Mamba: Bringing Multi-Dimensional ABR to WebRTC

TL;DR

Mamba is a multi-dimensional ABR algorithm for WebRTC that uses multi-agent reinforcement learning to adaptively optimize encoding parameters, significantly improving user QoE in real-time video communication.

Contribution

It introduces a novel MARL-based approach with curriculum learning for coordinated bitrate, resolution, and frame rate adaptation in WebRTC.

Findings

Outperforms existing ABR algorithms in QoE metrics

Effective in both lab and real-world scenarios

Enhances video quality and reduces latency

Abstract

Contemporary real-time video communication systems, such as WebRTC, use an adaptive bitrate (ABR) algorithm to assure high-quality and low-delay services, e.g., promptly adjusting video bitrate according to the instantaneous network bandwidth. However, target bitrate decisions in the network and bitrate control in the codec are typically incoordinated and simply ignoring the effect of inappropriate resolution and frame rate settings also leads to compromised results in bitrate control, thus devastatingly deteriorating the quality of experience (QoE). To tackle these challenges, Mamba, an end-to-end multi-dimensional ABR algorithm is proposed, which utilizes multi-agent reinforcement learning (MARL) to maximize the user's QoE by adaptively and collaboratively adjusting encoding factors including the quantization parameters (QP), resolution, and frame rate based on observed states such as network conditions and video complexity information in a video conferencing system. We also introduce curriculum learning to improve the training efficiency of MARL. Both the in-lab and real-world evaluation results demonstrate the remarkable efficacy of Mamba.