Study and performance of localization methods in IP based networks: Vivaldi algorithm
Journal of Network and Computer Applications
| Hi-index | 0.00 |
Network coordinate algorithm provides anefficient way to predict communication latency between hosts withlimited measurements. Hosts are embedded into a coordinate space andassigned a set of coordinates which reflect their positions in thegeometric space. The distance between the coordinates of two hostscan be a predictor of the actual communication latency between them.How to improve the prediction accuracy is a problem which challengesthe practical applications of network coordinate algorithm. In orderto improve the short distance prediction accuracy of traditionalnetwork coordinate algorithm such as Vivaldi, we propose a newhierarchical network coordinate algorithm based on communitystructure detection. The whole network is organized into N-levelhierarchies without any requirement of fixed nodes asinfrastructure. Each host is assigned multiple coordinatescorresponding to the N-level hierarchies so that different sets ofcoordinates satisfy different ranges of distance prediction.Simulation results show that the new hierarchical network coordinatealgorithm provides more accurate short distance predictions withoutdegrading the accuracy of long distance predictions. Internetapplications such as the nearest server selection will benefit fromthe more accurate prediction of short distance using this newhierarchical network coordinate algorithm.