Optimizing clustering algorithm in mobile ad hoc networks using simulated annealing

• Published in: 2003 IEEE Wireless Communications and Networking, 2003. WCNC 2003 Vol. 3; pp. 1492 - 1497 vol.3
• Main Authors: Turgut, D., Turgut, B., Elmasri, R., Le, T.V.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 2003
• Subjects: Ad hoc networks; Clustering algorithms; Computational modeling; Computer science; Computer simulation; Intelligent networks; Iterative algorithms; Mobile ad hoc networks; Search methods; Simulated annealing
• ISBN: 0780377001; 9780780377004
• ISSN: 1525-3511; 1558-2612
• DOI: 10.1109/WCNC.2003.1200607

In this paper, we demonstrate how simulated annealing algorithm can be applied to clustering algorithms used in ad hoc networks; specifically our recently proposed weighted clustering algorithm (WCA) is optimized by simulated annealing. As the simulated annealing stands to be a powerful stochastic search method, its usage for combinatorial optimization problems was found to be applicable in our problem domain. The problem formulation along with the parameters is mapped to be an individual solution as an input to the simulated annealing algorithm. Input consists of a random set of clusterhead set along with its members and the set of all possible dominant sets chosen from a given network of N nodes as obtained from the original WCA. Simulated annealing uses this information to find the best solution defined by computing the objective function and obtaining the best fitness value. The proposed technique is such that each clusterhead handles the maximum possible number of mobile nodes in its cluster in order to facilitate the optimal operation of the MAC protocol. Consequently, it results in the minimum number of clusters and hence clusterheads. Simulation results exhibit improved performance of the optimized WCA than the original WCA.