Quality Of Multimedia Streaming Research Articles

Reinforcement Learning Based Multipath QUIC Scheduler for Multimedia Streaming.

With the recent advances in computing devices such as smartphones and laptops, most devices are equipped with multiple network interfaces such as cellular, Wi-Fi, and Ethernet. Multipath TCP (MPTCP) has been the de facto standard for utilizing multipaths, and Multipath QUIC (MPQUIC), which is an extension of the Quick UDP Internet Connections (QUIC) protocol, has become a promising replacement due to its various advantages. The multipath scheduler, which determines the path to which each packet should be transmitted, is a key function that affects the multipath transport performance. For example, the default minRTT scheduler typically achieves good throughput, while the redundant scheduler gains low latency. While the legacy schedulers may generally give a desirable performance in some environments, however, each application renders different requirements. For example, Web applications target low latency, while video streaming applications require low jitter and high video quality. In this paper, we propose a novel MPQUIC scheduler based on deep reinforcement learning using the Deep Q-Network (DQN) that enhances the quality of multimedia streaming. Our proposal first takes into account both delay and throughput as a reward for reinforcement learning to achieve a low video chunk download time. Second, we propose a chunk manager that informs the scheduler of the video chunk information, and we also tune the learning parameters to explore new random actions adequately. Finally, we implement our new scheduler on the Linux kernel and give results using the Mininet experiments. The evaluation results show that our proposal outperforms legacy schedulers by at least 20%.

Autonomous framework simulation tools for real-time multimedia streaming over wireless ad-hoc networks

Real-time video communication has become one of the most significant applications extensively used by homogeneous/heterogeneous wireless network technologies, such as Wi-Fi, the Internet of things, the wireless sensor network (WSN), 5G, etc. This leads to enhanced deployment of multimedia streaming applications over wireless network technologies. In order to accomplish the optimal performance of real-time multimedia streaming applications over the homogeneous/heterogeneous wireless network, it is therefore necessary to develop a simulation tool-set that effectively measures the quality of service (QoS) for different multimedia streaming applications over transport layer protocols. This paper proposes an autonomous simulation tool (AST) that is entirely independent from the source code of transport layer protocols. Furthermore, the AST is integrated into NS-2 to evaluate the QoS of real-time video streaming over numerous transport layer protocols and it uses new QoS measurement tools to test the video delivery quality based on I-frames to speeds up the assessment of multimedia streaming quality and ensure high accuracy of performance metrics. The simulation results show that using the AST to simulate real-time multimedia stream results in between 13% and 36% higher delivery ratio and 150–250% less cumulative jitter delay compared with using baseline simulation tools. Also, the AST guarantees an optimal QoS performance measurements in terms of the peak signal-to-noise Ratio and visual quality of the received video.

Optimized congestion aware and rate-adaptive multimedia communication for application in fragile environment during disaster management and recovery

Delivery of multimedia content over network has always been given vital importance since its inception. Different models for content delivery have been proposed with respect to multiple factors like bandwidth, quality of content, importance, and delivery method, etc. In this paper, we propose a Hyper Text Transfer Protocol-based rate adaptive transcoding and streaming technique that maintains its multimedia streaming quality, especially when network grows congested and fragile during Disaster Management and Recovery operations. The proposed solution provides users high-quality audio and video contents even when network resources are limited and fragile. We developed a real-time test bed that can deliver high-quality streaming videos by transcoding the original video in real time to a scalable codec, which allows streaming adaptation according to network dynamics. We also performed validation tests to ensure multimedia delivery in cloud computing, functionality on our hosted website for live communication between transmitter and receiver. From our experiments, we observed that the rate adaptive transcoding and streaming can efficiently deliver information services such as live multimedia in emergency and disasters under very low bandwidth environments.

Subjective Assessment of Region of Interest-Aware Adaptive Multimedia Streaming Quality

Adaptive multimedia streaming relies on adjusting the video content's bit-rate meet network conditions in the quest to reduce packet loss and resulting video quality degradations. Current multimedia adaptation schemes uniformly adjust the compression over the entire image area. However, research has shown that user attention is focused mostly on certain image areas, denoted areas of maximum user interest (AMUI), and their interest decreases with the increase in distance to the AMUI. The region of interest-based adaptive multimedia streaming scheme (ROIAS) is introduced to perform bit-rate adaptation to network conditions by adjusting video quality relative to the AMUI location. This paper also extends ROIAS to support multiple areas of maximum user interest within the same video frame. This paper presents the performance analysis of ROIAS in terms of the impact on user perceived video quality measured using subjective video quality assessment techniques based on human subjects. The tests use a wide range of video clips, which differ in terms of spatial and temporal complexity and region of interest location and variation. A comparative evaluation of both subjective and objective video quality test results is performed and demonstrate the benefit of using ROIAS for adaptive video quality delivery.

MWAHCA: a multimedia wireless ad hoc cluster architecture.

Wireless Ad hoc networks provide a flexible and adaptable infrastructure to transport data over a great variety of environments. Recently, real-time audio and video data transmission has been increased due to the appearance of many multimedia applications. One of the major challenges is to ensure the quality of multimedia streams when they have passed through a wireless ad hoc network. It requires adapting the network architecture to the multimedia QoS requirements. In this paper we propose a new architecture to organize and manage cluster-based ad hoc networks in order to provide multimedia streams. Proposed architecture adapts the network wireless topology in order to improve the quality of audio and video transmissions. In order to achieve this goal, the architecture uses some information such as each node's capacity and the QoS parameters (bandwidth, delay, jitter, and packet loss). The architecture splits the network into clusters which are specialized in specific multimedia traffic. The real system performance study provided at the end of the paper will demonstrate the feasibility of the proposal.

A Quality-Centric TCP-Friendly Congestion Control for Multimedia Transmission

In this paper, we propose a quality-centric congestion control for multimedia streaming over wired IP networks, which we refer to as media-TCP-friendly congestion control (MTCC). Our solution adapts the sending rate to both the network condition and the application characteristics by explicitly considering the distortion impacts, delay deadlines, and interdependencies of different video packet classes. The media-aware solution is able to provide differential services for transmitting various packet classes and thereby, further improves the multimedia streaming quality compared to the conventional network-aware congestion control. We use finite-horizon Markov decision process (FHMDP) to determine the optimal congestion control policy that maximizes the long-term multimedia quality, while adhering to the horizon- <formula formulatype="inline" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex Notation="TeX">$K$</tex></formula> TCP-friendliness constraint, which ensures long-term fairness with existing TCP applications. Moreover, the proposed MTCC is able to achieve quality-based fairness among multimedia users. We derive sufficient conditions for multiple multimedia users to achieve quality-based fairness using MTCC congestion control. Note that the proposed solution only modifies the adaptation mechanism of the TCP congestion window size at the sender, without changing the design at the receiver side (i.e., each current TCP receiver can correctly receive and process MTCC streams). Our simulation results show that MTCC achieves more than 3 dB improvement in terms of PSNR over the conventional TCP congestion control approaches, with the largest improvements observed for real-time streaming applications requiring stringent playback delays.

Designing a collector overlay architecture for fault diagnosis in video networks

To prevent subscriber churn, network service providers of VoD, SDV and IPTV have a pressing need to pro-actively detect, isolate and fix outages within an access network. Network induced degradations prove to be detrimental for streaming applications. This typically leads to a poor quality of experience (QoE) for subscribers. By monitoring key functional points of the access network for traces of degradation, service providers can devise mechanisms to mitigate the problem.In this work we propose a hierarchy of exporters, collectors and ANCON (ANalysis and CONtrol) nodes that can semi-autonomously monitor, detect and isolate impairments within an access network. Exporters on the data plane gather and disseminate statistics for individual subnets, which are streamed onto “collector” nodes on an orthogonal plane. Collector nodes aggregate traffic from various exporters, and stream them onto the root of the tree (ANCON). With an even placement of exporters, root cause analysis can now take the granularity of loss rates or delay rates in individual segments or subnets of an access network. As an extension to our architecture, we show that the overlay can support instrumentations of quality evaluation for streaming video. As an example, we use a simple MOS plugin that is in part an extension of the ITU-T Erlang model to predict the quality of a video stream much before it reaches the end user. We show that our overlay can support a wide variety of quality evaluation metrics. Through extensive simulations and an implementation, we discuss issues of engineering such an overlay, isolating impairments in access networks, instrumenting MOS plugins and predicting video quality of multimedia streams in transit.

Real-Time Adaptive Content-Based Synchronization of Multimedia Streams

Traditional synchronization schemes of multimedia applications are based on temporal relationships between inter- and intrastreams. These schemes do not provide good synchronization in the presence of random delay. As a solution, this paper proposes an adaptive content-based synchronization scheme that synchronizes multimedia streams by accounting for content in addition to time. This approach to synchronization is based on the fact that having two streams sampled close in time does not always imply that these streams are close in content. The proposed scheme primary contribution is the synchronization of audio and video streams based on content. The secondary contribution is adapting the frame rate based on content decisions. Testing adaptive content-based and adaptive time-based synchronization algorithms remotely between the American University of Beirut and Michigan State University showed that the proposed method outperforms the traditional synchronization method. Objective and subjective assessment of the received video and audio quality demonstrated that the content-based scheme provides better synchronization and overall quality of multimedia streams. Although demonstrated using a video conference application, the method can be applied to any multimedia streams including nontraditional ones referred to as supermedia like control signals, haptic, and other sensory measurements. In addition, the method can be applied to synchronize more than two streams simultaneously.

SASHA—A Quality-Oriented Handover Algorithm for Multimedia Content Delivery to Mobile Users

The convergence of the existing network access technologies to a common IP-based architecture and the increase in popularity of accessing video content over the Internet makes IPTV a promising solution for media and entertainment industries. Additionally, video content delivery to the increasingly popular mobile devices over heterogeneous wireless networks makes IPTV even more appealing. However the distribution of multimedia content over heterogeneous wireless networks to mobile devices involves significant technical challenges related to mobility management and quality of service provisioning. The existing solutions do not consider quality of service as a decision making parameter for mobility management in general and handover management in particular. This paper proposes the smooth adaptive soft handover algorithm (SASHA), a novel quality-aware approach to handover based on load balancing among different networks using a comprehensive, quality of multimedia streaming (QMS), function for decision making. SASHA represents the handover management solution at the core of the more comprehensive multimedia mobility management system (M3S), a quality oriented mobility management framework for multimedia applications which maximizes user perceived quality by efficiently exploiting all available communication resources. Simulation-based testing results are presented, outlining the performance of SASHA in different mobility scenarios. The evaluation is performed for different number of nodes performing handover simultaneous and for various situations in terms of networks' overlapping area. The results shown indicate how SASHA outperforms other three mobility management solutions in terms of quality, scalability and resilience to the dynamics of the networks' overlapping area.

Objective Assessment of Region of Interest-Aware Adaptive Multimedia Streaming Quality

Adaptive multimedia streaming relies on controlled adjustment of content bitrate and consequent video quality variation in order to meet the bandwidth constraints of the communication link used for content delivery to the end-user. The values of the easy to measure network-related Quality of Service metrics have no direct relationship with the way moving images are perceived by the human viewer. Consequently variations in the video stream bitrate are not clearly linked to similar variation in the user perceived quality. This is especially true if some human visual system-based adaptation techniques are employed. As research has shown, there are certain image regions in each frame of a video sequence on which the users are more interested than in the others. This paper presents the Region of Interest-based Adaptive Scheme (ROIAS) which adjusts differently the regions within each frame of the streamed multimedia content based on the user interest in them. ROIAS is presented and discussed in terms of the adjustment algorithms employed and their impact on the human perceived video quality. Comparisons with existing approaches, including a constant quality adaptation scheme across the whole frame area, are performed employing two objective metrics which estimate user perceived video quality.

Multimedia streaming in large-scale sensor networks with mobile swarms

Sensor networking technologies have developed very rapidly in the last ten years. In many situations, high quality multimedia streams may be required for providing detailed information of the hot spots in a large scale network. With the limited capabilities of sensor node and sensor network, it is very difficult to support multimedia streams in current sensor network structure. In this paper, we propose to enhance the sensor network by deploying limited number of mobile "swarms". The swarm nodes have much higher capabilities than the sensor nodes in terms of both hardware functionalities and networking capabilities. The mobile swarms can be directed to the hot spots in the sensor network to provide detailed information of the intended area. With the help of mobile swarms, high quality of multimedia streams can be supported in the large scale sensor network without too much cost. The wireless backbone network for connecting different swarms and the routing schemes for supporting such a unified architecture is also discussed and verified via simulations.

QoS-adaptive bandwidth scheduling in continuous media streaming

In this paper, we present a QoS-adaptive admission control and resource scheduling framework for continuous media (CM) servers. The framework consists of two parts. One is a reserve-based admission control mechanism in which new streams, arriving during periods of congestion, are offered lower QoS, instead of being blocked. The other part is a scheduler for continuous media with dynamic resource allocation to achieve higher utilization than non-dynamic schedulers by effectively sharing available resources among contending streams and by reclamation which is a scheduler-initiated negotiation to reallocate resources among streams to improve overall QoS. This soft-QoS framework recognizes that CM applications can generally tolerate certain variations on QoS parameters; that is, it exploits the findings about human tolerance to degradation in quality of multimedia streams. Using our policy, we could increase the number of simultaneously running clients that could be supported and could ensure a good response ratio and better resource utilization under heavy traffic requirements. Our admission control and scheduling strategy provides three principle advantages over conventional mechanisms. First, it guarantees better total system utilization. Second, it provides better disk utilization and larger admission ratio for input CM streams, which is a major advantage. Third, it still presents acceptable play-out qualities compared to the conventional greedy admission control algorithm.