Javier A. Barria

Communication Infrastructures for Distributed Control of Power Distribution Networks

Power distribution networks with distributed generators (DGs) can exhibit complex operational regimes which makes conventional management approaches no longer adequate. This paper looks into key communication infrastructure design aspects, and analyzes two representative evolution cases of Active Network Management (ANM) for distributed control. Relevant standard initiatives, communication protocols and technologies are introduced and underlying engineering challenges are highlighted. By analyzing two representative case networks (meshed and radial topologies) at different voltage levels (33 and 11 kV), this paper discusses the design considerations and presents performance results based on numerical simulations. This study focuses on the key role of the telecommunications provision when upgrading and deploying distributed control solutions, as part of future ANM systems.

Multiresolution FIR neural-network-based learning algorithm applied to network traffic prediction

In this paper, a multiresolution finite-impulse-response (FIR) neural-network-based learning algorithm using the maximal overlap discrete wavelet transform (MODWT) is proposed. The multiresolution learning algorithm employs the analysis framework of wavelet theory, which decomposes a signal into wavelet coefficients and scaling coefficients. The translation-invariant property of the MODWT allows alignment of events in a multiresolution analysis with respect to the original time series and, therefore, preserving the integrity of some transient events. A learning algorithm is also derived for adapting the gain of the activation functions at each level of resolution. The proposed multiresolution FIR neural-network-based learning algorithm is applied to network traffic prediction (real-world aggregate Ethernet traffic data) with comparable results. These results indicate that the generalization ability of the FIR neural network is improved by the proposed multiresolution learning algorithm.

Detection and Classification of Traffic Anomalies Using Microscopic Traffic Variables

This paper proposes a novel anomaly detection and classification algorithm that combines the spatiotemporal changes in the variability of microscopic traffic variables, namely, relative speed, intervehicle time gap, and lane changing. When applied to real-world scenarios, the proposed algorithm can use the variances of statistics of microscopic traffic variables to detect and classify traffic anomalies. Based on a simulation environment, it is shown that, with minimum prior knowledge and partial availability of microscopic traffic information from as few as 20% of the vehicle population, the proposed algorithm can still achieve 100% detection rates and very low false alarm rates, which outperforms previous algorithms monitoring loop detectors that are ideally placed at locations where anomalies originate.

Data fusion for modern engineering applications: an overview

An overview of data fusion approaches is provided from the signal processing viewpoint. The general concept of data fusion is introduced, together with the related architectures, algorithms and performance aspects. Benefits of such an approach are highlighted and potential applications are identified. Case studies illustrate the merits of applying data fusion concepts in real world applications.

Lexicographic maximin optimisation for fair bandwidth allocation in computer networks

AbstractThis paper addresses the problem of bandwidth allocation in multi-application computer network environments. Allo-cations are determined from the solution of a multiple objective optimisation problem under network constraints, wherethe lexicographic maximin criterion is applied to solve the problem and guarantees fairness and efficiency properties to thesolution. An algorithm based on a series of maximum concurrent multicommodity flow subproblems is proposed. Numer-ical results show the advantage of the approach compared to other traditional bandwidth allocation solutions. 2007 Elsevier B.V. All rights reserved. Keywords: OR in telecommunications; Bandwidth allocation; Lexicographic optimisation; Fairness; Utility theory 1. IntroductionThe performance of network applications is directly affected by the amount of available bandwidth alongend-to-end paths. Optimizing the allocation of bandwidth on network links is therefore a fundamental issuetoward the improvement of network services.In the network context, it is often possible to map the amount of allotted bandwidth to the expected level ofperformance experienced by the application. From microeconomic theory (Mas-Colell et al., 1995), such map-ping is known as the utility function associated to the application, or simply, application utility.Utility functions can be determined either qualitatively, through typical application behaviour (Shenker,1995), or quantitatively, through mean opinion scores (MOS), distortion rates (Berger, 1971) or closed-formexpressions (Breslau and Shenker, 1998; Liao and Campbell, 2001; Salles and Barria, 2004). From the userpoint of view, her personal goal may be to maximise utility. In this sense, the bandwidth allocation problemcan be generally formulated as a constrained multiple objective optimisation problem (CMOP): multiple util-ity functions to be maximised under constraints given by the limited amount of resources (link bandwidth)distributed according to the network topology.

Reinforcement Learning for Resource Allocation in LEO Satellite Networks

In this paper, we develop and assess online decision-making algorithms for call admission and routing for low Earth orbit (LEO) satellite networks. It has been shown in a recent paper that, in a LEO satellite system, a semi-Markov decision process formulation of the call admission and routing problem can achieve better performance in terms of an average revenue function than existing routing methods. However, the conventional dynamic programming (DP) numerical solution becomes prohibited as the problem size increases. In this paper, two solution methods based on reinforcement learning (RL) are proposed in order to circumvent the computational burden of DP. The first method is based on an actor-critic method with temporal-difference (TD) learning. The second method is based on a critic-only method, called optimistic TD learning. The algorithms enhance performance in terms of requirements in storage, computational complexity and computational time, and in terms of an overall long-term average revenue function that penalizes blocked calls. Numerical studies are carried out, and the results obtained show that the RL framework can achieve up to 56% higher average revenue over existing routing methods used in LEO satellite networks with reasonable storage and computational requirements

A Coxian model for channel holding time distribution for teletraffic mobility modeling

In this letter, we derive an algebraic set of equations that examines the relationships between the cell residence times and the handoff calls channel holding time. When the cell residence times have an Erlang or Hyper-Erlang distribution, the channel holding times can be represented by a Coxian model. An algorithm is presented to compute the parameters of the equivalent Coxian model. The analytical models proposed in this letter provide a flexible framework for further studies into the optimization and performance evaluation aspects of teletraffic mobile systems.

Distributed Classification of Traffic Anomalies Using Microscopic Traffic Variables

This paper proposes a novel anomaly classification algorithm that can be deployed in a distributed manner and utilizes microscopic traffic variables shared by neighboring vehicles to detect and classify traffic anomalies under different traffic conditions. The algorithm, which incorporates multiresolution concepts, is based on the likelihood estimation of a neural network output and a bisection-based decision threshold. We show that, when applied to real-world traffic scenarios, the proposed algorithm can detect all the traffic anomalies of the reference test data set; this result represents a significant improvement over our previously proposed algorithm. We also show that the proposed algorithm can effectively detect and classify traffic anomalies even when the following two cases occur: 1) the microscopic traffic variables are available from only a fraction of the vehicle population, and 2) some microscopic traffic variables are lost due to degradation in vehicle-to-vehicle (V2V) or vehicle-to-infrastructure communications (V2I).

Fair and efficient dynamic bandwidth allocation for multi-application networks

The large diversity of applications and requirements posed to current network environments make the resource allocation problem difficult to work out. This paper proposes a dynamic algorithm based on weighted fair queueing (WFQ) to promote fairness (in the Rawlsian sense) and efficiency (in the Paretian sense) in the allocation of bandwidth for multi-application networks. Utility functions are used to characterize application requirements and provide the informational basis from where the algorithm operates. Aggregation techniques are employed to ensure scalability in the network core. Simulation results confirm a significant improvement of our approach over traditional bandwidth allocation algorithms (maxmin and proportional fairness). The algorithm also provides low errors (below 10% when compared to the zero-delay centralized approach) whenever response time does not exceed 1000 times the timescale involving flow arrivals and departures.

On the Use of LEO Satellite Constellation for Active Network Management in Power Distribution Networks

The passive nature of power distribution networks has been changing to an active one in recent years as the number of small-scale Distributed Generators (DGs) connected to them rises. The consensus of recent research is that current slow central network control based upon Supervisory Control and Data Acquisition (SCADA) systems is no longer sufficient and Distributed Network Operators (DNOs) wish to adopt novel management mechanisms coupled with advanced communication infrastructures to meet the emerging control challenges. In this paper, we address this issue from the communication perspective by exploiting the effectiveness of using a Low Earth Orbit (LEO) satellite network as the key component of the underlying communication infrastructure to support a recently suggested active network management solution. The key factors that would affect the communication performance over satellite links are discussed and an analytical LEO network model is presented. The delivery performance of several major data services for supporting the management solution is evaluated against a wide range of satellite link delay and loss conditions under both normal and emergency traffic scenarios through extensive simulation experiments. Our investigation demonstrates encouraging results which suggests that a LEO network can be a viable communication solution for managing the next-generation power energy networks.