An intelligent fuzzy clustering approach for energy-efficient data aggregation in wireless sensor networks

Thesis Type: Doctorate

Institution Of The Thesis: Orta Doğu Teknik Üniversitesi, Faculty of Engineering, Department of Computer Engineering, Turkey

Approval Date: 2018


Consultant: ADNAN YAZICI


Technological developments have made the generation and usage of wireless sensor nodes possible. Although an individual node is capable of gathering data alone, these nodes generally cooperate to extract high-level semantic information from the sensed region. Networks consisting such nodes are referred to as Wireless Sensor Networks (WSNs). There is generally a balance between energy-efficiency and accuracy, which are two desirable but incompatible features of these networks, because of the resource-restricted nature of the utilized devices. The balance, which can also be called the trade-off, is tried to be optimized by efficient algorithms that mostly utilize manually-value-assigned parameters through trial-and-error processes. However, this assignment process nearly always fails in finding the optimum blend of parameters, renders the implementation vague and inapplicable for most cases, and generally biases the obtained result. In this dissertation, an Intelligent Fuzzy Clustering Approach for Energy-Efficient Data Aggregation in Wireless Sensor Networks is proposed. The proposed approach is a distribution-agnostic approach that runs and scales efficiently for sensor network applications. Additionally, along with the proposal, an optimization framework is utilized to tune the parameters used in the fuzzy clustering process in order to optimize the performance of a given WSN. This dissertation also includes performance comparisons and experimental evaluations of the proposal with the selected state-of-the-art algorithms. The experimental results reveal that the proposal performs better than any of the compared protocols under the same network setup considering metrics used for comparing energy-efficiency and network lifespan of the protocols. Besides, along with the proposed optimized fuzzy network clustering protocol, an empirical study on multi-modal object classification problem in wireless sensor networks is conducted in detail and obtained results are presented as well in order to corroborate the object classification accuracy aspect of the proposed protocol.