RESILIENCE OF NETWORK PROTOCOLS TO DEEPFAKE DETECTION TRAFFIC

Abstract

This research analyzes the robustness of current multimedia protocols in managing traffic produced by real-time, high-volume deepfake detection. The widespread occurrence of deepfakes on online platforms necessitates swift and precise detection algorithms for multimedia data, highlighting the demand for optimized network protocols. This paper assesses the existing functionalities of multimedia protocols, including RTP and RTSP, in facilitating deepfake detection and proposes improvements to enhance robustness, speed, and accuracy. The results underscore the necessary protocol modifications to accommodate real-time deepfake traffic requirements and reduce network stress.