| Summary: | 博士 === 國立高雄應用科技大學 === 電子工程系碩士班 === 104 === As an emerging technology, a wireless sensor network (WSN) consists of spatially distributed sensor nodes to collect the selected information in the target environment. WSNs have been widely used in variety of fields from civil to military. Challenges in many WSN applications like optimal deployment, clustering topology, localization, task scheduling, security, energy aware routing, quality of service, and data aggregation and fusion arise from subtle requirements of such problems, particularly whenever the size of a sensor network is large.
Computational intelligence (CI) is a promising answer for these challenges. CI is a set of nature-inspired computational methodologies to address complex real-world problems. In CI, the mathematical model or tool of intelligence capable of inputting data, processing data, and producing results can be used to exploit the representing paralleling, generating reliable responses, and facing up high fault tolerance. Complex and dynamic environments like WSNs, CIs could bring reasonably about flexibility, autonomous behavior, and robustness against topology of medium changes, communication failures and scenario alterations.
Paradigms of CI like fuzzy logic (FL), swarm intelligence (SI), and evolutionary algorithms (EA) have been applied successfully to WSNs environments.
This dissertation intends to bridge the gap between theory and practice and attempts to learn how to analyze, redesign or improve the methodologies of CI for solving various WSN problems. In this dissertation, several improved and analyzed algorithms are parallel bat algorithm (BA), hybrid particle and bee algorithm, compact BA, diversity grey wolf optimization (GWO), etc. Beside of discussing the advantages and disadvantages of CIs over traditional methods, the results of this dissertation are reviewed briefly the selected proposed methods and compared their performances with others related methods in the literature.
The proposed methods include the proposed fuzzy logic topology for prolonging the WSN lifetime, a self-configuration chromosome genetic algorithm for global optimization the communication distances in WSN, a hybrid particles and bees (HPB) for topology control WSN problem, bat algorithm (BA) for the unequal clustering formation in WSNs, the communication strategies particles and bats for the base stations (BS) optimization in WSN, an energy-based cluster head selection algorithm (ECHA) for optimal selecting cluster head (CH) based on effective of the distances of normal node to CH and CHs to the BS.
|
|---|
