ABSTRACT

Computer networks are prone to hacking, viruses and other malware; a Network Intrusion Detection System (NIDS) is needed to protect the end-user machines from threats. An effective NIDS is therefore a network security system capable of protecting the end user machines well before a threat or intruder affects. NIDS requires a space efficient data base for detection of threats in high speed conditions. A bloom filter is a space efficient randomized data structure for representing a set in order to support membership queries. These Bloom filters allow false positive results (FPR) but the space saving capability often outweigh this drawback provided the probability of FPR is controlled. Research is being done to reduce FPR by modifying the structure of bloom filters and enabling it to operate in the increasing network speeds, thus variant bloom filters are being introduced. The aim of this paper is to survey the ways in which Bloom filters have been used and modified to be used in high speed Network Intrusion Detection Systems with their merits and demerits.