Sampling method based on HBF for recording more flows

For the limit of false positives probability in Bloom Filter (BF), this paper proposed a novel mechanism based on Hierarchy Bloom Filter (HBF) for recording network flows statistic. By extending the standard structure of Bloom Filter to multi-layer with the false positives probability which predefined each layer, the mechanism could not only record more flows, but also adapt dynamically when the number of flows rapidly rising. The experimental simulation results shown that compared with a single layer of Bloom Filter, the algorithm can significantly improve the accuracy of recording flows information.