Simulated Annealing Based Neural Network For Dynamic Clustering In Wireless Sensor Network

  Vartika Sharma,  Kritika Raj Sharma,  Susheva Sharma

Wireless Sensor Networks finds its utility in several applications and has gained significant attention from researchers in recent times. Clustering forms one of the main phases for efficient implementation of transmission algorithms. There have been attempts to address this problem using static clustering but these approaches are unable to utilize several information of the nodes which can further improve the efficiency of the network. This paper proposes a novel approach for dynamic clustering in WSN using an improved Simulated Annealing based Neural Network. Clusters are formed dynamically based on the information present in the nodes. A cost function based on residual energy and distance is optimized when a threshold criterion is satisfied. Re-clustering of nodes takes place when the residual energy of battery falls below the proposed threshold. Routing is done using DSR algorithm. Simulated Annealing technique is used to optimize the weights of Neural Network in back propagation method. Results show several improvements over the static clustering techniques in terms of throughput, delay and energy.

Keywords- Dynamic Clustering, Simulated Annealing, Neural Networks, Energy Aware Clustering, Wireless Sensor Networks

Unique Identification Number - IJEDR1501009
Page Number(s) -
Pubished in - Volume 3 | Issue 1 | Jan 2015
DOI (Digital Object Identifier) -    Arni University
Publisher - IJEDR (ISSN - 2321-9939)

  Vartika Sharma,  Kritika Raj Sharma,  Susheva Sharma,   "Simulated Annealing Based Neural Network For Dynamic Clustering In Wireless Sensor Network", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.3, Issue 1, pp., Jan 2015, Available at :http://www.ijedr.org/papers/IJEDR1501009.pdf