Spam Detection and Prevention Framework Over Internet Telephony

Abstract

Spam Over IP telephony is considered to be a threat in communications and it has the potential to become even more annoying than E-mail spam because recipients will be disturbed by each received spam. So exploring the properties of spam from conversations to detect them and use these properties to prevent spam before inter- rupting users. Due to live communication nature of VOIP, the time available to filter and analyze the VOIP spam is very less as compared to the time available to detect E-mail spam. The aim of this dissertation work is to propose a framework which will able detect and prevent the spam (SIP invite flooding via unregistered user). This approach helps to improve the Quality-of-Service(QOS) of legitimate users during the SIP flooding attack. This report contains, a general overview of Spam, VOIP spam and its comparison with E-mail spam, analytical survey of existing AntiSPIT frameworks as well as study and implementation of SPIT generation(SIP invite flooding via unregistered user). The implementation of AntiSPIT framework, experimental work and analysis of results based on the Quality-of-Service parameters are discussed.