Musicians'-inspired clustering protocol for efficient energy Wireless Sensor Networks

A Decentralized Fuzzy C-Means-Based Energy-Efficient Routing Protocol for Wireless Sensor Networks

“…The competition between candidate nodes to be a CH is based on the following factors:

the residual energy in the candidate node;

the location of each candidate node within a cluster; and

the location of each candidate node with regards to the BS.

These factors are the main components of our proposed objective function that is used in the election process of CHs. The proposed objective function is described as follows (The early version of the proposed objective function was presented in [17]): $\begin{array}{c}{\text{CH}}_{\text{obj}}=\underset{\forall {\text{cd}}_i\in {\text{CD}}_c}{\max }\left\{\frac{E_{{\text{cd}}_i}\times q}{\alpha \times f_{\mathrm{1}}+\left(\mathrm{1}-\alpha \right)\times f_{\mathrm{2}}}\right\},\end{array}$ where $\begin{array}{c}{f}_{\mathrm{1}}=\underset{j=\mathrm{normal1}}{\overset{n_{\text{alive}}}{{\displaystyle false\sum }}}||{\text{node}}_{jc}-{\text{cd}}_i||,\\ f_{\mathrm{2}}=||{\text{cd}}_i-\text{BS}||.\end{array}$ In this objective function, E cd i is the residual energy of the candidate cluster head cd i ∈ cluster C j . q , which is set as q = 1000, is a constant term for a particular WSN and is used to avoid the objective function value from approaching zero.…”

“…These factors are the main components of our proposed objective function that is used in the election process of CHs. The proposed objective function is described as follows (The early version of the proposed objective function was presented in [ 17 ]): where In this objective function, E cd i is the residual energy of the candidate cluster head cd i ∈ cluster C j . q , which is set as q = 1000, is a constant term for a particular WSN and is used to avoid the objective function value from approaching zero.…”

Clustering-based solutions for energy efficiency, adaptability and resilience in IoT networks

“…(iv) We summarize the contributions of DeCoRIC in Section 3.4. [43,44], residual energy of nodes [45,46] or other parameters. The clustering problem was formulated as a linear programming problem in [47], representing a trade-o↵ between energy consumption and the quality of communication.…”

“…As the RSSI threshold increases, DeCoRIC marks more nodes as potential neighbors, with a mean and worst case of 13 and 25 CHs at -85 dBm; a mean and worst case of 17 and 33 CHs at -65 dBm. In comparison, most clustering schemes (including LEACH and BEEM) predefine the number of CHs to be between 5-25% (see [45,47,34,49]) of the total number of nodes with no consideration for the network structure, often resulting in disconnected clusters. The outliers in the number of CHs for DeCoRIC can be attributed to the randomness of the topologies since nodes that are farther than the radio range of the RSSI threshold become members of di↵erent clusters.…”