Recent Advances in Clock Synchronization for Packet-Switched Networks

Many clock synchronization protocols based on message passing, e.g., the Network Time Protocol (NTP), assume symmetric network delays to estimate the one-way packet transmission time as half of the round-trip time. As a result, asymmetric network delays caused by either network congestion or malicious packet delays can cause significant synchronization errors. This article exploits sinusoidal voltage signals of an alternating current (AC) power grid to limit the impact of the asymmetric network delays on these clock synchronization protocols. Our extensive measurements show that the voltage signals at geographically distributed locations in a city are highly synchronized. Leveraging calibrated voltage phases, we develop a new clock synchronization protocol that we call Grid Time Protocol (GTP), which allows direct measurement of one-way packet transmission times between its slave and master nodes, subject to an analytic condition that can be easily verified in practice. The direct measurements render GTP resilient against asymmetric network delays under this condition. A prototype implementation of GTP maintains sub-millisecond synchronization accuracy for two nodes tens of kilometers apart in the presence of malicious packet delays. The result has been demonstrated for both Singapore and Hangzhou, China. Simulations driven by real network delay measurements between Singapore and Hangzhou under both normal and congested network conditions also show the synchronization accuracy improvement by GTP. We believe that GTP is suitable for grid-connected distributed systems that are currently served by NTP but desire higher resilience against unfavorable network dynamics and packet delay attacks.

Recent advances in the industrial internet of things (IIoT) and cyber–physical systems drive Industry 4.0 and have led to remote monitoring and control applications that require factories to be connected to remote sites over wide area networks (WAN). The adequate performance of remote applications depends on the use of a clock synchronization scheme. Packet delay variations adversely impact the clock synchronization performance. This impact is significant in WAN as it comprises wired and wireless segments belonging to public and private networks, and such heterogeneity results in inconsistent delays. Highly accurate, hardware–based time synchronization solutions, global positioning system (GPS), and precision time protocol (PTP) are not preferred in WAN due to cost, environmental effects, hardware failure modes, and reliability issues. As a software–based network time protocol (NTP) overcomes these challenges but lacks accuracy, the authors propose a software–based clock synchronization method, called CoSiWiNeT, based on the random sample consensus (RANSAC) algorithm that uses an iterative technique to estimate a correct offset from observed noisy data. To evaluate the algorithm’s performance, measurements captured in a WAN deployed within two cities were used in the simulation. The results show that the performance of the new algorithm matches well with NTP and state–of–the–art methods in good network conditions; however, it outperforms them in degrading network scenarios.

This article presents the design of a timing system for accelerator facilities, which relies on a general networking approach based on standard Ethernet protocols that keeps all the devices synchronized to a common time reference. The case of the IFMIF-DONES infrastructure is studied in detail, providing a framework for the implementation of the timing system. The network time protocol (NTP) with software timestamping and the precision time protocol (PTP) with hardware timestamping are used to synchronize devices with sub-millisecond and sub-microsecond accuracy requirements, respectively. The design also considers the utilization of IEEE 1588 high accuracy default PTP profile (PTP-HA) to provide sub-nanosecond accuracy for the most demanding components. Three different solutions for the design of the timing system are discussed in detail. The first solution considers the deployment of one time-dedicated network for each synchronization protocol, while the second one proposes the integration of the synchronization data of NTP and PTP into the networks of the facility. The third solution relies on the single distribution of PTP-HA to all the systems. The final design aims to be fully based on standard technologies and to be cost-efficient, seeking for interoperability and scalability, and minimizing the impact on other systems in the facility. An experimental setup has been implemented to evaluate and discuss the suitability of the solutions for the timing system by studying the synchronization accuracy obtained with NTP, PTP and PTP-HA under different network conditions. It includes a timing evaluation platform that tries to resemble the network architecture foreseen in the facility. The measured results revealed that PTP is the most limiting protocol for the second solution. Using the default PTP configuration, it tolerates less than 20% of maximum bandwidth utilization for symmetric bidirectional flows, and around 30% in the case of unidirectional flows (server to client or client to server), with the current setup and using switches without enabled timing support. This case study provides a better understanding of the trade-off between bandwidth utilization, synchronization accuracy and cost in these kinds of facilities.

White-box switches have emerged as a pivotal component in modern networking architectures due to their flexibility, cost-effectiveness, and programmability. Within a data center network, achieving precise time synchronization within these switches is imperative for various critical functions such as packet scheduling, network security, and compliance with regulatory requirements. Conventional time synchronization protocols, like the Network Time Protocol (NTP), fail to meet the stringent time synchronization demands of the white-box switches because they are software-based. In contrast, the Precision Time Protocol (PTP), a hardware-based solution, offers sub-microsecond accuracy and low-latency time synchronization. However, despite the advantages of PTP, integrating it into white-box switches may pose challenges due to additional hardware requirements. This paper elaborates on a study conducted to understand the role of white-box switches and their precise time synchronization in data center networking, as well as a comparative measurement between NTP and PTP. Ultimately, this research contributes to a nuanced understanding of the role of PTP in modern white-box switches and lays down decision making processes regarding time synchronization protocols in networking deployments.

Clock synchronization plays a crucial role in Wireless Sensor Networks (WSNs). Assuming that there is no clock skew between sensor nodes, the Maximum Likelihood Estimate (MLE) of clock offset was derived by [1] for clock synchronization protocols assuming exponential random delays and a two-way message exchange mechanism as in TPSN (Timing-sync Protocol for Sensor Networks [2]) or NTP (Network Time Protocol). The MLE is appropriate for the case that the random delays in WSNs are exponentially distributed. However, the performance of the MLE is deteriorated considerably in the case that the underlying distribution is contaminated by or mixed with other distributions. Hence, a robust estimator is needed. In this paper, clock offset estimators based on the robust M-estimation method are proposed and it is shown that the proposed estimators present excellent performance in the mean squared error (MSE) sense under the condition that the underlying distribution is mixed with other distribution.

Clock synchronization is essential not only for precise data integration and sensor reading fusion in wireless sensor networks which consists of distributed sensors but for packet-switched networks. Over past years, clock synchronization techniques have been proposed for different environments and applications. In this paper, we propose a novel clock synchronization protocol which does not require timestamp exchanges. This proposed protocol focuses on enhancing energy efficiency and communication bandwidth efficiency, and can be realized without any additional message transmissions. This paper shows maximum likelihood estimator of clock skew and clock offset and the corresponding Cramer-Rao lower bounds under Gaussian delay model. Moreover, it is demonstrated that this protocol has the same performance to traditional timestamping techniques through simulations.

An improved approach, called receiver only synchronization (ROS), relative to timing synch protocol for sensor networks (TPSN) was devised based on a two-way timing message exchange mechanism between two nodes. In ROS, all the nodes in their transmission range overhear the messages and adjust their clock parameters without having to communicate by themselves, hence saving a lot of energy. In this paper, we are proving that in ROS, one-way timing message transmission from the root node to the neighboring nodes located within its transmission range is superior to the two-way timing message exchange. Also, we are showing that choosing the more accurate model instead of the linear regression model yields further performance gains for the clock parameters estimation.

With the publication of version 2.0 of the IEEE1588 Standard for accurate time transfer over packet networks in 2008, the Precision Time Protocol (PTP) quickly became the only viable clock synchronization technology for all application domains relying on Ethernet as a transport medium. PTP was deliberately defined in highly generic terms allowing it to be tailored to applicationspecific requirements via PTP profiles. The SMPTE ST 2059-2 standard is the PTP profile for the broadcasting industry and a crucial element for any All-IP Studio deployment. In 2019 the IEEE published a new version of the IEEE1588 standard (PTP v2.1), where PTP was enhanced with several interesting optional features aimed primarily at improving the overall reliability of PTP, especially for larger deployments. — In this paper, we will explain all major new features of PTP v2.1 and investigate to what extent broadcasting applications may benefit from adopting them as part of a new version of the ST 2059-2 PTP profile. Specific focus will be put on possible implications with respect to implementation efforts and operating requirements. We will analyze whether the overall reliability of PTP can be improved by adding new features such as PTP security or extended monitoring. The high accuracy extensions to PTP v2.1 will be described together with a rough effort estimate required to develop and deploy products supporting sub-nanosecond accuracy. — The paper will conclude with a list of recommendations summarizing the benefits of these new features for deploying broadcasting networks at scale.

In this chapter is introduced the issues involved in the Wind Farms Sensorial Data Acquisition and Processing. This chapter is organized in five sub chapters summarized afterwards. The first sub chapter is the introduction. The second sub chapter makes an overview of a wind maintenance system, describing in detail the software related to the acquisition system, the information system and other software. This sub chapter explains also the operation of the acquisition system, including algorithms, hardware and firmware details. The third sub chapter deals with algorithms that manage the results of a methodology presented in the second sub chapter, with the objective to illustrate the operation of the system. The penultimate sub chapter will present results including simulation and real operation of the system, data details for clock synchronization protocols with improved changes, acquisition time and a SVM (Support Vector Machines) classifier applied to sensorial wind data. Finally we will make the chapter conclusions and present the references used in this chapter. The contribution of this chapter is in the design of the architecture proposed with emphasis for synchronous data acquisition in different geographic points. An improvement for PTP (Precision Time Protocol) is included to achieve fast time convergence in the initial phase of a clock synchronization setup. The control and setup of acquisition timings also play an important role in the system behaviour. This chapter also includes different alternatives for this subject. Given the current energy framework and global climate change, the emphasis on renewable energy has grown a lot. One of the most important renewable energies is from wind that has given great contribution for this new paradigm. There are, however, many aspects that must be considered and are related to its framework as an energy environmentally friendly. This growth in wind farms has the effect of the increase in diversity of the type of equipment in wind turbines. Moreover, the average life of each wind generator and readiness of this kind of technology means that there is a legacy of equipments for different ages and maintenance needs. An information system for maintenance, called SMIT (Terology Integrated Modular System) is used as a general base to manage the assets and for the strategic lines to the evolution of

If all clocks within a distributed system share the same notion of time, the application domain can gain several advantages. Among those is the possibility to implement real-time behavior, accurate time stamping, and event detection. However, with the wide spread application of clock synchronization another topic has to be taken into consideration: the fault tolerance. The well known clock synchronization protocol IEEE1588 (precision time protocol, PTP), is based on a master/slave principle, which has one severe disadvantage. This disadvantage is the fact that the failure of a master automatically requires the re-election of a new master. The start of a master election based on timeout and thus takes a certain time span during which the clocks are not synchronized and thus running freely. Moreover the usage of a new master also requires new delay measurements, which prolong the time of uncertainty as well. This paper analyzes the results of such a master failure and proposes democratic master groups instead of hot-stand-by masters to overcome this problem by. It is shown by means of simulation that the proposed solution will not deteriorate the accuracy of the slave clocks in case of a master failure.

Time synchronization is indispensable for convenient network management, device monitoring, security, and other fundamental operations in industrial wireless sensor networks (IWSNs). Over the past few decades, a wide variety of highly accurate clock synchronization protocols have been investigated by employing powerful statistical signal processing techniques. However, most two-way exchange estimation schemes do not readjust the node's local clock upon every resynchronization before the clock parameters are estimated. And it may not be appropriate in IWSNs where time synchronization is consistently required. Based on the two-way message exchange mechanism, this paper investigates the clock synchronization schemes of active node and overhearing node with immediate clock readjustment. The maximum-likelihood estimators of the clock skew and the corresponding Cramer–Rao lower bounds are derived assuming Gaussian delays. Simulation and experimental results validate the performance of the proposed estimators.

The key point of the existing Ethernet test and measurement is that the method needs to achieve precision clock synchronization between different terminals. In this paper, an improved method was proposed with consideration of the time drift and propagation delay, expressing master-slave synchronization as a linear relationship and inducting the line of least squares fitting algorithm that traded as linear matrix form. The precision time protocol (PTP) was also analyzed in this paper. The experiment results show that the synchronization accuracy reaches sub-microsecond grade, which is the same as the PTP.

Clock synchronization is an essential problem faced by packet transport network. PTP (precise time protocol) can be used to solve this problem.Its feasibility has been verified by related literatures. In this paper, we use the OPNET simulation platform to further study the key factors affect the clocks' accuracy of packet-based synchronization schemes, e.g., queuing disciplines, network traffic load, whether low-pass filter is used or not, and deployment of the PTP enabled router. We intend to answer the question how much influence can each factor make to the clocks' synchronization performance.

Time synchronization has played an important role in modern measurement and control systems which require high time accuracy to maintain optimal performance in sample acquisition, signal control and safety protection. The Ethernet/IP based IEEE 1588 precise time protocol (PTP) provides us a prospective solution. By means of taking hardware timestamps, the time synchronization accuracy can be kept in the range of sub-microsecond with minimal network load and low demand on computing resources. Based on the OMNeT++ simulator, we build the PTP ordinary clock model and simulate the typical process of the PTP clock synchronization. The complete clock synchronization process and its performance are both analyzed and evaluated.

Base stations in mobile networks have very strict frequency synchronisation (also referred to as syntonisation) requirements. As backhaul networks are migrated to asynchronous packet-switching technology, timing over packet using the IEEE 1588 Precision Time Protocol (PTP) replaces current synchronisation methods that rely on the synchronous bit clock of the network. With PTP, base-station clocks derive their frequency from the inter-packet delays of Sync messages sent by a high-quality time source at regular time intervals. Packet-delay variation (PDV) thus has a major impact on the achievable synchronisation quality, and the amount of PDV of a backhaul network determines whether the network can support frequency synchronisation of base stations. We present a simulation and an analytical approach to assessing the suitability of backhaul networks for clock synchronisation using PTP and derive two methods for reducing PDV.KeywordsPrecision Time ProtocolFrequency SynchronisationMobile Backhaul Networks