Buffer-aware ABR optimization for QoE enhancement in video streaming

In video streaming services, network dynamics are prone to causing issues such as buffering, which degrades the Quality of Experience (QoE) for users. Existing Adaptive Bitrate (ABR) algorithms have limitations: traditional methods struggle to handle buffer fluctuations, while machine learning-based approaches overlook the impact of buffer dynamics on decision-making. This paper proposes a buffer-aware ABR optimization framework that integrates LSTM-based buffer prediction with reinforcement learning for bitrate selection, incorporating a buffer-aware module to extract temporal features. Experimental results demonstrate that the proposed algorithm outperforms existing ones in terms of video bitrate, buffering time, and bitrate smoothness, thereby enhancing QoE.