Intelligent Spectrum Decision ML Algorithms for WSN

Sivaranjani, S.; Mohanraj, S.; Kavitha, Veeraruna

doi:10.17485/ijst/2018/v11i19/123224

Cited by 1 publication

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The recompenses of heuristic approach is that it always tries to provide the acceptable solutions to the given problem by using trial and error approach in a small duration whereas the metaheuristic approach like Genetic algorithm, Differential Evolution (DE), Simulation Annealing (SA) (17) are generalized and black box procedures that can be used in various areas, it always tries to provide solution that might not be the best but optimal solution for a given problem (18) . The approach used in (19) tries to use DE together with optimization of particle swarm for the purpose of node localization. The approach specified in (8) uses fuzzy to select the cluster head providing intermediate solution rather than the optimal solution as it requires entire network information, hence this approach does not suit in the environment where the nodes are deployed randomly.…”

Section: Introductionmentioning

confidence: 99%

An efficient hybrid model for cluster head selection to optimize wireless sensor network using simulated annealing algorithm

Umashankar¹,

Anitha²,

Mallikarjunaswamy³

2021

IJST

View full text Add to dashboard Cite

Objective: Energy efficiency aspect in wireless sensor networks (WSN) can be achieved by small sized rechargeable and easily replaceable batteries. The lifetime of wireless sensor network can be improved by identifying the efficient and reliable nodes as a cluster heads using Hybrid Simulated Annealing algorithm. The proposed algorithm identifies cluster head to reduce overhead and is capable of handling high volume of nodes with minimum node death rate. Methods: This study proposed initialization of population vectors using the opposite point procedure, self-adaptive control approach by node mutation rate, crossover rate, node capacity and cluster head allocation Methods. Findings: A case study in the proposed work is found to be better in throughput, accuracy, efficiency, energy utilization, batteries recharge ability and replacement procedures compared to the conventional methods. By the analysis and comparison of the proposed method with existing methods, it is identified that the reduction of the number of dead nodes gradually increases the throughput and lifetime of the nodes with respect to the number of iterations. Novelty: To overcome the limitations of conventional Low Energy Adaptive Clustering Hierarchy (LEACH), harmony search algorithm (HSA), modified HSA and differential evolution, we propose a hybrid optimal model using simulated annealing algorithm which includes a node capability function. It is used to improve the network lifetime of the cluster heads and sensor nodes. The proposed method have capability of batteries recharge ability and replacement option to improve network throughput and reliability of network.

show abstract