Sharma S. Сombined metaheuristic clustering methods for energy efficient protocols of heterogeneous wireless sensor networks

The thesis is devoted to the improvement of clustering protocols for heterogeneous wireless sensor networks (HWSN) based on the use of heuristic and metaheuristic clustering methods in order to save energy and increase the lifetime of the network. To achieve the goal the following contributions were made : the method of multipurpose quality assessment for communication protocols of heterogeneous WSN is improved which takes into account both spatial and energy characteristics of heterogeneous WSNs. A generalized model of meta-heuristic optimization was proposed. Preliminary assessment methods of the initial population of clustering was proposed, taking into account the numerical characteristic of spatial compactness of clusters based on the k-means method, which has reduced the sensitivity in selection of the initial population. The combined methods of multi-purpose meta-heuristic clustering (GA, PSO and BBO) were developed.