Improved clustering techniques in wireless sensor networks

With developments in technology, ad hoc Wireless Sensor Networks are gaining prominence in monitoring and surveillance applications, especially in remote regions and in terrains which are dangerous for human intervention.An example of such an application is in controlling and monitoring equipments (such as transformers and circuit breakers) in a high voltage substation.Since the sensor networks are often employed in regions which are inaccessible, it is crucial that these networks function for long periods of time.A lot of research is focused on developing techniques to reduce energy consumption, and thereby increase the network's lifetime.An example of a technique that enables efficient energy consumption in sensor networks is clustering.The main contribution of this thesis is to propose modifications to the existing clustering algorithms (viz., LEACH and EECH) in order to reduce energy consumption, and thereby enhance the network lifetime.Simulations were performed to compare the performance of the modified algorithms (viz., LEACH-Improved and EECH-Improved) with the original algorithms.The results of these simulations show that the modifications proposed enhance the performance of the clustering algorithms.The performance metrics used for comparison are the energy consumption in the network, the amount of data successfully transmitted from the network to the end user, the lifetime of the network, and the latency in the network.In monitoring applications, latency in the network is often a crucial parameter as it is essential that the monitored parameters are transmitted to the end user with minimum delay.Further, latency is often the determining parameter in fault detection applications.An important contribution of this thesis is in providing an analysis of the latency when different algorithms are used.The factors that have an impact on the latency experienced by nodes in the network have been analyzed.Further, as part of this thesis, we have analyzed the dependency of the delay experienced by a node on its geographical location in different types of networks.The simulations show that the delay experienced by a node is dependent on the clustering algorithm, as well as its geographical position in the network.The analysis presented in this thesis can aid researchers in choosing efficient clustering algorithms for different networks.First and foremost, I wish to thank my supervisor and mentor, Prof. Fabrice Labeau for his guidance and support, as well as his belief and confidence in me.Without him, this thesis work would not have been possible.He changed my perspective on how certain problems are solved.Prof. Labeau's attitude towards his students and the lab has made working with him a remarkably pleasant and enjoyable experience.I would also like to thank the teaching and non-teaching staff at McGill University for the numerous ways in which they were a source of help and support.Thanks are also due to my lab-mates for being such good friends.I have had a great time with them on as well as off the campus.Heartfelt gratitude and thanks to my parents, sister and family for being supportive and egging me on to perform my very best.I thank my grandparents who believe I can never fail.I'll always be grateful to my family.Thanks are due to all my friends at Montreal, in particular

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.