A Self-Stabilizing Method for Dynamically Redundant Autonomously Distributed Real-Time Video

This document is a Technical Note that reports on the design and proof-of-concept of a self-distributed Selective Forwarding Unit (SFU) architecture for real-time video streaming. Unlike a full research article, this note emphasizes the practical motivation, implementation details, and observed behaviors of the system. The central idea is to address the limitations of traditional prediction-based scaling by introducing a finite state machine (FSM)-based self-stabilizing method that enables autonomous scale-out and scale-in of distributed SFUs under dynamic traffic conditions.