(Not) yet another policy for scalable video delivery to mobile users

Abstract

In this work, we provide a methodology to analyze optimal adaptation policies for scalable video delivery in mobile environments. Typically, download policies for adaptive video are tuned to very specific system settings. The aim of this work is not to propose a new policy, but instead to understand how the optimal policy changes according to the operating environment and the system characteristics of a mobile video client. Armed with this insight, we can design or adapt policies for SVC adaptive video delivery for a broader range of settings.Using a semi-Markov decision process (SMDP), we find optimal video retrieval policies for a single user, subject to different limits on buffer capacity and different wireless environments. We apply a decision tree classifier to the output of the SMDP to derive simple approximate policies for 55 scenarios and use these to derive high-level rules on the relationship between optimal download policy and the underlying channel settings. For example, we show that the optimal policy is more conservative in slowly varying channels, and becomes more greedy in fast changing channels, and that instantaneous channel state is relevant to the decision-making process only in a setting with a very limited buffer capacity and slow-varying channel.

A Survey of Anticipatory Mobile Networking: Context-Based Classification, Prediction Methodologies, and Optimization Techniques

“…A DTMC assumes the time the system spends in each state is equal for all states. This time depends on the prediction application and can range from a few hundred milliseconds to predict wireless channel quality [62], to tens of seconds for user mobility prediction [19], [53], to hours for Internet traffic [93]. For tractability reason, the state space is often compressed by means of simple heuristics [20], [53], [102], K-means clustering [62], [136], equal probability classification [102], and density-based clustering [136].…”

“…In particular, the authors show the proportional dependency between utility and buffer size, as well as the complexity of the two algorithms. Furthermore, a Markov model is adopted to represent different user's achievable rates [61] and channel states [62]. The transition matrix is derived empirically to minimize the number of video stalls and their duration over a 10-second horizon.…”

“…In [157], a large video receiver buffer is modeled for storing video on demand but only a small portion of the buffer is used in the optimization, while the rest of the buffer follows a heuristic download policy. [60], [62] solve this problem by increasing the duration of the time slot such that more video can be downloaded in each slot and, therefore, the buffer is filled entirely based on the optimal policy. This, in turn, comes at the cost of lower accuracy, since the assumption is that the system is static within the duration of a time slot.…”

Restless Video Bandits: Optimal SVC Streaming in a Multi-User Wireless Network

“…With this technique, any arbitrary QA mechanism can be modeled as a set of policy matrices, each representing a particular channel state. For the remainder of our analysis, we consider three different QA policies: 1) Diagonal Buffer Policy (DBP): Results from existing research [7], [8] suggest that it is optimal to pre-fetch lower layers first, and fill higher layers after. In this policy, which we call the diagonal policy, the user starts pre-fetching sub-segments from the lowest layer until the difference between the sub-segments of that layer and the one above reaches a certain pre-fetch threshold, at which point it switches to the layer above, and this continues for all layers.…”

Eliminating bandwidth estimation from adaptive video streaming in wireless networks

“…This approach is effective for video freeze avoidance as the buffer is filled with the BLs through the progressive download. Lastly, Hosseini et al [39] analyzed optimal adaptation policies using a semi-Markov decision process for scalable video delivery in mobile environments. One of their important findings is that instantaneous channel quality is only useful for making decisions in a scenario with both a very small client buffer and slowly-varying channel.…”