Fairness-oriented multi-stream rate adaptation using scalable video coding

Device-to-device (D2D) communications have been regarded as a promising technology to meet the dramatically increasing video data demand in the 5G network. In this paper, we consider the power control problem in a multiuser video transmission system. Due to the non-convex nature of the optimization problem, it is challenging to obtain an optimal strategy. In addition, many existing solutions require instantaneous channel state information (CSI) for each link, which is hard to obtain in resource-limited wireless networks. We developed a multi-agent deep reinforcement learning-based power control method, where each agent adaptively controls its transmit power based on the observed local states. The proposed method aims to maximize the average quality of received videos of all users while satisfying the quality requirement of each user. After off-line training, the method can be distributedly implemented such that all the users can achieve their target state from any initial state. Compared with conventional optimization based approach, the proposed method is model-free, does not require CSI, and is scalable to large networks. INDEX TERMS Multiuser video transmission, multi-agent deep reinforcement learning, power control, quality of experience.

show abstract

“…The authors in [36] further simplify this model to reduce complexity by eliminating the parameter α i , i.e.,…”

Section: B Video Transmission Modelmentioning

confidence: 99%

Smart Power Control for Quality-Driven Multi-User Video Transmissions: A Deep Reinforcement Learning Approach

Zhang

Mao

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The minimum information rate in bps, F k (d e,v ), required to transmit the e-th sub-stream with the given expected distortion d e,v , is a function of the discrete values in D v . Following the approach in [12] [13] and recently extended in [14], the expected R-D relationship is modeled by using a parametric function Although this framework holds for any monotonic strictly decreasing function F v (D), in this paper we consider the following parametric model:…”

Section: Video Coding and Adaptationmentioning

confidence: 99%

“…This is obtained through a dynamic rate adaptation strategy, consisting in maximizing the overall video quality while minimizing a weighted quality difference among the different videos under minimum and maximum rate constraints. This strategy has been first proposed by the authors in [14] by addressing the problem for both empirical and semi-analytical R-D model and without considering weighted quality difference. In [14] it has been shown that, when the parametric R-D model is sufficiently accurate, it can be used to relax the multi-objective optimization problem leading to a much simpler constraint satisfaction problem.…”

Section: Adaptation Algorithmsmentioning

confidence: 99%

“…This strategy has been first proposed by the authors in [14] by addressing the problem for both empirical and semi-analytical R-D model and without considering weighted quality difference. In [14] it has been shown that, when the parametric R-D model is sufficiently accurate, it can be used to relax the multi-objective optimization problem leading to a much simpler constraint satisfaction problem. Here, we follow this approach by defining an adaptation strategy which derives the transmission rates as the solution of the following set of equations and constraints:…”

Section: Adaptation Algorithmsmentioning

confidence: 99%

“…The algorithm proposed in [14] can be suitable extended to solve the problem considered here. Let x v , y v ∈ {0, 1}, v ∈ V, be binary variables that indicate whether (1) or not (0) the two constraints F v >F min v and F v <F max v are active for the video v. We then define:…”

Section: Adaptation Algorithmsmentioning

confidence: 99%

See 2 more Smart Citations

Cross-layer optimization for m-health SVC multiple video transmission over LTE uplink

Cicalo

Mazzotti

Moretti

et al. 2013

2013 IEEE 15th International Conference on E-Health Networking, Applications and Services (Healthcom 2013)

Self Cite

View full text Add to dashboard Cite

M-health services are expected to become increasingly relevant in the management of emergency situations, enabling real-time support of remote medical experts. In this context, the transmission of health-related information from an ambulance to a remote hospital is a challenging task, due to the variability and the limitations of the mobile radio link. In particular, the transmission of multiple video streams can improve the efficacy of the tele-consultation service, but requires a large bandwidth to meet the desired quality, not always guaranteed by the mobile network. In this paper we propose a novel cross-layer adaptation strategy for multiple SVC videos delivered over a single LTE channel, which dynamically adjusts the overall transmitted throughput to meet the actual available bandwidth, while being able to provide high quality to diagnostic video sequences and lower (but fair) quality to less critical ambient videos. After having introduced a realistic LTE uplink scenario, including an advanced resource allocation strategy, we show through numerical simulations that the proposed solution is capable to achieve an optimal end-to-end video quality for both the diagnostic and the ambient videos.

show abstract

Empirical evaluation of H.264/SVC streaming in resource-constrained multihomed mobile networks

Nightingale

Wang

Grecos

2012

Multimed Tools Appl

View full text Add to dashboard Cite

Fairness-oriented multi-stream rate adaptation using scalable video coding

Cited by 12 publications

References 15 publications

Smart Power Control for Quality-Driven Multi-User Video Transmissions: A Deep Reinforcement Learning Approach

Smart Power Control for Quality-Driven Multi-User Video Transmissions: A Deep Reinforcement Learning Approach

Cross-layer optimization for m-health SVC multiple video transmission over LTE uplink

Empirical evaluation of H.264/SVC streaming in resource-constrained multihomed mobile networks

Contact Info

Product

Resources

About