Facial reconstruction

Search LJMU Research Online

Browse Repository | Browse E-Theses

A middleware framework for wireless sensor network

Javed, H (2007) A middleware framework for wireless sensor network. Doctoral thesis, Liverpool John Moores University.

[img] Text
505937.pdf - Published Version

Download (20MB)

Abstract

Advances in wireless and Micro-Electro-Mechanical Systems (MEMS) technology has given birth to a new technology field sensor networks. These new technologies along with pervasive computing have made the dream of a smart environment come true. Sensors being small and capable of sensing, processing and communicating data has opened a whole new era of applications from medicine to military and from indoors to outdoors. Sensor networks although exciting have very limited resources, for example, memory, processing power and bandwidth, with energy being the most precious resource as they are battery operated. However, these amazing devices can collaborate in order to perform a task. Due to these limitations and specific characteristics being application specific and heterogeneous there is a need to devise techniques and software which would utilize the meager resources efficiently keeping in view the unique characteristics of this network. This thesis presents a lightweight, flexible and energy-efficient middleware framework called MidWSeN which combines aspects of queries, events and context of WSN in a single system. It provides a combination of core and optional services which could be adjusted according to the resources available and specific requirements of the application. The availability of multiple copies of services distributed across the network helps in making the system robust. This middleware framework introduces a new Persistent Storage Service which saves data within the sensor network on the nodes for lifetime of the network to provide historical data. A Priority algorithm is being also presented in this thesis to ensure that enough memory is always available. A novel context enhanced aggregation has also been presented in this thesis which aggregates data with respect to context. Application management service (AMS) provides Service optimization within the network is another novel aspect of the proposed framework. To evaluate the functionality of the work presented, different parts of the framework have also been implemented. The tests and results are detailed to prove the ideas presented in the framework. The work has also been evaluated against a set of requirements and compared against existing works to indicate the novel aspects of framework. Finally some ideas are presented for the future works.

Item Type: Thesis (Doctoral)
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Computer Science
Date Deposited: 14 Mar 2017 11:48
Last Modified: 14 Mar 2017 11:48
URI: http://researchonline.ljmu.ac.uk/id/eprint/5928

Actions (login required)

View Item View Item