Citation Context

...er at a target level. In [17], the authors implement a different buffer-aware rate selection algorithm and experimentally measure user preferences for gradual and infrequent playback rate changes. In =-=[23]-=-, theauthors model the rate selection problem as a Markov Decision Process and use a dynamic programming technique to choose a streaming strategy that improves QoE. In [13], the authors use simulatio...

Citation Context

...ption problem of video decoding can be effectively modelled as an MDP. In [17], the authors used MDP to model the prefetching decision problem in adaptive multimedia. Xing et al. [8] and Xiang et al. =-=[7]-=- have considered MDP to optimize rate adaption of streaming video where the uncertainty in network bandwidth is modelled as a Markov chain with its own bandwidth states. The MDP formulation we used to...

Citation Context

...r and more robust signal for video rate selection. Rate Selection Algorithms. To the best of our knowledge, all previous work on client-side rate selection algorithms is based on bandwidth estimation =-=[10, 14, 9]-=-. The work most similar to ours is [12], where the authors design the algorithm as a PID controller with the buffer level as the primary feedback signal. However, their video rate selector still uses ...

Citation Context

...tested four different methods to improve the performance. In [5], how to reduce unnecessary video quality variations using a probing method to identify the effective available bandwidth was given. In =-=[4]-=-, [8], the authors designed the optimal rate adaptation algorithm for streaming scalable video coding (SVC) over HTTP using MDP. With SVC, each video frame is encoded into a base layer and several enh...

Citation Context

...ression rate and speed. Adaptive streaming can be post-processed on-the-fly, e.g. using a PI bit-rate controller in the server side as in [18] or a Markov Decision Process (MDP) based algorithm as in =-=[19]-=-. On the decoder side, using feedback is mainly motivated by energy consumption considerations, therefore considering the voltage and working frequency as control variables. For example, in [20] a fee...

Citation Context

...he dissertation and suggest the future research directions. 1.5 Bibliographic Notes Most of the work in this dissertation have appeared in research papers. The work in Chapter 2 has been published in =-=[51, 50]-=-. The work in Chapter 3 has appeared in [49]. The work in Chapter 4 is based on research paper [48]. 9Chapter 2 Dynamic Rate Adaptation for Adaptive Video Streaming in Wireless Networks In this chapte...

Citation Context

...uest frequency of contents received by each router. In contrast to described work, our proposed strategy addresses the issue of the management of cache under video delivery scenarios. The problem of video caching has been addressed in previous literature as a core challenge in the Internet under HTTP. Thus, Dynamic Adaptive Streaming over HTTP (DASH) [5] was developed to handle heterogeneous network condition. The DASH rate adaptation algorithm working on clients’ devices is designed to change requested video rate on-the-fly based on users’ available bandwidth. Therefore, some research (e.g., [6], [7]) has been devoted to that adaptation strategy, considering playback smoothness, average quality, playback interruption, etc. However, these attempts are largely HTTP based, and do not address ICN architectures with intrinsic caching capabilities. Grandl et al. in [8] pointed out the issue of caching competition between contents in different bit rates in CCN. Thus they proposed DASH-INC which only keeps the highest rate in cache and performs transcoding once a lower rate is requested. However, with the increasing demand in video traffic, one cannot simply assume that in-node processing fo...

Citation Context

... on past throughput and buffer level. Compare to them, our MDP-based adaptation makes online decisions based on expected future channel conditions and provides the optimal actions in each state. Xiang=-=[22]-=- investigated MDPbased adaptive streaming using SVC, but only layers of future segments can be requested in this work, which does not fully utilize the flexibility of SVC to improve downloaded segment...