Evgenia F. Adamopoulou

Enhancing Channel Estimation in Cognitive Radio Systems by means of Bayesian Networks

This paper proposes enhancements to the channel(-state) estimation phase of a cognitive radio system. Cognitive radio devices have the ability to dynamically select their operating configurations, based on environment aspects, goals, profiles, preferences etc. The proposed method aims at evaluating the various candidate configurations that a cognitive transmitter may operate in, by associating a capability e.g., achievable bit-rate, with each of these configurations. It takes into account calculations of channel capacity provided by channel-state estimation information (CSI) and the sensed environment, and at the same time increases the certainty about the configuration evaluations by considering past experience and knowledge through the use of Bayesian networks. Results from comprehensive scenarios show the impact of our method on the behaviour of cognitive radio systems, whereas potential application and future work are identified.

Enhanced estimation of configuration capabilities in cognitive radio

Cognitive radio is a highly promising answer to the complexity and heterogeneity characterizing the beyond 3G wireless scenario. In this context, this article advances from the field of interference sensing to the fields of (basic) reasoning and robust reasoning. Interference sensing is concerned with the acquisition of interference related measurements for frequency bands of interest. The article describes how a cognitive radio system can reason on these measurements to obtain estimations for the capabilities of alternate configurations, especially in terms of achievable transmission capacity and coverage. Subsequently, it focuses on robust reasoning, namely, on enhancing these estimations by employing machine learning, which constitutes an important aspect of cognitive radio. Several relevant solutions are sketched and explained, with a view to providing a complete picture.

Terminal management and intelligent access selection in heterogeneous environments

This paper presents a mobile terminal architecture for devices operating in heterogeneous environments, which incorporates intelligence for supporting mobility and roaming across legacy access networks. It focuses on the structure and functionality of the proposed scheme that supports terminal-initiated and terminal-controlled access network selection in heterogeneous networks. It discusses the decomposition of the proposed Terminal Management System into separate modules, responsible for retrieving link-layer measurements from the attachment points in the terminal’s neighborhood, for handling the user’s profile and for performing intelligent access network selection. This latter function aims at independently determining the optimal local interface and attachment point through which applications can be obtained as efficiently as possible, by taking into account network status and resource availability, user preferences and service requirements.

Dynamic Backhaul Resource Allocation: An Evolutionary Game Theoretic Approach

The recently deployed 4G access technology promises to satisfy the increasing demand of bandwidth consuming applications by providing high network capacity, low latency and seamless mobility. Towards this direction, concept solutions concerning the integration of wireless and optical networks have been proposed. However, the majority of these approaches assume the conventional fixed commitment of resources to the base stations, an inefficient and costly process, especially in case the Passive Optical Network (PON) belongs to a different operator. As a result, new, more dynamic backhaul resource allocation approaches are required. In this paper, the authors study the problem of committing resources of the backhaul network to a base station by employing evolutionary game theory in order to model the interactions between the subscribers and the base station. The asymptotic stability of the proposed scheme is proven under the replicator dynamics model. Finally, the impact of time delay in the proposed scheme is also investigated.

Modelling user preferences and configuring services in B3G devices

This paper discusses a management architecture for devices operating in heterogeneous environments, that enables access network selection through terminal-controlled, preference-based mechanisms. In this domain two problems are identified, mathematically formulated and solved: Intelligent Access Selection (IAS) and Modelling and Adaptation to User Preferences (MAUP). Their objective is to compute the optimal allocation of services to access networks and quality levels, and to dynamically determine user preferences according to the usage context, respectively. A greedy algorithm is proposed for the IAS problem, while the MAUP problem is handled through the construction of a Bayesian network that allows inference and learning of profile and usage patterns. Extensive simulation results of the proposed methods and algorithms are also presented.

Service configuration and user profiling in 4G terminals

This paper presents a middleware platform for managing devices that operate in heterogeneous environments. The proposed management framework supports terminal-controlled, preference-based access network selection. Two separate problems are identified in this domain: one involving the computation of optimal allocations of services to access networks and quality levels (service configuration), and one concerning the dynamic inference of the user’s preferences, according to the usage context (user profiling). This paper includes an approach to the definition, mathematical formulation and solution of both these problems. Indicative results of the proposed solution methods are presented in the context of a real-life scenario simulating a day in the life of an ordinary user.

Providing recommendations on location-based social networks

During the last decade, in parallel with the rapid growth of mobile communications and devices, location-based social networks have met a tremendous growth with the acceptance of the public being constantly increasing. Users have access to a plethora of venues and points of interest, while they are able to share their visits to various locations along with comments and ratings about their experience (a process which is often referred to as “check-ins”). Location recommendations based on users’ needs have been a subject of interest for many researchers, while location prediction schemes have been developed in order to provide user’s possible future locations. In this paper, we present a novel method for predicting a user’s location based on machine learning techniques. In addition, following the incremental trend towards data accumulation in social networks, we introduce a clustering based prediction method in order to enhance the recommender system. For the prediction process we propose a probabilistic neural network and confirm its superior performance against two other types of neural networks, while for the clustering process we use a K-means clustering algorithm. The dataset we used was based on input from a well-known location-based social network. Prediction results can be used in order to make appropriate suggestions for venues or points of interests to users, based on their interests and social connections.

Energy-efficient routing based on vehicular consumption predictions of a mesoscopic learning model

Graphical abstractDisplay Omitted HighlightsIntroduction of a context-aware routing methodology.Ability to sense the environment and to perceive FEVs special characteristics.A rather unbiased (MPE=0.85%) and accurate (MAPE=10.56%) consumption model.Proposed consumption model vs. a reference model: MPE 20x better, MAPE 2x better.Energy efficient vs. fastest routes: save 21% energy, only 10% longer in time. This paper proposes an alternative approach for determining the most energy efficient route towards a destination. An innovative mesoscopic vehicular consumption model that is based on machine learning functionality is introduced and its application in a case study involving Fully Electric Vehicles (FEVs) is examined. The integration of this model in a routing engine especially designed for FEVs is also analyzed and a software architecture for implementing the proposed routing methodology is defined. In order to verify the robustness and the energy efficiency of this methodology, a system prototype has been developed and a series of field tests have been performed. The results of these tests are reported and significant conclusions are derived regarding the generated energy efficient routes.

Distributed web-based management framework for ambient reconfigurable services in the intelligent environment

Existing and emerging technologies in the areas of mobile computing, wireless communications/networking, sensor and control devices, context awareness, user interfaces, etc., provide the ground for the support of human activities in a certain space. More specifically, these recent advances now allow the gradual “disappearance” of computers and/or other end-user devices into the environment creating a system that can facilitate everyday living. Such an intelligent environment system offers personalised, context-aware services that can support and improve everyday life. In spite of the large number and variety of devices, networking technologies and ambient intelligence subsystems there is a lack of a framework that brings the different relevant actors together and exploits the full potential of emerging technologies to meet the requirements of an intelligent environment system, not only in the context of the home but also in the corporate and public sectors. Intelligent environments necessitate new, advanced management mechanisms. This paper presents an approach for a Distributed Web-based managementframework for ambient reconfigurable services in the intelligentenvironment (DAFNE).

Advanced Driver Assistance System Supporting Routing and Navigation for Fully Electric Vehicles

The emergence of Fully Electric Vehicles has sparkled visions of pollution- and noise free cities. However, towards this challenging end, a lot has yet to be accomplished. One of the first priorities should be placed on improving the reliability and energy efficiency of the fully electric vehicles. This paper presents a new Advanced Driver Assistance System that has been implemented, which automatically helps the driver to save more energy while on-trip, by choosing the most energy efficient routes and by providing recommendations whenever necessary. This advanced functionality is based on the collection and exploitation of experiences - through machine learning.