizes and realistic deployment scenarios. It achieves close-to-optimum solution quality with a significant reduction in computation time without retraining. [ABSTRACT FROM AUTHOR]

While Deep Reinforcement Learning has emerged as a de facto approach to many complex experience-driven networking problems, it remains challenging to deploy DRL into real systems. Due to the random exploration or half-trained deep neural networks during the online training process, the DRL agent may make unexpected decisions, which may lead to system performance degradation or even system crash. In this paper, we propose PnP-DRL, an offline-trained, plug and play DRL solution, to leverage the batch reinforcement learning approach to learn the best control policy from pre-collected transition samples without interacting with the system. After being trained without interaction with systems, our Plug and Play DRL agent will start working seamlessly, without additional exploration or possible disruption of the running systems. We implement and evaluate our PnP-DRL solution on a prevalent experience-driven networking problem, Dynamic Adaptive Streaming over HTTP (DASH). Extensive experimental results manifest that 1) The existing batch reinforcement learning method has its limits; 2) Our approach PnP-DRL significantly outperforms classical adaptive bitrate algorithms in average user Quality of Experience (QoE); 3) PnP-DRL, unlike the state-of-the-art online DRL methods, can be off and running without learning gaps, while achieving comparable performances. [ABSTRACT FROM AUTHOR]

Grid-connected power converters that integrate frequency support and inertia emulation functionalities have emerged in the last years. These features are promising for renewable energy-based power generation plants, as it permits such systems to contribute to enhance the power system stability. This paper discusses about three different active power control solutions that can be applied to grid-connected converters equipped with a synchronous power controller. For each solution, the detailed analytical relationship between the control parameters and the power loop dynamics is illustrated, and the local stability and dynamics are analyzed. The analysis and validation of frequency support functionalities are particularly addressed in this study, where the results obtained in a 10-kW experiment endorse the analysis; finally permitting to conclude that the synchronous power control offers good performance in terms of inertial response and droop characteristics. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) is becoming truly ubiquitous in our everyday lives, but it also faces unique security challenges. Intrusion detection is critical for the security and safety of a wireless IoT network. This article discusses the human-in-theloop active learning approach for wireless intrusion detection. We first present the fundamental challenges against the design of a successful intrusion detection system for a wireless IoT network. We then briefly review the rudimentary concepts of active learning and propose its employment in the diverse applications of wireless intrusion detection. An experimental example is also presented to show the significant performance improvement of the active learning method over the traditional supervised learning approach. While machine learning (ML) techniques have been widely employed for intrusion detection, the application of human-in-the-loop ML that leverages both machine and human intelligence to intrusion detection of IoT is still in its infancy. We hope this article can assist readers in understanding the key concepts of active learning and spur further research in this area. [ABSTRACT FROM AUTHOR]

The confluence of two recent technological trends - the maturation of ROADM technology and the rise of SDN - has produced the perfect opportunity to streamline carrier metro network operations through the use of automation. This article introduces readers to the carrier metro network and describes the challenges of automating carrier metro network operations. To this end, we illustrate both the inherent complexity of today's metro network and the legacy processes that have been put in place to manage this complexity. Our discussion covers the essential details regarding current practices in metro network operation, and touches on the economic and technical factors that have contributed to these processes remaining in place for so long. We explain how the application of lean methodology can streamline and automate metro network processes - enabling carriers to reduce work in progress, thereby simultaneously decreasing both operational expenditures and time to market. [ABSTRACT FROM PUBLISHER]

The increasing amount of renewable power generation systems is a challenging issue for the control and operation of the electrical networks. One of the main issues is their lack of inertia, which is becoming a greater problem as much as the share of the power plants based on traditional synchronous generators gets reduced. In this regard, the new grid codes ask these plants to provide new functionalities such as the frequency support and inertia emulation. In this paper, a synchronous power controller for grid-connected converters is proposed as a good solution for the renewable generation systems with energy storage. It provides inertia, damping, and flexible droop characteristics. Different from the faithful replication of the swing equation of synchronous machines, an alternative control structure is proposed, by which the damping and inherent droop slope can be configured independently to meet the requirements in both dynamics and frequency regulations. Analysis and experimental results are both shown to validate the proposed controller. [ABSTRACT FROM PUBLISHER]

Device-to-device (D2D) communication has emerged as a promising technique for improving capacity and reducing power consumption in wireless networks. Most existing works on D2D communications either targeted CDMA-based single-channel networks or aimed at maximizing network throughput. In this paper, we, however, aim to enable green D2D communications in OFDMA-based wireless networks. We formally define an optimization problem based on a practical link data rate model, whose objective is to minimize total power consumption while meeting user data rate requirements. We propose solving it using a joint optimization approach by presenting two effective and efficient algorithms, which both jointly determines mode selection, channel allocation and power assignment. It has been shown by extensive simulation results that the proposed algorithms can achieve over 68% power savings, compared to several baseline methods. [ABSTRACT FROM PUBLISHER]

With the exploding popularity of wireless communication, the radio spectrum has become a scarce commodity. To further improve the network capacity, various solutions have been proposed to increase spectrum efficiency and network throughput. Small cell network is one of these new trends for next generation mobile network design. One model is using relay stations (RS) as small cell providers to achieve extended coverage, lower cost, and higher network capacity. Considering multiple related physical constraints such as channel capacity, signal to noise ratio ( SNR) requirement of subscribers, relay power and network topology, this paper studies a joint signal-aware RS placement and power allocation problem with multiple base stations in wireless relay networks. We presented approximation schemes which first find a minimum number of RS, using maximum transmission power, to cover all the subscribers meeting each SNR requirement, and then ensure communications between any subscriber to a base station by adjusting the transmission power of each RS. Numerical results are presented to confirm the theoretical analysis of our schemes, and to show strong performances of our solutions. [ABSTRACT FROM PUBLISHER]

In this paper, we propose a service-aware approach to enhance survivability in virtualized data centers. The idea is to create and maintain a Survivable Virtual Infrastructure (SVI) for each service or tenant, which includes Virtual Machines (VMs) hosting the corresponding application and their backup VMs. A fundamental problem is to determine how to map each SVI to a data center network with minimum operational costs while satisfying each VM's resource requirements and bandwidth demands between VMs before and after failures. This problem can be naturally divided into two subproblems: VM Placement (VMP) and Virtual Link Mapping (VLM). We first present a general optimization framework. Then we propose an efficient algorithm for VMP, and a polynomial-time optimal algorithm for VLM, which can be used as subroutines in the framework. We also present an effective heuristic algorithm that jointly solves two subproblems. It has been shown by extensive simulation results based on the real VM workload traces collected from Syracuse University's green data center that compared to the First Fit Decreasing (FFD) and shortest path routing based baseline algorithm, the proposed algorithms significantly reduce the reserved bandwidth, and yield comparable results in terms of the number of active servers. \beginkeywordsCloud Computing, Data Center, Service-aware, Survivability, Virtual Machine Management. \endkeywords [ABSTRACT FROM PUBLISHER]

Advances in the development of colorless and nondirectional reconfigurable optical add-drop multiplexers (ROADMs) enable flexible predeployment of optoelectronic regenerators (reshaping, retiming, and reamplifying known as 3R) in future optical networks. Compared to the current practice of installing a regenerator only when a circuit needs them, predeployment of regenerators in specific sites will allow service providers to achieve rapid provisioning such as bandwidth-on-demand service and fast restoration. Concentrating the predeployment of regenerators in a subset of ROADM sites will achieve high utilization and reduces the network operational costs. We prove the resulting optimization problem is NP-hard and provide the proof. We present an efficient heuristic for this problem that takes into account both the cost of individual circuits (regenerator cost and transmission line system cost) and the number of regenerator sites. We validate our heuristic approach with integer linear programming (ILP) formulations for a small network. Using specific network examples, we show that our heuristic has near-optimal performance under most studied scenarios and cost models. We further enhance the heuristic to incorporate the probability of demand for each circuit. This enables a reduction in the number of regenerator sites by allowing circuits to use costlier paths if they have lower probability of being needed. We also evaluate the heuristic to determine the extra regenerator sites required to support diverse routing. In this paper, we provide detailed analysis, pseudocodes, and proofs for the models presented in our previous work [Nat. Fiber Optic Engineers Conf., 2012, NW3F.6; 9th Int. Conf. on Design of Reliable Communication Networks (DRCN), 2013, 139] and compare the heuristic results with ILP for a small-scale network topology. [ABSTRACT FROM PUBLISHER]

Links in an IP network are unidirectional but currently are routed over bidirectional optical circuits with identical A ? Z and Z ? A capacities. Potentially there could be savings if the two directions of an optical circuit could be treated independently. In this study of possible savings, we have quantified the asymmetry of traffic on a current large IP backbone. A theoretical greenfield network with similarly asymmetric traffic is modeled, and it is shown that the use of unidirectional circuits to satisfy traffic demands provides significant equipment savings. We consider this study of the benefits of this approach to be a first step in deciding whether it makes economic sense to tackle the hurdles that would face such a major change in network design and operations. [ABSTRACT FROM PUBLISHER]

System state estimation and control are important issues to ensure the stability and reliability of the smart grid system. In this paper, the problem of how to securely estimate the system state and control the smart grid is studied. In the setup studied, the sensor(s) and the controller communicate with each other through a wireless channel subjected to monitoring by an eavesdropper. The channel capacity requirement that ensures negligible information leakage to the eavesdropper about the system state and control messages is studied from the information theoretic perspective. Two scenarios with single sensor or multiple sensors are studied. Numerical simulations are used to evaluate the capacity requirement in typical configurations of the smart grid. [ABSTRACT FROM PUBLISHER]

The rapidly growing demands for bandwidth-intensive mobile broadband services have triggered tremendous efforts to develop the long-term-evolution (LTE)-Advanced and beyond cellular networks, which are widely deemed as a major advancement of the existing LTE networks. Constrained by limited spectral resources, wireless communication researchers and engineers have proposed a number of technologies to efficiently utilize spectral resources in multiple dimensions. Specifically, with a large number of transmit antennas deployed at a base station, massive multiple-input multiple- output (MIMO) (also called full-dimension MIMO [FD-MIMO] in the Third Generation Partnership Project [3GPP] community) is expected to achieve enormous spectral efficiency by exploiting degrees of freedom in the spatial domain. Recent theoretical and experimental findings highlight that massive MIMO is a promising solution to improve data rates , link reliability, and power savings for cellular networks. Another attractive technology, small cells, has also shown great potential for enhancing system throughputs of LTE-Advanced and beyond cellular networks. In the small cell solution, lower-cost base stations are densely deployed, and the size of a cell is substantially reduced, thus enabling great improvement of frequency reuse ratios in the geographical domain and significant enhancement of the spectral efficiency of cellular networks. Also, other technologies such as deviceto-device (D2D) and cognitive radio have also emerged as promising solutions to boost the spectral efficiency of cellular networks in various scenarios. [ABSTRACT FROM PUBLISHER]