الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 16 |  |  |
| | | from | 16 | | |
Abstract
The interest of research on Heterogeneous Wireless Sensor Networks (HWSNs) has been increasing in recent past. Energy Efficient Heterogeneous Clustered scheme (EEHC) is one of the schemes for HWSNs. Based on this scheme, a new clustering algorithm for HWSNs is proposed and it uses weighted election probabilities for the election of Cluster Heads (CHs).
The scheme is implemented along three scenarios that may send sensor remaining energy to Base Station (BS), along three different rates: none in Impl-a, every round in Impl-b, and every epoch in Impl-c. In each scenario, the proposed algorithm allows BS to elect CHs based on suggestions of requesting nodes to be CHs and it allows Sensor Nodes (SNs) to send their data directly to either BS or their nearest CHs based on distances between each node and each ofBS and CHs.
The proposed clustering algorithm aims mainly to improve network stability period, and throughput. It can be used mainly in data-sensitive applications, as nuclear safety, volcano detection, forest fires detection, and military field where amount of data is more important than quality of data in terms of redundancy removal.
The proposed algorithm was tested under three HWSN architectures: Two-Level, Three-Level, and Multi-Level, using Matlab simulator. Simulation results showed that the proposed algorithm had initially outperformed the considered common existing algorithms, which are Low-Energy Adaptive Clustering Hierarchy (LEACH), Design of distributed Energy-Efficient Clustering algorithm (DEEC), and EEHC, along the network stability period or First Nodes Died (FND), Half Nodes Died (HND) and the network throughput.