The Internet of Things (IoT) has driven the rapid development of large-scale smart streaming media services, making it one of the most important services on the Internet. The key challenge hindering the rapid development of large-scale smart streaming over IoT is how to address the dynamic and limited network bandwidth to ensure the stability and fluency of streaming media services. Some recent studies have proposed the solutions based on Scalable Video Coding (SVC) on a system called DASH to ensure the quality of streaming media services. However, in a heterogeneous network environment, the fluctuation of bandwidth may cause frequent video switching behavior of the quality levels and even video freezes. To tackle this problem, this paper proposes a dynamic cache stack algorithm with an average bandwidth restraint. With the help of a SVC flexible adjustment mechanism and the excellent regulation ability of the system DASH, the proposed algorithm provides a smooth and high-quality video playback under the available average bandwidth. At the same time, it reduces video freezes and improves the video service quality from all aspects. Compared with existing adaptive algorithms, experimental results show that the proposed algorithm has a better performance in the quality lifting speed and the average quality, in addition to the smoothness and fluency of the video playback.

在物联网（IOT）的互联网推动了大规模的智能流媒体业务的快速发展，使其成为互联网上最重要的服务之一。阻碍物联网大规模智能流媒体快速发展的关键挑战是如何解决动态有限的网络带宽，以确保流媒体服务的稳定性和流畅性。最近的一些研究提出了基于可伸缩视频编码的解决方案（SVC）在称为DASH的系统上，以确保流媒体服务的质量。但是，在异构网络环境中，带宽的波动可能会导致质量级别的频繁视频切换行为，甚至视频冻结。为了解决这个问题，本文提出了一种具有平均带宽约束的动态缓存堆栈算法。借助SVC灵活的调整机制和系统DASH的出色调节能力，该算法可在可用平均带宽下提供流畅且高质量的视频播放。同时，它从各个方面减少了视频冻结并提高了视频服务质量。与现有的自适应算法相比，