A VoIP SPAM reduction framework

Abstract

SPAM over Internet Telephony (SPIT) has become a critical problem especially because VoIP systems are widely deployed. Identifying VoIP SPAM is more problematic compared with e-mail SPAM mainly because of the different nature of each of them. In this paper, a new approach is introduced to reduce VoIP SPAM by ranking VoIP callers based on a set of parameters. These parameters include the caller's reputation, the feedbacks (if any) collected from the callee, and whether the callee responds or ignores the call. Additionally, a set of dummy directory numbers is introduced in the callee's domain. These numbers work as traps for certain kinds of VoIP spammers who try to guess the directory numbers in the victim's domain or overwhelm this domain by dialing random directory numbers. The VoIP caller is scored based on an identification process that uses these parameters and, hence, the caller's reputation is adjusted. This score is communicated to the callee for each incoming call and the final decision to accept/reject the call is taken by the callee himself. Additionally, this score can be utilized as an acceptance/rejection probability to automatically accept//reject the incoming call. A theoretical analysis is carried out for the proposed framework. Furthermore, the effectiveness of the proposed framework in dealing with VoIP SPAM is verified by simulation studies.