Niki Pissinou

Film video modeling

In this paper, we present our concept and design of a general film model that represents the structural, semantic and syntactic elements of film. The purpose of this model is to serve as a basis for film video annotation and subsequent retrieval. It is informed by contemporary film theory and analysis to adequately represent the film content and its attributes. The implementation of such a model enables users to access a much faster array of multi-modal film content than is presently possible.

Data Reduction in Low Powered Wireless Sensor Networks

A Wireless sensor network (WSN) [1] is a special kind of a peer to peer network where the nodes communicate with the sink wirelessly to transmit the sensed information. In contemporary world, a WSN utilizes different technological advancements in low power communications and Very Large Scale Integration (VLSI) [16] to support functionalities of sensing, processing and communications. WSN’s have penetrated in all walks of life ranging from health care, environmental monitoring to defense related applications. One of the major challenges faced byWSN’s is that of energy conservation. Researchers all over the world have been trying their best to make sure that WSN efficiently make use of the energy resources and increase their life span. In this Chapter one such technique for energy conservation called ’Data Reduction’ is discussed.

A Framework for Temporal Object Databases

The 3 Dimensional Information Space (3DIS) is a rich, extensible, object database model for information management. Here, we concentrate on extending the 3DIS to incorporate the semantics of time, and to support temporal data and the temporal evolution of data. We present a unique approach towards the synthesis of an integrated 3DIS model that supports the temporal aspects of data modeling in addition to the structural and dynamic ones, while preserving its original three dimensional geometric space features. This research presents a step towards defining the concepts and techniques for incorporating time in object databases, and provides a concrete experimental framework for demonstration thereof.

A parallel algorithm for the Steiner tree problem

The Steiner problem in graphs is to find the minimum cost tree which spans at least a given subset of V, where V is the set of vertices in the graph. This problem has a wide variety of practical applications, including wire routing in physical VLSI design. Since the problem is NP-complete, much work has been devoted to developing approximate solutions such as the one proposed by V.J. Rayward-Smith (1988). In this paper, we identify the Rayward-Smiths algorithm as being appropriate for parallelization, and describe a parallel implementation of that algorithm. We also show that linear speedup is achievable on the PRAM model with the number of processors p/spl les/|V|.<<ETX>>