WCA: A Weighted Clustering Algorithm for Mobile Ad Hoc Networks
Cluster Computing
Time Series Models for Internet Data Traffic
LCN '99 Proceedings of the 24th Annual IEEE Conference on Local Computer Networks
Distributed Clustering for Ad Hoc Networks
ISPAN '99 Proceedings of the 1999 International Symposium on Parallel Architectures, Algorithms and Networks
Topology properties of Ad hoc networks with random waypoint mobility
ACM SIGMOBILE Mobile Computing and Communications Review
IEEE/ACM Transactions on Networking (TON)
Temporal behavior analysis of mobile ad hoc network with different mobility patterns
Proceedings of the International Conference on Advances in Computing, Communication and Control
Time series models for internet traffic
INFOCOM'96 Proceedings of the Fifteenth annual joint conference of the IEEE computer and communications societies conference on The conference on computer communications - Volume 2
| Hi-index | 0.00 |
Clustering partitions the ad hoc network into several groups of nodes to induce a hierarchical architecture. Each group of nodes is called a cluster and is managed by a manager called cluster-head. One of the popular technique to cluster ad hoc network is based on node weights. The node weights are assigned on the basis of certain node parameters like average numbers of neighbours, sum of distances of neighbours etc. In node weight based clustering, the cluster formation and maintenance are decided by the weights of neighbouring nodes. In this article, We explore the impact of different mobility pattern on the weight based clustering algorithms. We have simulated the network using four different mobility patterns: (i) Random Way Point, (ii) Restricted Random Way Point, (iii) Gauss Markov and (iv) Random Direction mobility. We have also tried to find out the effect of average speed of nodes on clustering the network under different mobility patterns. The weights of mobile nodes are represented as a time series and modelled by Autoregressive model of order p i. e. AR(p). The order p of the model is found to lye between 1 and 3. The fitted model is then used to make prediction about the node weights. The predicted node weights are close the actual node weights as indicated by the statistical analysis.