Multiple Bloom filters
Yuanhang Yang, Shuhui Chen · 2017
A standard technique from the cryptanalysis is to use exhaustive search that consists of systematically enumerating all possible candidates for the solution and checking whether each candidate satisfies the hash value. But this will take a lot of storage space and the time spent on query will be very long. In this paper, we introduce multiple Bloom filter, which improves performance when the password is attacked by brute-force or dictionary. Bloom filters can be used represent the password dictionary, which requires less storage space and searching time. For example, Bloom filters have been suggested as a mean for searching keywords in a huge database. In this setting, users do not search the whole of keywords, but instead query the table of a Bloom filter that represents the full set of keyword. Our goal is to recover the passwords from hash values that have been stored in or transmitted by a computer system. By using Bloom filters, we can just query a Bloom filter table instead of gigabyte dictionaries, correspondingly, the time spent on searching becomes much fewer.