La clusterisation des reseaux sans fil ad hoc: Une approche efficace et performante

In recent years, most research is focusing on clustering approaches for Ad hoc networks because of its effectiveness in building a virtual backbone formed by a set of suitable clusterhead (CH) to guarantee the communications across clusters. Clustering has been proven to be a promising approach for mimicking the operation of the fixed infrastructure and managing the resources in Ad hoc networks. An Ad hoc network is characterized by a collection of wireless nodes which communicate with each other using high-frequency radio waves. These nodes arbitrarily and randomly change their locations and capabilities without the aid of any fixed infrastructure.;We propose a clustering algorithm in order to elect suitable nodes' representatives, i.e. CH and to store minimum topology information by reducing the propagation of routing information which facilitates the spatial reuse of resource and increase the system capacity. We use the node degree, the remaining battery power, the transmission power, and the node mobility for the CH's election. Each CH will act as a temporary base station within its zone or cluster and communicates with other CHs. Thus, packets for route finding may only spread among CHs instead of flooding among all nodes. On the other hand, the topology change information caused by movement of some nodes is limited in adjacent clusters, not in the whole network.;The clusters must adapt dynamically to the environment changes, we proposed a distributed maintenance procedure that allows managing nodes' adhesion, nodes' handoff and the reelection of the CHs. In parallel, an elaborate analytical model was proposed to estimate the quality of service parameters (local allowed saturation throughput, delay and packet error rate). Based on the results from the analytical model, we implement a new admission control algorithm in order to determine the number of members inside a cluster that can be accommodated while satisfying the constraints imposed by the current applications. In this matter, the estimated knowledge of the number of members sharing a cluster might effectively drive congestion avoidance on the CH and interclusters load-balancing to achieve better network resource utilization. The obtained results will help us to readjust the used parameters of the clustering algorithm in order to provide better maintenance and quality of service guarantees depending on the used applications.;Through numerical analysis and simulations, we have studied the performance of our model and compared it with that of other existing algorithms. The results demonstrate the superior performance of the proposed model in terms of number of clusters, number of re-affiliations, number of transitions (state change) on CH, quality of service, load balancing and scalability. We also observed how the connectivity and the stability are maximized when the number of nodes increases in presence of mobility.;Keywords: Ad hoc networks, clusters, maintenance, quality of service, scalability.