Energy efficient in cluster head and relay node selection for wireless sensor networks

• Main Author: Azamuddin, Ab. Rahman
• Format: Thesis
• Language: English
• Published: 2021
• Subjects: QA75 Electronic computers. Computer science
• Online Access: http://umpir.ump.edu.my/id/eprint/39217/1/ir.Energy%20efficient%20in%20cluster%20head%20and%20relay%20node%20selection%20for%20wireless%20sensor%20networks.pdf; http://umpir.ump.edu.my/id/eprint/39217/

Wireless Sensor Networks (WSNs) are defined as networks of nodes that work in a cooperative way to sense and control the surrounding environment. However, nodes contain limited energy which is the key limiting factor of the sensor network operation. In WSN architecture, the nodes are typically grouped into clusters where one node from each cluster is selected as the Cluster Head (CH) and relays utilisation to minimise energy consumption. Currently, the selection of CH based on a different combination of input variables. Example of these variables includes residual energy, communication cost, node density, mobility, cluster size and many others. Improper selection of sensor node (i.e. weak signal strength) as CH can cause an increase in energy consumption. Additionally, a direct transmission in dual-hop communication between sensor nodes (e.g. CH) with the base station (BS) uses high energy consumption. A proper selection of the relay node can assist in communication while minimising energy consumption. Therefore, the research aim is to prolong the network lifetime (i.e. reduce energy consumption) by improving the selection of CHs and relay nodes through a new combination of input variables and distance threshold approach. In CH selection, the Received Signal Strength Indicator (RSSI) scheme, residual energy, and centrality variable were proposed. Fuzzy logic was utilized in selecting the appropriate CHs based on these variables in the MATLAB. In relay node selection, the selection is based on the distance threshold according to the nearest distance with the BS. The selection of the optimal number of relay nodes is performed using K-Optimal and K-Means techniques. This ensures that all CHs are connected to at least one corresponding relay node (i.e. a 2-tier network) to execute the routing process and send the data to BS. To evaluate the proposal, the performance of Multi-Tier Protocol (MAP) and Stable Election Protocol (SEP) was compared based on 100, 200, and 800 nodes with 1 J and random energy. The simulation results showed that our proposed approach, refer to as Energy Efficient Cluster Heads and Relay Nodes (EECR) selection approach, extended the network lifetime of the wireless sensor network by 43% and 33% longer than SEP and MAP, respectively. This thesis concluded that with effective combinations of variables for CHs and relay nodes selection in static environment for data routing, EECR can effectively improve the energy efficiency of WSNs.