用量子计算优化视频流，限速下提升观看体验
BALANCE: Bitrate-Adaptive Limit-Aware Netcast Content Enhancement Utilizing QUBO and Quantum Annealing
- 将码率分配转为量子优化问题，动态约束数据上限
- 相比传统方法，在相同流量下提升用户体验评分
- 适合有数据套餐限制的用户，尤其移动端场景
在数据用量限制日益严格的背景下，如何在用户设定的数据限额内优化视频流质量仍是重大挑战。本文提出一种基于量子计算的新型框架BALANCE，通过视觉复杂度与预估数据消耗智能预选视频片段，并利用VMAF指标提升用户体验（QoE）。将码率分配问题建模为无约束二次二值优化（QUBO），对比松弛变量法与动态惩罚法（DPA），结果表明DPA在数据限额增加时持续优于前者，提供更优且符合约束的解。实验显示该方法在同等数据限制下显著提升用户满意度。
原文摘要 · Abstract (English)
In an era of increasing data cap constraints, optimizing video streaming quality while adhering to user-defined data caps remains a significant challenge. This paper introduces Bitrate-Adaptive Limit-Aware Netcast Content Enhancement (BALANCE), a novel Quantum framework aimed at addressing this issue. BALANCE intelligently pre-selects video segments based on visual complexity and anticipated data consumption, utilizing the Video Multimethod Assessment Fusion (VMAF) metric to enhance Quality of Experience (QoE). We compare our method against traditional bitrate ladders used in Adaptive Bitrate (ABR) streaming, demonstrating a notable improvement in QoE under equivalent data constraints. We compare the Slack variable approach with the Dynamic Penalization Approach (DPA) by framing the bitrate allocation problem through Quadratic Unconstrained Binary Optimization (QUBO) to effectively enforce data limits. Our results indicate that the DPA consistently outperforms the Slack Variable Method, delivering more valid and optimal solutions as data limits increase. This new quantum approach significantly enhances streaming satisfaction for users with limited data plans.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。