A Multilayered Fusion Method for SPITs Detection

SPam over Internet Telephony (SPIT) had been become a very serious issue in recent years. The threat is going to spin out from the well known email spam problem by bot nets being reprogrammed to initiate not just spam emails but also Voice over IP (VoIP) calls. Preventing SPIT is a new problem, because many well-established methods for blocking email spam cannot be applied. Though the need for SPIT prevention is urgent, SPIT prevention research is still at a very early stage. In this paper, we propose a multilayered fusion method for SPIT calls detection. Before a calls is connected, the historical signal information about the caller number is collected to judge whether it's from a suspicious caller, then during the call communication period, some acoustic parameters of the speech are calculated to judge whether it's a TTS(text to speech) voice, after the connection is released, two parameters about user behavior pattern are calculated to judge whether the call is from a machine. These three layers can work separately, but this paper proposes a fusion method to combine the information from these layers to get a more accurate decision as well.