视频端到端的时延测试方法研究

 在各类视频监控网络中,视频端到端的时延测试是一个令人困扰的问题,各设备制造商均无合适的端到端视频时延的测试工具。本文分析了视频端到端时延测试的主要难点,提出了一种适合视频端到端时延的测试方法,通过GPS或Internet网络解决了时延测试关键的同步问题,并利用数字图像识别技术自动分析出依附在视频上的同步时钟信息。     

基于GPS的端到端时延测量方法设计方案

端到端时延是评估Internet的网络性能的重要参数之一,也是近年来的研究热点。由于端到端时延的测量存在收发时钟不同步问题,所以常用的网络时延测量方法大多是通过测往返时延来间接求得端到端时延,测量结果的误差较大。文中基于GPS接收机的同步方法设计出网络时延测量仪来测量端到端时延,解决了收发时钟不同步的问题,提高了测量的精度。     

基于线性规划的Internet端到端时延的估计

测量Internet端到端时延特征是研究Internet端到端分组行为的重要内容之一,它能够应用于QoS(Quality of Service),SLA(Service Level Agreement)的管理、拥塞控制算法研究等许多方面.常用的端到端时延测量方法大多依赖于GPS接收机或采用NTP协议来实现收发端时钟的同步,但由于GPS接收机价格较高不可能每台主机都能配备, NTP协议的精度不能满足要求。该文基于线性规划的方法估计收发时钟的频差、相对时钟偏差等参数,以获得端到端时延的估计。作者在几条不同的链路上进行了测试,结果表明该方法能有效消除收发时钟不同步的影响。     

互联网端到端时延分析及测量方法

由于端到端网络时延的测量中存在收发时钟不同步的问题，在测量中大多是通过测往返时延来间接求得端到端时延，测试结果误差较大。本文通过对IP网络端到端时延性能进行分析，提出主动探测方法：在互联网上通过在一端发送带有时间戳的IP数据包，在另一端记录该测量分组的到达时间戳来获得端到端的时延测量值，然后用线性规划的方法来消除收发时钟的初始相位差和相对频差等参数，估计出网络的端到端时延真实值。并以实例测试对此方法进行验证。     

基于GPS的Internet端到端时延测量的实现

网络行为测量是网络行为分析的基础,而端到端的时延又是网络行为测量中一个重要的参数.但是由于传统的端到端测量存在时钟不同步问题,因此测量的误差较大.文中采用基于GPS接收机的同步方法设计出网络时延测量仪来测量端到端时延,解决了收发时钟不同步的问题,提高了测量的精度.     

时间触发总线时间同步和端到端时延测试方法

分析了TTP总线时间同步机制和基于中断方式的端到端数据传输模式,以及节点间时间同步、端到端时延和抖动等参数,设计了同步精度和时延测试方法,构建测试原型系统和配置场景并开发测试用例。结果表明,各节点间的时间偏差小于总线配置场景所允许的同步精度上限5us,端到端时延263.30～267.43us。因时间槽设置为250us,可知主机应用层与总线控制器同步方式下数据传输的端到端时延主要取决于时间槽所占时间。此测试方法和所用的原型系统可用于TTP总线通信系统设计和部署的优化。     

非时钟同步的单向排队时延测量估计方法

提出了一种测量和估计网络端到端单向排队时延的新方法。与消除时钟偏差和时钟频差的现有方法相比,新方法完全不需要测量端之间的时钟同步。探测分组之间的发送和到达时间间隔在两端分别测量,然后利用傅立叶域-时间域的迭代重构算法估计端到端单向排队时延的分布特性。仿真和分析结果表明新方法具有很好的实时性、准确性和顽健性。     

思博伦LTE端到端业务质量测试解决方案

移动通讯演进到LTE,要求端到端全IP传输,并在此基础上,对业务的支持从单一的话务业务扩展到包括数据、语音、视频等各种应用服务。由于相对于有线网络,无线网络存在高时延、低带宽和不稳定等因素,所以在测试方面,除了进行L2～L3性能测试以外,需要从用户体验角度出发,进行真实业务服务质量的验证。思博伦通信赵隽琪所撰《思博伦LTE端到端业务质量测试解决方案》一文详细介绍了思博伦通信LTE端到端业务质量测试解决方案。此方案可以从基本功能和性能,QoS能力,真实应用业务仿真和质量测试,现网应用场景测试等方面对LTE系统进行端到端的研发测试、选型验证和运维验证。     

DRX对TD-LTE网络端到端时延的影响

针对LTE系统的DRX(非连续接收)机制进行介绍,系统分析DRX机制引入后对TD-LTE网络端到端时延性能的影响,并基于DRX机制打开和关闭前提对TD-LTE网络Ping包端到端时延进行对比测试分析。测试结果表明,DRX机制引入后,不合理的参数设置以及不同的UE(用户终端)处理能力将对网络时延性能产生影响。     

基于端到端信令挖掘的VoNR语音时延切片分析方法及应用

VoNR端到端时延是影响5G语音感知的重要因素。传统的时延分析依赖于性能指标关联和现场测试分析，成本高昂且效率低下。本文提出一种基于端到端信令的VoNR语音时延切片分析方法，将呼叫接续时延的统计分析细化至每条信令和每个网元，准确统计业务流程中每条信令消息在每个网元消耗的时间，结合分段时延切片分析工具定位影响VoNR端到端时延的网元并实施优化，与只统计整体时延的分析方法相比，更加高效和精准。通过实际运用，该方法极大地支撑了5G语音质量的优化提升。     

RTEthernet中改进的FTA时钟同步算法研究

针对以太网存在节点时钟漂移、网络链路延迟、同步能力差等问题,文中基于RTEthernet协议的通信原理,构建了时钟同步系统的模型。文中研究了RTEthernet的组成、工作原理,考虑了影响实时以太网时钟同步精密度的三大因素：漂移率、网络传输延迟和时钟拜占庭故障,并基于此分析了原始的FTA时钟同步算法,发现它在拜占庭故障增多的情况下容错性能明显降低,进而引入“容错中值”的思想进行改进并提出了RTE-FTM算法。通过CANoe仿真平台,对系统(7个节点)中存在2个拜占庭故障与不存在拜占庭故障进行对比分析。结果表明系统的精密度损失率降低了3.1%,并由此验证了该算法的收敛性和有效性。     

面向时间同步网络开通运维的智能时钟解决方案

本文提出了基于SDN的端到端的智能时钟网络架构和解决方案,该方案可以自动准确地进行同步规划,快速高效地发现现网配置隐患,进行自动同步恢复和告警定位,并实现同步性能实时监控,从而全面支撑时间同步网络的开通运维.     

Multimedia group synchronization protocols for integrated servicesnetworks

Unlike traditional data traffic, real time multimedia traffic requires synchronization. Temporal relationships among media must be maintained. Yet delay jitter and the absence of a global clock may disrupt these temporal relationships. This paper introduces new group synchronization protocols for real-time, multimedia applications, including teleconference, teleorchestration and multimedia on demand services. The proposed protocols achieve synchronization for all configurations (one-to-one, one-to-many, many-to-one, and many-to-many), and does so without prior knowledge of the end-to-end delay distribution, or the distribution of the clock drift. The only a-priori knowledge the protocols require is an upper bound on the end-to-end delay. The paper concludes with simulation experiments showing that the protocols work effectively in both LAN and WAN environments     

TTP总线在分布式飞控计算机系统中的应用

TTP总线具有高确定性、低时延、容错支持等优点,在构建有硬实时需求的安全关键分布式容错系统中有较大优势。文章简要总结了飞控计算机系统对系统总线的需求,介绍了TTP总线分布式时钟同步与确定性通信技术,分析了TTP总线容错特性与安全性,设计了基于TTP总线的三余度分布式飞控计算机系统,测试验证了系统中的TTP总线同步精度、通信延迟与抖动、TTP总线对飞控功能的支持等,能够满足飞控计算机系统的通信需求。     

Measurement Scheme for One‐Way Delay Variation with Detection and Removal of Clock Skew

One‐way delay variation (OWDV) has become increasingly of interest to researchers as a way to evaluate network state and service quality, especially for real‐time and streaming services such as voice‐over‐Internet‐protocol (VoIP) and video. Many schemes for OWDV measurement require clock synchronization through the global‐positioning system (GPS) or network time protocol. In clock‐synchronized approaches, the accuracy of OWDV measurement depends on the accuracy of the clock synchronization. GPS provides highly accurate clock synchronization. However, the deployment of GPS on legacy network equipment might be slow and costly. This paper proposes a method for measuring OWDV that dispenses with clock synchronization. The clock synchronization problem is mainly caused by clock skew. The proposed approach is based on the measurement of inter‐packet delay and accumulated OWDV. This paper shows the performance of the proposed scheme via simulations and through experiments in a VoIP network. The presented simulation and measurement results indicate that clock skew can be efficiently measured and removed and that OWDV can be measured without requiring clock synchronization.     

基于灰色预测容错时钟同步算法

针对分布式实时系统中无主式时钟同步存在时钟拜占庭故障和节点通信链路丢失故障的问题,文中提出一种基于灰色预测容错时钟同步算法。该算法基于广播式通信网络LL模型,使用GM(1,1)的灰色预测方法对前轮次的校正偏差值进行分析,从而预测出该节点在故障伦次中的校正偏差值,再通过计算得到修正值。实验结果表明,文中提出的灰色预测算法能够容忍拜占庭故障,同时可克服节点通信链路丢失故障带来的问题,提升了FTA算法的普适性。通过数据对比分析结果表明,该算法的时钟同步精密度相比于原始算法提高了24.3%;相较于其他算法,文中算法在复杂度上也有一定的优势。     

基于连续时间戳的时钟同步与保持的技术研究

为解决分布式系统中时钟同步问题,提出了时钟服务器被动应答式的同步结构、基于连续时间戳统计平均值的误差、时延计算方法和基于设定时钟精度的时钟保持技术。在软件同步的情况下,提高了同步精度,减少了因时钟同步而产生的网络通信量、各分布式系统的CPU开销。     

A Novel Artificial Intelligence Based Timing Synchronization Scheme for Smart Grid Applications

The smart grid control applications necessitate real-time communication systems with time efficiency for real-time monitoring, measurement, and control. Time-efficient communication systems should have the ability to function in severe propagation conditions in smart grid applications. The data/packet communications need to be maintained by synchronized timing and reliability through equally considering the signal deterioration occurrences, which are propagation delay, phase errors and channel conditions. Phase synchronization plays a vital part in the digital smart grid to get precise and real-time control measurement information. IEEE C37.118 and IEC 61850 had implemented for the synchronization communication to measure as well as control the smart grid applications. Both IEEE C37.118 and IEC 61850 experienced a huge propagation and packet delays due to synchronization precision issues. Because of these delays and errors, measurement and monitoring of the smart grid application in real-time is not accurate. Therefore, it has been investigated that the time synchronization in real-time is a critical challenge in smart grid applications, and for this issue, other errors raised consequently. The existing communication systems are designed with the phasor measurement unit (PMU) along with communication protocol IEEE C37.118 and uses the GPS timestamps as the reference clock stamps. The absence of GPS increases the clock offsets, which surely can hamper the synchronization process and the full control measurement system that can be imprecise. Therefore, to reduce this clock offsets, a new algorithm is needed which may consider any alternative reference timestamps rather than GPS. The revolutionary Artificial Intelligence (AI) enables the industrial revolution to provide a significant performance to engineering solutions. Therefore, this article proposed the AI-based Synchronization scheme to mitigate smart grid timing issues. The backpropagation neural network is applied as the AI method that employs the timing estimations and error corrections for the precise performances. The novel AIFS scheme is considered the radio communication functionalities in order to connect the external timing server. The performance of the proposed AIFS scheme is evaluated using a MATLAB-based simulation approach. Simulation results show that the proposed scheme performs better than the existing system.