Odysseas Sekkas

Sensation: a middleware integration platform for pervasive applications in wireless sensor networks

In this paper we focus on the issue of application development for wireless sensor networks (WSN). Currently, such networks are extensively used in various business domains. However, their highly customized operating systems and application middleware render the application development for multiple WSNs rather cumbersome. Applications based on multiple WSNs are typical in the emerging pervasive computing paradigm adopted in numerous domains (e.g., telemedicine). A WSN can be considered as a source of information similarly to a database. Motivated by the layered driver approach introduced in the ODBC/JDBC frameworks, we propose a middleware integration architecture. Our architecture presents a unified and developer-friendly interface and abstract data model towards the application. Such interface conceals the peculiarities of the underlying WSNs as their coordination and data retrieval software is integrated in the lower layer of the proposed middleware framework. We present the design of the discussed architecture that is based on open standards like XML.

Commercial wireless sensor networks: technical and business issues

In this paper we study the opportunities of commercial exploitation of applications based on sensor networks. Such applications are quite famous nowadays in many different domains of everyday life (e.g., health monitoring, traffic monitoring). We introduce a new business model for sensor based services. Such business model involves entities like the mobile network operator and capitalizes on existing standards for application service providers (e.g., the OSA/Parlay standard). We also present the required technical framework that would facilitate the introduction and rapid adoption of the proposed business schemes.

Performance evaluation of a self-evolving trust building framework

A self-evolving reputation scheme for trust establishment in distributed peer networks is presented and evaluated. The framework, called ad-hoc trust framework (ATF), incorporates subjective behavior of end-users, direct observations of behaviors, recommendations, and history of evidences to assess the trustworthiness of peer entities. It considers several idiosyncrasies of the wireless self-organized networks, such as lack of computational resources. ATF is associated with a generic model for the evaluation of the trustworthiness of adjacent or distant nodes. It relies on a sophisticated reputation method, called trustspan, to contact only trusted peers for recommendations, and, thus, it minimizes communication costs for trust building, accelerating the trust evolution process. To evaluate the performance of the ATF framework we have deployed a large number of simulation scenarios. The performance assessment results show that ATF achieves to rapidly identify selfish nodes with high accuracy, and with relatively low communication costs.

Context Fusion: Dealing with Sensor Reliability

Context-aware applications sense, combine and reason about contextual information in order to determine and adapt to the current users context. A very important problem associated with context is the inherent ambiguity and inaccuracy. Contextual information is typically pervaded with imperfect sensing (e.g., noise of sensor readings). A novel context fusion model that represents, determines and reasons about context based on the reliability on sensor readings is proposed. This model adopts dynamic Bayesian networks and fuzzy-set theory in order to deal with the reliability of contextual data at the context inference phase.

Context Fusion through Imprecise Reasoning

Pervasive computing is an emerging computing paradigm that provides intelligent context-aware applications. Such applications handle contextual information in order to determine the current users situation. Contextual information is typically inaccurate (e.g., noise of sensor readings). A novel context fusion engine* that models, determines and reasons about the current users situation is proposed. This engine, based on dynamic Bayesian networks and fuzzy logic, deals with the reliability of sources and approximate contextual reasoning.

MNISIKLIS:Indoor Location Based Services for All

MNISIKLIS is an integrated system aiming to provide universal, indoor locationbased services focusing on navigation. This paper presents the overall MNISIKLIS architecture and certain implementation details. In the context of the Design for All approach, the system targets to the support of several types of users, including persons with disabilities as well as elderly, by exploiting multimodal interaction. Moreover, the system implements effi cient path fi nding algorithms and provides advanced user experience through highly personalized services. MNISIKLIS adopts Semantic Web technologies (e.g., ontologies and reasoning methods) for representing an d managing application models. Furthermore, MNISIKLIS exploits modern positioning techniques in order to achieve high quality positioning. The paper discusses the algorithms and the models that accommodate the services provided by the system. Additionally, an analysis of the positioning subsystem, the user interaction subsystem and the peripheral infrastructure is given. Hence, a new paradigm in the area of location-based systems is presented.

Probabilistic Information Dissemination for MANETs: The IPAC Approach

Efficient data dissemination in MANETs presents a significant challenge. Epidemic dissemination is introduced as a method to reliably spread information (context) across a network in which no direct path from source to destination can be secured. We propose a probabilistic broadcast scheme instead of the flooding technique, thus, reducing significantly the volume of message transmissions seen throughout the network. Simulation results prove the efficiency of such a scheme, which achieves full coverage of the network with disseminated context.

A Multi-level Data Fusion Approach for Early Fire Detection

Wireless Sensor Networks (WSN) allows large scale deployments for environmental monitoring applications especially in the Wild land Urban Interface (WUI) (i.e. in areas where forests and rural lands interface with homes, other buildings and infrastructures). In such areas early fire detection is of great importance as the consequences of a fire are catastrophic. Towards this direction the SCIER project envisages the deployment of Wireless Sensor Networks at the WUI using a multi-level scheme of data fusion to enhance the performance of the early fire detection process.

An adaptive epidemic information dissemination model for wireless sensor networks

We propose an adaptive bio-inspired information dissemination model that exploits the specific characteristics of the sampled/generated data stream (DS) in a wireless sensor network. Our model extends the basic epidemic algorithm by adapting key operational parameters (i.e., the forwarding probability and validity period) of the data dissemination process. The main idea is that the forwarding probability is tuned according to the variability of the involved DS. Our findings from the introduction of this adaptive epidemic are quite promising. Our scheme supersedes conventional probabilistic information dissemination algorithms in terms of efficiency and reliability.

A scheduling framework for enterprise services

Application-level scheduling is a common process in the enterprise domain, but in order to be productive it should be time-effective, interoperable and reliable. In this paper we present the design and implementation aspects of a modular scheduling solution, which aims to fulfill the above-mentioned requirements. The proposed solution constitutes an integral part of an existing service provisioning platform targeted to the location based services (LBS) domain; however application to other service-oriented architectures is also possible. A careful design process is followed in order to identify the schedulers functional subsystems and the relationships between them. The implementation is based on the J2EE framework, thus guaranteeing independence from underlying technologies and applicability in enterprise environments. The high quality and scalability of the scheduler is also validated through an intensive performance analysis.