Ad-hoc network refers to a wireless network comprising of non-stationary nodes, which autonomously collaborate between themselves for transmitting communication. Wireless sensor networks (WSN) are a subclass, where the nodes feature topological limitation in creating contact with the outside world. Replenishment of energy in WSN is usually not possible due to inhospitable environment. Accordingly, activity switching and clustering are two major approaches adopted for energy conservation and prolonging lifetime in MANETs. For large networks, soft computing has been repeatedly proven to achieve near to optimal solutions in routing and clustering problems. A genetic algorithm based approach towards efficient election of cluster heads has been proposed. The objective functions targeting optimal selection and placement of cluster centers has been derived. The problem is genetically encoded and standard multi-objective genetic algorithm is used for evaluation. Subsequent results and comparative discussion is provided.

No abstract

Enhanced mobile communication in current days demands a drastic evolution in resource management strategies to meet the quality of service (QoS) requirements for the network. Call scheduling is such an important scheme for efficient utilization in spite of scarcity of resources. In this paper, a priority driven call scheduling technique is proposed for efficient routing in mobile networks. The proposed scheduler is addressed with two conflicting objectives: to minimize the mean routing cost, and to achieve uniformity in the overall system utilization. In this context, a Multi Objective Genetic algorithm (MOGA) based approach is introduced for finding an optimal route from several alternatives pertaining to the network constraints. This results in effective scheduling and simultaneous improvement in call acceptance for the system. The proposed model takes into account the significance of priority in a comprehensive manner. The performance of the proposed model is evaluated in network scenarios with different parameters and service requirements. In addition, results from simulation studies show the quality of approximation obtained by the proposed model.

Mobile nodes depend upon mutual operation to perform data transmission in an ad hoc network. In absence of central controlling authority, the authenticity of individual nodes plays an important role. Nodes depend upon a mutual understanding for deciding the malfunctioning of any member. The reliability of this decision is vital for forwarding data within the network. Before applying any standard routing protocol, an efficient network partitioning between the reliable and malfunctioning is thus necessary. This has been approached in this work. The success rate and fault tolerance of the approach has also been discussed.

No abstract