Energy Savings for Cellular Network with Evaluation of Impact on Data Traffic Performance

We present a concrete methodology for saving energy in future and contemporary cellular networks. It is based on re-arranging the user-cell association so as to allow shutting down under-utilized parts of the network. We consider a hypothetical static case where we have complete knowledge of stationary user locations and thus the results represent an upper bound of potential energy savings. We formulate the problem as a binary integer programming problem, thus it is NP-hard, and we present a heuristic approximation method. We simulate the methodology on an example real cellular network topology with traffic- and user distribution generated according to recently measured patterns. Further, we evaluate the energy savings, using realistic energy profiles, and the impact on the user-perceived network performance, represented by delay and throughput, at various times of day. The general findings conclude that up to 50% energy may be saved in less busy periods, while the performance effects remain limited. We conclude that practical, real-time user-cell re-allocation methodology, taking into account user mobility predictions, may thus be feasible and bring significant energy savings at acceptable performance impact.

Cybersecurity Solutions for Active Power Distribution Networks

Teklemariam Tsegay Tesfay

An active distribution network (ADN) is an electrical-power distribution network that implements a real-time monitoring and control of the electrical resources and the grid. Effective monitoring and control is realised by deploying a large number of sensing and actuating devices and a communication network to facilitates the two-way transfer of data. The reliance of ADN operations on a large number of electronic devices and on communication networks poses a challenge in protecting the system against cyber-attacks. Identifying these challenges and commissioning appropriate solutions is of utmost importance to realize the full potential of a smart grid that seamlessly integrates distributed generation, such as renewable energy sources. As a first step, we perform a thorough threat analysis of a typical ADN. We identify potential threats against field devices, the communication infrastructure and servers at control centers. We also propose a check-list of security solutions and best practices that guarantee a distribution network's resilient operation in the presence of malicious attackers, natural disasters, and other unintended failures that could potentially lead to islanded communication zone. For the next step, we investigate the security of MPLS-TP, a technology that is mainly used for long-distance inter-domain communication in smart grid. We find that an MPLS-TP implementation in Cisco IOS has serious security vulnerabilities in two of its protocols, BFD and PSC. These two protocols control protection-switching features in MPLS-TP. In our test-bed, we demonstrate that an attacker who has physical access to the network can exploit the vulnerabilities in order to inject forged BFD or PSC messages that affect the network's availability. Third, we consider multicast source authentication for synchrophasor data communication in grid monitoring systems (GMS). Ensuring source authentication without violating the stringent real-time requirement of GMS is challenging. Through an extensive review of existing schemes, we identified a set of schemes that satisfy some desirable requirements for GMS. The identified schemes are ECDSA, TV-HORS and Incomplete- key-set. We experimentally compared these schemes using computation, communication and key management overheads as performance metrics. A tweak in ECDSA's implementation to make it use pre-generated tokens to generate signatures significantly improves the computation overhead of ECDSA, making it the preferred scheme for GMS. This finding is contrary to the generally accepted view that asymmetric cryptography is inapplicable for real-time systems. Finally, we studied a planning problem that arises when a utility wants to roll out a software patch that requires rebooting to all PMUs while maintaining system observability. The problem we address is how to find a partitioning of the set of the deployed PMUs into as few subsets as possible such that all the PMUs in one subset can be patched in one round while all the PMUs in the other subsets provide full observability. We show that the problem is NP-complete in the general case and and formulated it as binary integer linear programming (BILP) problem. We have also provided an heuristic algorithm to find an approximate solution. Furthermore, we have identified a special case of the problem where the grid is a tree and provided a polynomial-time algorithm that finds an optimal patching plan that requires only two rounds to patch the PMUs.

EPFL2017

Modeling and Measuring Performance of Data Dissemination in Opportunistic Networks

Nikodin Ristanovic

In this thesis we focus on understanding, measuring and describing the performance of Opportunistic Networks (ONs) and their applications. An “opportunistic network” is a term introduced to describe a sparse, wireless, ad hoc network with highly mobile nodes. The opportunistic networking paradigm deviates from the traditional end-to-end connectivity concept: Forwarding is based on intermittent connectivity between mobile nodes (typically, users with wireless devices); complete routes between sources and destinations rarely exist. Due to this unique property of spontaneous link establishment, the challenges that exist in ONs are specific. The unstructured nature of these networks makes it difficult to give any performance guarantees on data dissemination. For this reason, in Part I of this thesis we explore the dynamics that affect the performance of opportunistic networks. We choose a number of meaningful scenarios where our models and algorithms can be validated using large and credible data sets. We show that a drift and jump model that takes a spatial approach succeeds in capturing the impact of infrastructure and mobile-to-mobile exchanges on an opportunistic content update system. We describe the effects of these dynamics by using the age distribution of a dynamic piece of data (i.e., information updates) as the performance measure. The model also succeeds in capturing a strong bias in user mobility and reveals the existence of regions, whose statistics play a critical role in the performance perceived in the network. We exploit these findings to design an application for greedy infrastructure placement, which relies on the model approximation for a large number of nodes. Another great challenge of opportunistic networking lies in the fact that the bandwidth available on wireless links, coupled with ad hoc networking, failed to rival the capacity of backbones and to establish opportunistic networks as an alternative to infrastructure-based networks. For this reason, we never study ONs in an isolated context. Instead, we consider the applications that leverage interconnection between opportunistic networks and legacy networks and we study the benefits this synergy brings to both. Following this approach, we use a large operator-provided data set to show that opportunistic networks (based on Wi-Fi) are capable of offloading a significant amount of traffic from 3G networks. At the same time, the offloading algorithms we propose reduce the amount of energy consumed by mobiles, while requiring Wi-Fi coverage that is several times smaller than in the case of real-time offloading. Again we confirm and reuse the fact that user mobility is biased towards certain regions of the network. In Part II of this thesis, we treat another issue that is essential for the acceptance and evolution of opportunistic networks and their applications. Namely, we address the absence of experimental results that would support the findings of simulation based studies. Although the techniques such as contact-based simulations should intuitively be able to capture the performance of opportunistic applications, this intuition has little evidence in practice. For this reason, we design and deploy an experiment with real users who use an opportunistic Twitter application, in a way that allows them to maintain communication with legacy networks (i.e., cellular networks, the Internet). The experiment gives us a unique insight into certain performance aspects that are typically hidden or misinterpreted when the usual evaluation techniques (such as simulation) are used. We show that, due to the commonly ignored factors (such as the limited transmission bandwidth), contact-based simulations significantly overestimate delivery ratio and obtain delays that are several times lower than those experimentally acquired. In addition to this, our results unanimously show that the common practice of assuming infinite cache sizes in simulation studies, leads to a misinterpretation of the effects of a backbone on an opportunistic network. Such simulations typically overestimate the performance of the opportunistic component, while underestimating the utility of the backbone. Given the discovered deficiencies of the contact-based simulations, we consider an alternative statistical treatment of contact traces that uses the weighted contact graph. We show that this approach offers a better interpretation of the impact of a backbone on an opportunistic network and results in a closer match when it comes to modeling certain aspects of performance (namely, delivery ratio). Finally, the security requirements for the opportunistic applications that involve an interconnection with legacy networks are also highly specific. They cannot be fully addressed by the solutions proposed in the context of autonomous opportunistic (or ad hoc) networks, nor by the security frameworks used for securing the applications with continuous connectivity. Thus, in Part III of this thesis, we put together a security framework that fits the networks and applications that we target (i.e., the opportunistic networks and applications with occasional Internet connectivity). We then focus on the impact of security print on network performance and design a scheme for the protection of optimal relaying capacity in an opportunistic multihop network. We fine-tune the parameters of our scheme by using a game-theoretic approach and we demonstrate the substantial performance gains provided by the scheme.

EPFL2012

Learning Approach to Delineation of Curvilinear Structures in 2D and 3D Images

Agata Justyna Mosinska

Detection of curvilinear structures has long been of interest due to its wide range of applications. Large amounts of imaging data could be readily used in many fields, but it is practically not possible to analyze them manually. Hence, the need for automated delineation approaches. In the recent years Computer Vision witnessed a paradigm shift from mathematical modelling to data-driven methods based on Machine Learning. This led to improvements in performance and robustness of the detection algorithms. Nonetheless, most Machine Learning methods are general-purpose and they do not exploit the specificity of the delineation problem. In this thesis, we present learning methods suited for this task and we apply them to various kinds of microscopic and natural images, proving the general applicability of the presented solutions. First, we introduce a topology loss - a new training loss term, which captures higher-level features of curvilinear networks such as smoothness, connectivity and continuity. This is in contrast to most Deep Learning segmentation methods that do not take into account the geometry of the resulting prediction. In order to compute the new loss term, we extract topology features of prediction and ground-truth using a pre-trained network, whose filters are activated by structures at different scales and orientations. We show that this approach yields better results in terms of conventional segmentation metrics and overall topology of the resulting delineation. Although segmentation of curvilinear structures provides useful information, it is not always sufficient. In many cases, such as neuroscience and cartography, it is crucial to estimate the network connectivity. In order to find the graph representation of the structure depicted in the image, we propose an approach for joint segmentation and connection classification. Apart from pixel probabilities, this approach also returns the likelihood of a proposed path being a part of the reconstructed network. We show that segmentation and path classification are closely related tasks and can benefit from the synergy. The aforementioned methods rely on Machine Learning, which requires significant amounts of annotated ground-truth data to train models. The labelling process often requires expertise, it is costly and tiresome. To alleviate this problem, we introduce an Active Learning method that significantly decreases the time spent on annotating images. It queries the annotator only about the most informative examples, in this case the hypothetical paths belonging to the structure of interest. Contrary to conventional Active Learning methods, our approach exploits local consistency of linear paths to pick the ones that stand out from their neighborhood. Our final contribution is a method suited for both Active Learning and proofreading the result, which often requires more time than the automated delineation itself. It investigates edges of the delineation graph and determines the ones that are especially significant for the global reconstruction by perturbing their weights. Our Active Learning and proofreading strategies are combined with a new efficient formulation of an optimal subgraph computation and reduce the annotation effort by up to 80%.

EPFL2019