一种基于拥塞发现的强化学习P2P网络视频点播预取策略

在P2P(Peer to Peer)网络视频点播中设计合理的视频内容预取策略,可以有效地解决视频观看中出现的缓冲等待时间过长、服务器负载过大等问题。而现有的P2P网络视频预取方法往往考虑了视频内容的发现而忽略了节点本身的状态监控,在网络拥塞发生时存在视频播放观看不佳的情况。提出了一种基于拥塞发现的强化学习P2P网络视频点播预取策略,通过监测节点的拥塞状况以及带宽等参数,采用Q-learning学习算法综合评估网络节点,引导视频预取节点的选择,减少了对网络拥塞节点视频内容的预取。实验结果表明,该方法能提高视频播放流畅性,避免用户在节点发生拥塞时等待时间过长,提高了视频播放效率。     

Video Packet Selection and Scheduling for Multipath Streaming

This paper addresses the problem of choosing the best streaming policy for distortion optimal multipath video delivery, under network bandwidth and playback delay constraints. The streaming policy consists in a joint selection of the network path and of the video packets to be transmitted, along with their sending time. A simple streaming model is introduced, which takes into account the video packet importance, and the dependencies between packets. A careful timing analysis allows to compute the quality perceived by the receiver for a constrained playback delay, as a function of the streaming policy. We derive an optimization problem based on a video abstraction model, under the assumption that the server knows, or can predict accurately the state of the network. A detailed analysis of constrained multipath streaming systems provides helpful insights to design an efficient branch and bound algorithm that finds the optimal streaming strategy. This solution allows to bound the performance of any scheduling strategy, but the complexity of the algorithm becomes rapidly intractable. We therefore propose a fast heuristic-based algorithm, built on load-balancing principles. It allows to reach close to optimal performance with a polynomial time complexity. The algorithm is then adapted to live streaming scenarios, where the server has only a partial knowledge of the packet stream, and the channel bandwidth. Extensive simulations show that the proposed algorithm only induces a negligible distortion penalty compared to the optimal strategy, even when the optimization horizon is limited, or the rate estimation is not perfect. Simulation results also demonstrate that the proposed scheduling solution performs better than common scheduling algorithms, and therefore represents a very efficient low-complexity multipath streaming algorithm, for both stored and live video services     

Reducing channel zapping time in live TV broadcasting over content centric networks

Channel zapping delay is a big challenge in delivering TV service over the Internet infrastructure. Previous research works have studied this delay, its components, and solutions to decrease it. Unfortunately, the best proposed solutions reduce the delay at the expense of increasing bandwidth usage or decreasing the received video quality. After channel switching, the Set Top Box (STB) or player application should buffer sufficient frames before starting to play the received video. However, the buffering process takes place at the playback rate and leads to a delay which is inversely related to the buffer duration. Regarding Information Centric Networking (ICN) paradigm, this paper introduces a new channel zapping protocol that aims to remove the synchronization and buffering delays while maintaining the bandwidth utilization and also the received video quality. The general idea of the proposed solution is to exploit the in-network caching feature of the ICN to retrieve the frames from the network at the network speed. Although the analyses show that the proposed zapping protocol eliminates the delay dependency to the buffer duration, network throughput becomes the bottleneck instead. So, novel solutions have been proposed to reduce the queuing delay as the main component of network delay. These solutions include two new caching algorithms, a new cache replacement algorithm, and applying scheduling methods to the forwarding queues. Simulation results show that increasing link rates, using the proposed caching and cache replacement algorithms, and applying an appropriate scheduling method will greatly reduce the zapping delay without sacrificing the bandwidth or video quality.     

Media streaming via TFRC: An analytical study of the impact of TFRC on user-perceived media quality

TCP-Friendly Rate Control (TFRC) is being adopted in Internet standards for congestion control of streaming media applications. In this paper, we consider the transmission of prerecorded media from a server to a client by using TFRC, and analytically study the impact of TFRC on user-perceived media quality, which is roughly measured by calculating the rebuffering probability. A rebuffering probability is defined to be the probability that the total duration of all rebuffering events experienced by a user is longer than a certain threshold. Several approaches are presented to help an application determine an appropriate initial buffering delay and media playback rate in order to achieve a certain rebuffering probability under a given network condition. First, we derive a closed-form expression to approximate the average TFRC sending rate, which could be used as the maximum allowed playback rate of a media stream. Second, we develop a queueing model for a TFRC client buffer with the traffic described by a Markov-Renewal-Modulated Deterministic Process (MRMDP), which captures the fundamental behavior of TFRC that predicts the immediate future TCP sending rate based on the history of past loss intervals. We present a closed-form solution and a more accurate iterative method to solve the queueing model and calculate the rebuffering probability.     

Adaptive topology formation for peer-to-peer video streaming

In this paper we propose an adaptive P2P video streaming framework to address the challenges due to bandwidth heterogeneity and peer churn on the Internet. This adaptive streaming framework consists of two major components, source rate adaptation and adaptive overlay topology formation, to maximize the video quality and fully utilize the overall peer upload capacity. In the source rate adaptation, the video server adapts the video source rate automatically based on the local measurement of peers’ download rates, so that the P2P network is not overloaded beyond its bandwidth capacity and peers are able to achieve smooth video playback. To combat bandwidth heterogeneity, we propose to construct a desirable link-level homogeneous overlay topology using a Markov chain Monte Carlo method, so that peers achieve an equal per-connection upload/download bandwidth. In this link-level homogeneous network, video flows do not encounter any bottlenecks along the delivery paths, and peers achieve high download rates to ensure smooth video playback. We also design a fully distributed algorithm to implement the dual mechanisms of the adaptive topology formation and the source rate maximization. To evaluate the performance of our streaming framework, we conduct both mathematical analysis and extensive simulations. The simulation results confirm our analysis and show that the proposed distributed algorithm is able to maximize the video playback quality with fast convergence.     

基于视点预测的全景视频码率自适应策略

为解决全景视频传输中存在的视频卡顿多、用户体验质量（quality of experience,QoE）低等问题,研究当前主流的视点自适应传输方案,提出一种基于视点预测的码率自适应策略（VPBAS）。首先,构建了一种基于长短期记忆网络和全卷积网络的视点预测模型,模型将视点数据和视频显著性信息进行特征融合,实现不同模态数据的相互补充和修正,提高视点预测的准确率;然后,客户端采用随机森林算法预测当前的可用带宽,并根据视点预测结果和可用带宽信息为视频分块选择码率。最后,客户端把选择的码率信息定期发送给服务器,服务器根据反馈的信息向客户端推送最佳码率的全景视频流,这种交互过程在视频播放期间不断地重复,直至客户端观看完毕。实验结果表明,与现有传输方案相比,VPBAS能有效提高带宽受限情况下的视频观看体验。     

改进的遗传算法无线视频传感器网络协作路由技术研究

提出了适用于无线视频传感器网络的基于能量感知的跨层交互多路径协作路由技术.该技术首先采用了基于视频传感器节点感知距离的遗传优化算法,预测传输视频数据的能耗和剩余能量,结合无线信道质量和视频编码算法建立一种跨层协同的工作体系,优化节点传输视频数据的能耗、时延和带宽等因素;然后建立应用层、网络层和物理层跨层协同工作体系.仿真实验和数学分析表明,该技术不仅能够较好地满足视频传感器网络应用业务的多样性QoS数据传输性能需求,而且可以充分利用视频传感器网络受限的计算、存储能力和能量等资源.     

Performance evaluation of smoothing algorithms for transmittingprerecorded variable-bit-rate video

The transfer of prerecorded, compressed variable-bit-rate video requires multimedia services to support large fluctuations in bandwidth requirements on multiple time scales. Bandwidth smoothing techniques can reduce the burstiness of a variable-bit-rate stream by transmitting data at a series of fixed rates, simplifying the allocation of resources in video servers and the communication network. This paper compares the transmission schedules generated by the various smoothing algorithms, based on a collection of metrics that relate directly to the server, network, and client resources necessary for the transmission, transport, and playback of prerecorded video. Using MPEG-1 and MJPEG video data and a range of client buffer sizes, we investigate the interplay between the performance metrics and the smoothing algorithms. The results highlight the unique strengths and weaknesses of each bandwidth smoothing algorithm, as well as the characteristics of a diverse set of video clips     

An optimal bandwidth allocation strategy for the delivery of compressed prerecorded video

The transportation of prerecorded, compressed video data without loss of picture quality requires the network and video servers to support large fluctuations in bandwidth requirements. Fully utilizing a client-side buffer for smoothing bandwidth requirements can limit the fluctuations in bandwidth required from the underlying network and the video-on-demand servers. This paper shows that, for a fixed-size buffer constraint, the critical bandwidth allocation technique results in plans for continuous playback of stored video that have (1) the minimum number of bandwidth increases, (2) the smallest peak bandwidth requirements, and (3) the largest minimum bandwidth requirements. In addition, this paper introduces an optimal bandwidth allocation algorithm which, in addition to the three critical bandwidth allocation properties, minimizes the total number of bandwidth changes necessary for continuous playback. A comparison between the optimal bandwidth allocation algorithm and other critical bandwidth-based algorithms using 17 full-length movie videos and 3 seminar videos is also presented.     

Low cost pre-stored video transmission across networks

This paper proposes a simple, practical and low-cost dynamic segmentation and bandwidth allocation scheme, called middle point, to transmit pre-stored video from source to playback destination across networks. The proposed scheme is based on two concepts, playback tunnel and bandwidth tunnel, to determine the segmentation boundary and transmission bandwidth with the objective of reducing transmission cost while guaranteeing the playback buffer neither underflows nor overflows. The performance of the proposed scheme is evaluated by a set of real-life MPEG video traces and also compared with the well-known optimal smoothing scheme. The obtained results show that the proposed scheme can be easily implemented with low complexity and low transmission cost.     

嵌入式多通道无线视频传输的码率自适应算法

无线视频传输和视频压缩技术是当前众多物联网（IoT）应用和嵌入式系统的基础和核心。而在多通道传输时无线网络状态的不断变化,会导致视频丢帧和延时抖动问题。虽然自适应算法能够在一定程度上解决在PC或服务器平台下的视频传输问题,但在嵌入式平台和无线网络下仍不能满足实时性和服务质量（QoS）要求。为此,基于DM368芯片设计了一从视频采集、压缩、WiFi传输、控制单元接收到上位机显示的完整平台。同时充分考虑了嵌入式平台的特点,提出一种结合信号质量、网络带宽、缓存状态和拥塞控制的码率自适应算法。该算法利用高斯函数统计网络带宽,使用分段反比例函数调整缓存状态,利用加权移动法对码率进行平滑,并使用极值抑制法进行码率均衡。该算法实现了码率的平滑调整,并被应用于所提平台来实现控制单元对多个WiFi相机的管理、多通道传输和负载均衡。以QoS为评价指标进行实验验证,结果表明：该算法在设计的嵌入式平台上性能良好,平滑性和缓存稳定性都有很大提升,多通道状态下的公平性和带宽利用率也有显著提高。在单相机信号质量动态变化或多相机竞争带宽等多种情况下,相对于MDI（McGinely Dynamic Indicator）算法,该算法的平滑性提升了16%~59%;相对于BBA（Buffer-Based Algorithm）,该算法的缓存抖动降低了15%~72%,时延抖动降低了12%~76%。     

基于MDP的移动流媒体能耗优化方法

针对移动流媒体应用的高能耗问题,面向异构平台环境,提出了基于马尔科夫决策过程的移动流媒体能耗优化方法。通过综合不同用户使用习惯、视频播放功耗、播放时长及当前网络状态,基于最大奖励原则对CPU资源进行合理分配,以达到降低功耗的目标。实验结果表明,相比于系统默认异核多处理调度策略,提出的能耗优化方法能在保持流媒体视频播放质量的同时,有效降低其在移动终端的能耗。     

P2P网络多视点立体视频传输的数据块调度策略

立体视频数据量大、编码结构复杂,因此视频数据块的重要性和紧急程度对视频的播放质量有更大的影响作用。综合考虑立体视频播放质量和网络传输效率,提出一种联合优先级和随机传输的P2P数据块调度算法。依据数据块的重要性、紧急性和稀有性计算传输优先级别,同时评估邻居节点传输能力,利用多余带宽传输随机数据块。该算法能够提供优越的视频播放质量,并能够使数据块均匀分布,保证系统传输性能。仿真实验通过分析视频PSNR、启动延迟、节点吞吐量和数据到达率等指标,证明了该算法的有效性。     

基于画面组单元时序控制算法的H.264实时流关键帧预处理

网络带宽特别是上行带宽受限情况下,基于H.264高压缩率的网络音视频通话或视频会议普遍会出现不同程度的丢包现象,从而造成流媒体播放的花屏等质量问题,影响视频通话或视频会议的效果。针对上述问题,提出了基于关键帧预处理的实时流媒体播放质量控制方法。该方法采用单元时序控制算法对视频关键坏帧进行实时侦测和取舍,进而减少花屏情况。使用该方法将降低后处理计算的时空成本,并提高流媒体播放的流畅性。通过原始帧播放、后处理播放、关键帧预处理播放3种处理方法的实验对比,证明基于播放单元时序控制算法的实时流媒体播放器,不但明显提高了播放的流畅性,播放后处理的计算复杂度也降低了40%以上。结果表明,该方法对提高播放质量、减少花屏情况有着突出的效果。     

端到端MPEG-4 FGS视频TCP友好的平滑传输

着重研究了Internet上MPEG-4 FGS(fine grained scalable)视频流的自适应平滑传输,其主要目的在于,在网络带宽变化的情况下,提供稳定的视频回放质量.提出了一种新的基于TFRC(TCP-friendly rate control)的MPEG-4 FGS端到端视频流传输系统框架,在此框架的基础上,首先假设完整的可用带宽变化已知,并且提出了一种离线的自适应平滑算法.此后,给出一种基于改进的ARAR(autoregressive autoregressive)预测技术的在线自适应平滑算法.最后,以NS-2为实验平台进行了模拟实验.模拟实验表明,提出的离线和在线自适应平滑算法可以充分利用可用网络带宽,并且能够在可用网络带宽持续波动的情况下保证接收方的回放尽可能地平稳,从而达到获得最佳视觉效果的目的.     

Multi-Radio Multi-Channel无线传感器网络中信道分配和路由策略

针对无线传感器网络中传输时延长、传输冲突大和吞吐量低等问题,提出了一种在Multi-Radio Multi-Channel无线传感器网络中信道分配和路由策略.该策略动态地建立k元n立方体拓扑结构,使用优化的静态信道分配算法提高节点的吞吐量,使用维序寻径的路由算法减少传输冲突.该方法适用于网络节点稠密、节点相互之间通信冲突大的情况,并且在单跳和多跳的网络环境下均适用.实验结果表明,基于k元n立方体这一拓扑结构的信道分配和路由策略与传统方法相比,有效地减少了端到端时延,降低了网络冲突,减少了节点能量消耗,延长了网络寿命,提高了网络吞吐量.     

PCRTT Enhancement for Off-Line Video Smoothing

An enhancement of the Piecewise Constant Rate Transmission and Transport (PCRTT) algorithm for reducing the burstiness of a video stream, based on smoothing constant interval is proposed. The new algorithm, called e-PCRTT, relies on geometrical consideration rather than traditional rate-control analysis. E-PCRTT is shown to construct transmission rate-plans with smaller buffer sizes, as compared to the original PCRTT. Alternatively, for the same buffer size, e-PCRTT reduces the number of bandwidth changes as compared to PCRTT. In addition, e-PCRTT produces a rate-plan that requires a smaller initial playback delay.     

Periodic broadcast with dynamic server selection

Service replication is an effective way to address resource requirements and resource availability problem. Dynamic service selection enables clients to choose a server offering the best performance. Proper server selection is especially important for video streaming over the Internet due to its high bandwidth requirements. However, given the length of a typical video transmission, the server priorly selected may no longer be an optimal one for the duration of the entire transmission. More importantly, a server may fail during the transmission of a video. In this paper we examine the possibility of switching to another server during an on-going transmission for Periodic Broadcast schemes. Due to the timing requirements typical for Periodic Broadcast the server switch may cause playback disruptions. We analyze the magnitude of the problem and propose an easy to implement solution. We define the criteria, additional to the bandwidth availability for example, according to which a new server should be selected. The client is also required to delay its playback by the amount of time bounded by the server transmission offset. In addition, we propose an alternative method to ensure uninterrupted playback that relies on proxy caching. Simulation results show that our approach can significantly reduce the likelihood of playback disruptions.

基于模糊控制的流媒体自适应传输算法

基于HTTP的动态自适应流媒体DASH传输协议可以使用户根据自身的终端显示能力和信道条件选择合适的视频质量,是网络视频服务技术的发展方向。如何根据网络吞吐量的变化自适应地选择视频码率,以获得最佳的用户体验质量QOE,在已有的DASH系统中还没有得到很好的解决。 提出了一种基于模糊控制的自适应传输算法,将缓存的视频余量以及用户申请的视频码率和网络吞吐量的码率失配度作为输入,将预期的缓存变化量作为输出,通过模糊逻辑实现以下控制目的：(1）将缓存稳定在一个安全的区间;(2）使传输视频的平均质量最大化;(3）避免因带宽波动所造成的视频播放中断。最后,分别在两种虚拟网络环境和两种实际网络环境下进行性能测试,实验结果表明,与已有的算法相比较,本文提出的算法可以给用户带来更好的QOE。     

基于蚁群算法和BP神经网络的信道分配策略的研究

研究无线传感器网络信道分配策略的主要目标是提高网络吞吐量和容量,减小网络的传输时延,最大限度的利用有限的网络带宽资源。多信道MAC协议的应用,可以有效地提高网络通信的可靠性和吞吐量,以及解决由于信道受干扰而造成的网络瘫痪等问题。根据无线传感器网络多信道的特点提出了一种基于蚁群算法的动态反馈负载均衡信道分配策略。本策略首先应用BP神经网络对信道负载情况进行预测,然后通过基于蚁群算法的负载均衡算法对信道进行筛选,最后利用最大离散化算法进行信道分配。在NS2平台下对所设计的协议进行了仿真实现,并与应用最为广泛的多信道MMAC协议以及SMAC进行了对比分析。根据仿真结果可知,本文设计的MAC协议在网络吞吐量、网络传输时延等性能方面比MMAC协议及SMAC都有了很大程度的提升。可以有效减小网络传输时延,提高网络吞吐量和抗干扰能力。