Using Parallel Bloom Filters for Multiattribute Representation on Network Services

摘要: One widely used mechanism for representing membership of a set items is the simple space-efficient randomized data structure known as Bloom filters. Yet, filters are not entirely suitable many new network applications that support services like representation and querying have multiple attributes opposed to single attribute. In this paper, we present an approach accurate efficient multiattribute using The proposes three variant structures filters: parallel filter (referred PBF) structure, PBF with hash table (PBF-HT), (PBF-BF). stores item in auxiliary HT BF provide functions capture inherent dependency all item. Compared standard represent attributes, proposed facilitates much faster query service both PBF-HT PBF-BF achieve lower false positive probability result save storage space. Simulation experimental results demonstrate can efficiently accurately quickly respond queries at cost relatively small probability.