Improving Live Streaming QoE Through HLS Parameter Tuning and Load Balancing to Mitigate Packet Loss

Live video streaming denotes a video distribution service that concurrently captures and transmits media material to all consumers in real time. In recent years, most of the internet has been used for video streaming, as platforms have transformed content consumption, providing immediate access to films, television programs, live events, and user-generated materials worldwide. Platforms like Twitch, YouTube, and Amazon Prime are built on technologies that facilitate efficient content delivery, adaptive playback, and personalized recommendations. Hypertext Transfer Protocol live streaming is a popular protocol for adaptive video delivery that adjusts to network bandwidth but not to packet loss, which can severely impact viewer Quality of Experience (QoE). This study addresses the challenge of maintaining live video streaming quality in environments with varying packet loss. To improve QoE, this study proposes optimizing HLS configuration parameters and evaluating the effects of two load balancing algorithms, round robin and ring hash, in a simulated testbed. The study investigates how adjusting the segment length, list length, and the group of pictures size affects the resilience of the system to packet loss, as assessed by objective evaluation metrics including peak signal-to-noise ratio and data loss percentage. Results show that the ring hash algorithm consistently outperforms round robin in reducing data loss, and with the optimal parameter configuration, data loss remained below 1.4% even under 5% network packet loss.