· · · Institutional Repository

Current home networks are not well designed to support quality of service (QoS). Successfully employing current QoS solutions in heterogeneous environments, such as home networks, requires compatibility between the many possible solutions for the different network technologies. Often the solutions need support from all devices throughout the network. Additionally, it is often difficult to configure QoS settings properly. An alternative approach that adds some QoS support to home networks is admission control based on path quality assessment. The service delivering device decides if a new service can be admitted based on a quick path quality measurement. Currently there are no tools available for performing a fast and accurate path quality measurement in heterogeneous network paths. This report describes the design and evaluation of a path capacity measurement tool for small­scale, heterogeneous, best­effort IP networks. A new probing method is developed that obtains the bottleneck capacity in paths consisting of different link layer technologies. Besides standard IP support, we do not put any requirements on the service receiving devices. The performance of the newly developed probing method is evaluated through simulations of Ethernet/802.11b networks in OPNET Modeler©. Simulation results show that the path capacity estimation tool provides accurate (typically ±0.5 Mbps) estimates for the path capacity (typically within 5 seconds) if employed in paths with low to moderate cross traffic intensities. For higher cross traffic intensities, the performance of the tool decreases.

Traffic grooming allows efficient utilization of network capacity by aggregating several independent traffic streams into a wavelength. In addition, survivability and impairment-awareness (i.e., taking into account the effect of physical impairments) are two important issues that have gained a lot of research interest in the area of optical networks. In this paper, we consider the survivable impairmentaware traffic grooming problem in Wavelength Division Multiplexing (WDM) optical networks, where the objective is to minimize the cost of traffic grooming and regeneration. Our approach to solve this problem is shown, using data obtained from a realistic network, to significantly outperform a sequential approach, which is usually used by practitioners.

The effect of virus spreading in a telecommunication network, where a certain curing strategy is deployed, can be captured by epidemic models. In the N-intertwined model proposed and studied in [1], [2], the probability of each node to be infected depends on the curing and infection rate of its neighbors. In this paper, we consider the case where all infection rates are equal and different values of curing rates can be deployed within a given budget, in order to minimize the overall infection of the network. We investigate this difficult optimization together with a related problem where the curing budget must be minimized within a given level of network infection. Some properties of these problems are derived and several solution algorithms are proposed. These algorithms are compared on two real world network instances, while Erdos-Renyi graphs and some special graphs such as the cycle, the star, the wheel and the complete bipartite graph are also addressed.

Wavelength Division Multiplexing (WDM) optical networks offer a large amount of bandwidth using multiple, but independent wavelength channels (or lightpaths), each operating at several Gb/s. Since the traffic between users is usually only a fraction of the capacity offered by a wavelength, several independent traffic streams can be groomed together. In addition, in order to reverse the effect of noise and signal degradations (physical impairments), optical signals need to be regenerated after a certain impairment threshold is reached. We consider survivable impairment-aware traffic grooming in WDM rings, which are among the most widely deployed optical network topologies. We first show that the survivable impairment-aware traffic grooming problem, where the objective is to minimize the total cost of grooming and regeneration, is NP-hard. We then provide approximation algorithms (for uniform traffic), and efficient heuristic algorithms whose performance is shown to be close to the lower-bounds (for non-uniform traffic) both when (1) the impairment threshold can be ignored, and (2) the impairment threshold should be considered.

With increasing penetrations of renewable distributed generations (DGs) and electrified vehicles (EVs), the volatility of the renewable sources and the huge load of the EVs induce tremendous challenges for the power grid. The two technologies also have considerable synergetic potential to alleviate these challenges if they are intelligently coordinated. The aim of this paper is to investigate how the (dis)charging of EVs could be intelligently coordinated with the production of the local DGs to reduce the peak load on the power grid. We consider a neighborhood energy community that is composed of prosumer households. Three EV (dis)charging scenarios are compared: the dumb strategy where all EVs are charged for the next commute as soon as they return from the previous commute, the centralized (dis)charging strategy where the EVs are managed by a centralized scheduling unit, and the distributed (dis)charging strategy where the households autonomously schedule their EVs while coordination is achieved through providing dynamic pricing based incentives. Our simulation results show that the distributed and centralized charging strategies can reduce the peak load up to 44.9% and 75.1%, respectively, compared to the dumb charging strategy. Moreover, the relative performance of the algorithms with respect to environmental values.

The ever increasing concerns for energy security, energy efficiency, and sustainable energy is offering various challenges for the power grid. With increasing penetration of the distributed generations, the electricity power system is facing an era of prosumerization, whereby all stakeholders can autonomously produce, consume, import and/or export power. The classical power grid with top-down organization and control does not fit this dynamics, hence reorganizing the rather old architecture of the system is indispensable. In this paper, we propose a generic architecture of the smart grid that fits the new scenario based on the concepts of holons. The proposed architecture of the system is composed of autonomous prosumers that are organized bottom-up in a recursive manner involving various aggregation layers, forming a dynamically reconfigurable system. A corresponding control architecture that employs a holonic approach to simultaneously capture the autonomy of the prosumers, the recursion and the dynamic reconfiguration of the proposed system is also proposed. We extend our work by proposing a service oriented architecture (SOA) framework to support our control architecture.

After the early land rush and fast exponential growth of online social networking platforms, concerns about how data placed in online social networks may be exploited and abused have begun to appear among mainstream users. Social networking sites have responded to these new public sentiments by introducing privacy filters to their site, allowing users to specify which aspects of their profile are visible to whom. In this paper, we demonstrate that such an approach to privacy and informational self-determination is largely futile: as we form social relations and build networks with those alike us, much of who we are and what we do can be reconstructed from unhidden parts of the social graph.

Spectrum-sliced elastic optical path (SLICE) technology offers a more flexible bandwidth allocation in optical networks than wavelength division multiplexing. It allows different connections to be served via different modulation formats. However, as with any optical network, the optical signal may be susceptible to signal impairments, especially when the signal traverses over long distances. The degree of impairment may differ per modulation type, but in any case must be taken into account. If impairment levels get too high, the signal needs to be regenerated by regenerators placed selectively (due to cost considerations) inside the network. In this paper we study the impairment-aware dynamic routing and subcarrier allocation problem in translucent SLICE networks. We propose an impairment-aware routing algorithm that tries to balance traffic flows evenly across the network to reduce the blocking probability. We consider two cases, namely (1) a modulation will be selected that is used by the entire connection, and (2) the modulation can be