Ad Hoc Networks: A Statistical Perspective

<DIV><BR></DIV>THIS BOOK IDENTIFIES THE TIME-DEPENDENT NETWORK PARAMETERS: NEIGHBOUR COUNT, LINK LOAD, PATH LENGTH, CLUSTER COUNT AND DELAY, AND PRESENTS A FIRST-OF-ITS-KIND DISCUSSION ON TEMPORAL PARAMETERS IN MOBILE AD HOC NETWORKS. FREQUENT TOPOLOGY CHANGES AND MULTIPLE LINK FAILURES OCCUR IN MOBILE AD HOC NETWORK DUE TO ARBITRARY AND RANDOM MOVEMENT OF NODES. THIS DYNAMIC ENVIRONMENT CHALLENGES THE DELIVERY OF DATA AND MAKES IT ESSENTIAL TO FIND BETTER MODELS FOR NETWORK PARAMETERS THAT ARE SHIFTING WITH TIME. THE PARAMETERS IDENTIFIED ARE PUT INTO THE FRAMEWORK OF TIME SERIES BECAUSE OF THEIR TEMPORAL CHARACTERISTIC, AND WHEN THEY ARE MODELLED USING TIME SERIES FRAMEWORK THEY EXHIBIT A SOUND FIT WITH AUTOREGRESSIVE AR(P) MODELS OF ORDER P. THE ORDER P IS EVALUATED FOR EACH FITTED MODEL AND FOUND TO LIE BETWEEN ONE AND THREE. THE BOOK ALSO ANALYSES THE DEPENDENCE OF END-TO-END DELAY OF AD HOC NETWORK ON VARIOUS EXTERNAL FACTORS SUCH AS THE NUMBER OF NODES, ROUTING PROTOCOL, MOBILITY MODELS AND PATH LENGTH AND DEVELOPS TWO PREDICTION MODELS. THE BOOK WILL BE USEFUL FOR RESEARCHERS AND PROFESSIONALS ALIKE.<P></P>