Quality of Experience in ICN: Keep Your Low- Bitrate Close and High-Bitrate Closer

Abstract

Recent studies into streaming media delivery suggest that performance gains from ubiquitous caching in Information-Centric Networks (ICN) may be negated by Dynamic Adaptive Streaming (DAS), the de facto method for retrieving multimedia content. Bitrate adaptation mechanisms, that drive video streaming, clash with caching mechanisms in ways that affect users' Quality of Experience (QoE). Cache performance also diminishes as consumers dynamically select content encoded at different bitrates. In this paper we use this evidence to draw a novel insight: in adaptive streaming over ICN, bitrates should be prioritized alongside popularity and hit rates. We build on this insight to propose RippleCache as a family of cache placement schemes that safeguard high-bitrate content at the edge and push low-bitrate content into the network core. Doing so reduces contention of cache resources, as well as congestion in the network. To validate RippleCache claims we construct two separate implementations. We design RippleClassic as a benchmark solution that optimizes content placement by maximizing a measure for ICNs shown to have high correlation with QoE. In addition, our lighter-weight RippleFinder is then re-designed with distributed execution for application in large-scale systems. RippleCache performance gains are reinforced by evaluations in NS-3 against state-of-the-art baseline approaches, using standard measures of QoE as defined by the DASH Industry Forum. Our results demonstrate that RippleClassic and RippleFinder deliver content that suffers less oscillation and rebuffering, all while achieving the highest levels of video quality; thus indicating overall improvements to QoE.

To DASH, or Not to DASH? Optimal Video Bitrate Selection and Edge Network Caching in MEC-Empowered Slice-Enabled Networks

“…To summarize, instead of focusing on on-off scenarios where video streaming is either based on cached content [10]- [11], or on backhaul connectivity [13], we focus on the emerging scenario where specific segments of the full video file can be cached near the edge network to fully leverage dedicated resources along the e2e CDN-to-user streaming path. We also extend prior works on oscillation dynamics [12] in the context of MEC-empowered slice-enabled networks, bringing to light the impact of traditional HTTP proxy strategies. To our knowledge, we address the problem of joint video bitrate selection and edge network caching under specific storage/rate performance guarantees for the first time.…”

“…In [12], the authors discuss the problem of oscillation dynamics created by DASH in Information Centric Networks (ICNs) and propose bitrate-based partitioning of available caches (higher bitrates are stored closer to the users). Assuming that each user is served through a single video streaming path and that all video files are partitioned into the same number of segments, they formulate the cache placement problem as a binary integer linear problem (BILP) and propose distributed heuristics to solve it in polynomial time.…”

A Taxonomy for Generative Adversarial Networks in Dynamic Adaptive Streaming Over HTTP

“…Rate Adaptation [17)]: GAN architectures utilized to adaptively adjust the video bitrate based on network conditions and user preferences, ensuring smooth and uninterrupted streaming. Rate adaptation refers to the process of dynamically adjusting the video bitrate during streaming based on network conditions and user preferences.…”

Admission control and end-to-end slicing for video streaming in MEC-empowered cellular networks

“…Network-agnostic DASH has been shown to sustain and even improve the users' QoE in pre-5G network deployments [2], where the radio resource schedulers at the cellular base stations admit and allocate resources on the basis of available resource blocks (i.e., frequency, time, power and space domains) and the available backhaul capacity. Nonetheless, an increasing body of recent works highlight that applicationlayer video bitrate adaptations by the end user side performs poorly when the underlying network infrastructure leverages edge network caching [3], [4], i.e., cache hits/misses mislead the video bitrate adaptation logic on the actual network status causing "oscillation dynamics". Vice versa, performance gains of edge network caching are diminished if cache-and networkagnostic DASH is employed.…”

“…Aiming to minimize access times in Information Centric Networks (ICNs), Li et al [12] formulate the problem of caching video segments at different bitrates as an Mixed Integer Linear Program (MILP), which is solved in polynomial time using greedy heuristics. In a more recent work [4], the authors bring to light the problem of oscillation dynamics created by DASH over ICNs and propose bitratebased partitioning of available caches along the e2e caching paths, i.e., higher bitrates should be cached closer to end users (Ripple-based principle). Liotou et al [13] propose a Knapsack formulation to adapt the video bitrate selection and radio resource management (RRM) on a per user and BS basis.…”