Cross-domain time synchronization method and cross-domain time synchronization system

Abstract

The application discloses a cross-domain time synchronization method and a cross-domain time synchronization system. The method is executed by a control plane and comprises the following steps: a cooperative controller acquires a synchronization path query request and determines a time synchronization path between a slave clock and an ancestor clock according to the synchronization path query request; the time synchronization path is issued to a target domain controller; the target domain controller comprises a source TSN domain controller, a target TSN domain controller and a non-TSN forwarding domain controller; the target domain controller performs time synchronization control according to the time synchronization path; the non-TSN forwarding domain controller forwards a time synchronization data stream by using a flow control strategy; the source TSN domain controller and the target TSN domain controller respectively collect timestamp information in respective domains and report the timestamp information to the cooperative controller; the cooperative controller acquires a time error compensation value of the slave clock according to the timestamp information and corrects the slave clock by using the time error compensation value.

Description

Technical Field

[0001] This application relates to the field of computer network technology, and in particular to a cross-domain time synchronization method and a cross-domain time synchronization system. Background Technology

[0002] Time-Sensitive Networking (TSN), as a next-generation Ethernet technology, provides bounded latency, low latency variation, and zero congestion loss communication guarantees through mechanisms such as network-wide time synchronization and time-aware shaping, meeting the deterministic communication needs within local area networks (LANs) such as factory or vehicle networks. However, the rise of emerging time-sensitive applications such as cloud-based programmable logic controllers, remote operation, and factory interconnection is driving the evolution of TSN from intra-domain to inter-domain to meet broader end-to-end connectivity requirements.

[0003] Interconnected time-aware systems within a TSN network adjust their activities in real time by sharing a clock, thus forming a time-aware network, also known as a gPTP domain. Networked devices within this domain, such as bridges, routers, and terminal stations, enable the TSN network to perform specific actions precisely in real time by supporting TSN network functions such as time synchronization and traffic scheduling. However, in multi-domain environments, it is often difficult to ensure that each network domain simultaneously possesses characteristics similar to TSN for end-to-end connectivity, especially when multiple TSN networks span across non-TSN networks. Therefore, how to interconnect multiple TSN networks through non-TSN networks is crucial.

[0004] TSN networks use the IEEE 802.1AS standard for time synchronization. This standard requires all devices in the network to have 802.1AS capability in hardware, such as peer-to-peer latency measurement mechanisms at Layer 2, to provide high-precision time synchronization guarantees. However, networking devices in non-TSN networks often struggle to support this requirement. Furthermore, due to the varying time synchronization capabilities of non-TSN networks (i.e., some data center networks achieve time synchronization support through dedicated network equipment, while most network devices lack this functionality), time synchronization accuracy can suffer significant losses due to path asymmetry and unknown synchronization message delays during time synchronization between TSN and non-TSN networks. Therefore, without modifying network device hardware, time synchronization becomes a challenging problem when building end-to-end connections between multiple TSN networks interconnected through non-TSN networks. Summary of the Invention

[0005] This application provides a cross-domain time synchronization method and system to improve the end-to-end time synchronization accuracy between TSN domains and non-TSN domains.

[0006] The embodiments of this application adopt the following technical solutions:

[0007] In a first aspect, embodiments of this application provide a cross-domain time synchronization method, wherein the cross-domain time synchronization method is executed by a combined control plane, the combined control plane including a cooperative controller and a domain controller, and the cross-domain time synchronization method includes:

[0008] The collaborative controller obtains a synchronization path query request, which is used to obtain the time synchronization path from the slave clock to the ancestor clock. The networks where the slave clock and the ancestor clock are located are both TSN networks. The TSN networks where the slave clock is located and the TSN networks where the ancestor clock is located are connected through a non-TSN network.

[0009] The cooperative controller determines the time synchronization path between the slave clock and the ancestor clock according to the synchronization path query request, and sends the time synchronization path to the target domain controller. The target domain controller includes the source TSN domain controller, the destination TSN domain controller, and the non-TSN forwarding domain controller.

[0010] The target domain controller performs time synchronization control according to the time synchronization path, wherein the non-TSN forwarding domain controller uses a flow control strategy to forward the time synchronization data stream, and the source TSN domain controller and the destination TSN domain controller respectively collect timestamp information within their respective domains and report it to the coordinating controller.

[0011] The collaborative controller obtains the time error compensation value of the slave clock based on the timestamp information, and performs time synchronization correction on the slave clock based on the time error compensation value.

[0012] Optionally, the flow control strategy includes non-shared network transmission control and shared network transmission control. The non-TSN forwarding domain controller uses the flow control strategy to forward the time synchronization data stream, including:

[0013] When the non-TSN forwarding domain controller plans its intra-domain time synchronization path according to the time synchronization path, it first executes non-common network transmission control. The non-common network transmission refers to adjusting the forwarding strategy of the non-time-synchronized data stream to avoid interference of the non-time-synchronized data stream to the time-synchronized data stream during the non-common network transmission control process.

[0014] And after the non-common network transmission control is completed according to the pre-configuration, the common network transmission control is executed, which refers to the common network transmission of non-time-synchronized data streams and time-synchronized data streams on the time-synchronized path within the domain.

[0015] Optionally, the timestamp information includes a first type of timestamp information collected by the source TSN domain controller and the destination TSN domain controller during the non-common network transmission control process and a second type of timestamp information collected by the source TSN domain controller and the destination TSN domain controller during the common network transmission control process. The cooperative controller obtains the time error compensation value of the slave clock based on the timestamp information, including:

[0016] The collaborative controller obtains the uplink and downlink fixed delays corresponding to the time synchronization path based on the first type of timestamp information, and obtains the end-to-end phase offset value based on the second type of timestamp information;

[0017] The uplink and downlink fixed delays and the end-to-end phase offset value are used as parameters of a preset combined error compensation algorithm, and the time error compensation value of the slave clock is obtained through the preset combined error compensation algorithm.

[0018] Optionally, the preset combined error compensation algorithm is a combination of a phase offset estimation algorithm and a phase offset control algorithm.

[0019] Optionally, the non-common network transmission control includes multiple time synchronization cycles, and the cooperative controller obtains the uplink and downlink fixed delays corresponding to the time synchronization path based on the first type of timestamp information, including:

[0020] The collaborative controller obtains the uplink and downlink delays corresponding to each time synchronization period based on the first type of timestamp information within each time synchronization period, and uses the minimum uplink and downlink delay as the uplink and downlink fixed delay.

[0021] Optionally, the collaborative controller obtains the uplink and downlink delays corresponding to each time synchronization period based on the first type of timestamp information within each time synchronization period, including:

[0022] The sum of the uplink node dwell time is obtained based on the uplink node entry timestamp and the uplink node exit timestamp, and the sum of the downlink node dwell time is obtained based on the downlink node entry timestamp and the downlink node exit timestamp;

[0023] The sum of uplink delays is obtained based on the uplink transmission medium length and the message transmission rate of the link transmission medium, and the sum of downlink delays is obtained based on the downlink transmission medium length and the message transmission rate of the link transmission medium.

[0024] The sum of the uplink node dwell times and the sum of the uplink delays are used as the uplink delay for each time synchronization period, and the sum of the downlink node dwell times and the sum of the downlink delays are used as the downlink delay for each time synchronization period.

[0025] Optionally, the cooperative controller determines the time synchronization path between the slave clock and the master clock based on the synchronization path query request, including:

[0026] The collaborative controller allocates a ancestor clock to the destination TSN domain controller based on the global network view, and calculates the time synchronization path according to a preset routing algorithm.

[0027] Optionally, calculating the time synchronization path according to a preset routing algorithm includes:

[0028] The time synchronization path within the TSN domain is calculated using the shortest path bridging algorithm, and the time synchronization path outside the TSN domain is calculated using the shortest path algorithm.

[0029] Optionally, the global network view can be obtained through the following steps:

[0030] The coordinating controller sends an intra-domain topology query request to the domain controller, enabling the domain controller to query the network topology within its domain based on the intra-domain topology query request and report it to the coordinating controller.

[0031] Based on the network topology within each domain reported by each domain controller, construct a global network view across domains.

[0032] Secondly, embodiments of this application also provide a cross-domain time synchronization system, which includes a control plane and a data plane. The control plane is used for the cross-domain time synchronization method and includes a cooperative controller and a domain controller. The data plane includes a TSN domain located at both ends and a non-TSN domain connecting the TSN domains at both ends.

[0033] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects:

[0034] In this embodiment, the collaborative controller and domain controller in the control plane work together to complete inter-domain and intra-domain time synchronization, ensuring the stability and accuracy of inter-domain time synchronization. As for the accuracy of intra-domain time synchronization, the TSN domain, based on the IEEE 802.1AS standard it follows, can achieve high-precision time synchronization within the TSN domain. Non-TSN domains, based on specific flow control strategies, can avoid the impact of non-TSN forwarding non-time-synchronized data streams on the accuracy of intra-TSN time synchronization, thus enabling high-precision time synchronization within non-TSN domains as well. Therefore, the cross-domain time synchronization scheme of this embodiment can achieve high time synchronization accuracy. Attached Figure Description

[0035] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0036] Figure 1 This is a flowchart illustrating a cross-domain time synchronization method in an embodiment of this application;

[0037] Figure 2 This is a schematic diagram of the system architecture of a cross-domain time synchronization system according to an embodiment of this application. Detailed Implementation

[0038] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0039] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0040] This application provides a cross-domain time synchronization method, which is executed by a combined control plane, including a cooperative controller and domain controllers, such as... Figure 1 The diagram illustrates a flowchart of a cross-domain time synchronization method according to an embodiment of this application. The cross-domain time synchronization method includes at least the following steps S110 to S140:

[0041] In step S110, the cooperative controller obtains a synchronization path query request. The synchronization path query request is used to obtain the time synchronization path from the slave clock to the ancestor clock. The networks where the slave clock and the ancestor clock are located are both TSN networks. The TSN networks where the slave clock is located and the TSN networks where the ancestor clock is located are connected through a non-TSN network.

[0042] The cross-domain time synchronization described in this application refers to end-to-end time synchronization between TSN domains across non-TSN domains. A TSN domain consists of a group of TSN devices managed under the same mechanism. Devices within each TSN domain are subject to unified management constraints within the domain and communicate with each other according to standard protocols. All TSN devices complete time synchronization based on IEEE 802.1AS to share a unified time reference established by a ancestral clock and support the implementation of other network functions, such as traffic scheduling. The network infrastructure of non-TSN domains is not limited to a single network; it can be a wired network, such as a Software Defined Network (SDN), or a wireless network, such as a 5G network.

[0043] The collaborative controller in this embodiment has a global network view and defines global network policies. It is responsible for coordinating domain controllers and uniformly scheduling network resources to achieve cross-domain interconnection even when the hardware composition and software deployment in each TSN domain and non-TSN domain are different. The core functions of the collaborative controller include, but are not limited to, topology management, resource management, connection management, path management, synchronization management, and traffic policies.

[0044] When a slave clock needs to synchronize its time, it sends a time synchronization request to the TSN domain controller closest to it (i.e., from the slave clock to the destination TSN domain controller). After receiving the time synchronization request, the destination TSN domain controller generates a synchronization path query request and sends it to the coordinating controller, so that the coordinating controller can obtain the synchronization path query request.

[0045] In step S120, the cooperative controller determines the time synchronization path between the slave clock and the ancestor clock according to the synchronization path query request, and sends the time synchronization path to the target domain controller. The target domain controller includes the source TSN domain controller, the destination TSN domain controller, and the non-TSN forwarding domain controller.

[0046] After the time synchronization path is constructed, the coordinating controller distributes the time synchronization path to each target domain controller associated with the time synchronization path. Each target domain controller then plans its own intra-domain time synchronization path based on the received time synchronization path.

[0047] by Figure 2Taking the illustrated application scenario as an example, assume that Domain C controller is the destination TSN domain controller, Domain A controller is the source TSN domain controller, and Domain B controller is a non-TSN forwarding domain controller. After receiving the time synchronization path from the coordinating controller, these three domain controllers plan the intra-domain time synchronization path from the Grandmaster Clock (GM) to the egress bridge node of TSN Domain A; Domain B controller plans the intra-domain time synchronization path from the ingress node of non-TSN Domain B to the egress node of non-TSN Domain B; and Domain C controller plans the intra-domain time synchronization path from the ingress bridge node of TSN Domain C to the egress bridge node of TSN Domain C. The egress bridge node of TSN Domain A can be understood as a switch (or other network device) within TSN Domain A that directly connects to a non-TSN domain, and the ingress bridge node of TSN Domain C can be understood as a switch (or other network device) within TSN Domain C that directly connects to a non-TSN domain. This completes the establishment of inter-domain and intra-domain time synchronization paths.

[0048] It should be noted that in other application scenarios, the domain controller of the TSN domain in this application embodiment can be understood as a centralized network configuration (CNC), for example... Figure 2 The domain A controller and domain C controller shown are specifically CNCs.

[0049] In step S130, the target domain controller performs time synchronization control according to the time synchronization path, wherein the non-TSN forwarding domain controller uses a flow control strategy to forward the time synchronization data stream, and the source TSN domain controller and the destination TSN domain controller respectively collect timestamp information within their respective domains and report it to the coordinating controller.

[0050] As mentioned above, the cross-domain time synchronization process in this application embodiment requires both TSN domain time synchronization and non-TSN domain time synchronization. In the TSN network, time synchronization follows the IEEE 802.1AS standard. This standard is based on the bidirectional exchange of timing information, selects the most accurate time source (i.e., the ancestor clock) within the domain as the time reference through the optimal master clock algorithm, establishes a master-slave hierarchical structure, and uses a peer-to-peer delay measurement mechanism to measure node dwell time and average link delay between adjacent nodes to achieve accurate time synchronization within the network domain.

[0051] Unlike IEEE 802.1AS, which can only be used in TSN domains that support the protocol, the Precision Time Protocol (PTP) allows non-PTP-aware network devices to transmit time in the network. Given the complexity and diversity of non-TSN domains, such as the coexistence of fully timed network devices, partially timed network devices, and non-timed-supported network devices in different scenarios, this application proposes a flow control strategy that can be universally applied to non-TSN domains composed of fully timed network devices and / or partially timed network devices and / or non-timed-supported network devices, while ensuring the time synchronization accuracy of non-TSN domains.

[0052] This flow control strategy converts the latency of the intra-domain time synchronization path outside the TSN domain into the latency between the last bridge node in the source TSN domain and the first bridge node in the destination TSN domain. The last bridge node in the source TSN domain can be understood as the exit bridge node on the intra-domain time synchronization path of the source TSN domain, and the first bridge node in the destination TSN domain can be understood as the ingress bridge node on the intra-domain time synchronization path of the destination TSN domain. Figure 2 Taking the network architecture shown as an example, the last bridge node of the source TSN domain is specifically the exit bridge node of TSN domain A, and the first bridge node of the destination TSN domain is specifically the ingress bridge node of TSN domain C.

[0053] In step S140, the collaborative controller obtains the time error compensation value of the slave clock based on the timestamp information, and performs time synchronization correction on the slave clock based on the time error compensation value.

[0054] according to Figure 1 As can be seen from the cross-domain time synchronization method shown, the embodiments of this application are not targeted at a single network scenario, but can be used in various network scenarios such as IT / OT network convergence, 5G-TSN network convergence, and wide area interconnection, and have good versatility. The embodiments of this application use the collaborative controller of the control plane and the domain controller to complete the inter-domain time synchronization and intra-domain time synchronization, which can ensure the stability and accuracy of inter-domain time synchronization. As for the accuracy of intra-domain time synchronization, the TSN domain can achieve high-precision time synchronization within the TSN domain based on the IEEE 802.1AS standard it follows. The non-TSN domain can avoid the impact of non-TSN forwarding non-time synchronized data streams on the accuracy of intra-TSN time synchronization based on specific flow control strategies, so that the intra-TSN time synchronization also has high accuracy. Therefore, the cross-domain time synchronization scheme of the embodiments of this application can achieve high time synchronization accuracy.

[0055] In some embodiments of this application, the cooperative controller in step S120 above determines the time synchronization path between the slave clock and the ancestor clock according to the synchronization path query request, specifically including:

[0056] The collaborative controller allocates a ancestor clock to the destination TSN domain controller based on the global network view, and calculates the time synchronization path according to a preset routing algorithm.

[0057] In some application scenarios of this embodiment, calculating the time synchronization path according to a preset routing algorithm specifically includes:

[0058] The time synchronization path within the TSN domain is calculated using the Shortest Path Bridging (SPB) algorithm, and the time synchronization path outside the TSN domain is calculated using a shortest path algorithm (such as Dijkstra). The co-controller then uses the calculated shortest path as the time synchronization path.

[0059] It is understood that in other application scenarios of this application, the collaborative controller may also use other routing algorithms to calculate the time synchronization path, and the embodiments of this application do not impose specific restrictions on this.

[0060] In some application scenarios of this application, the collaborative controller can construct a global network view through the following steps:

[0061] The coordinating controller sends an intra-domain topology query request to the domain controller, enabling the domain controller to query the network topology within its domain based on the intra-domain topology query request and report it to the coordinating controller.

[0062] Based on the network topology within each domain reported by each domain controller, construct a global network view across domains.

[0063] In practical applications, the collaborative controller can periodically maintain the global network view and perform collaborative control on end-to-end cross-domain time synchronization based on the global network view.

[0064] In some embodiments of this application, the flow control strategy in step S130 above includes non-common network transmission control and common network transmission control. Accordingly, the non-TSN forwarding domain controller uses the flow control strategy to forward the time synchronization data stream, specifically including:

[0065] When the non-TSN forwarding domain controller plans its intra-domain time synchronization path according to the time synchronization path, it first executes non-common network transmission control. The non-common network transmission refers to adjusting the forwarding strategy of the non-time-synchronized data stream to avoid interference of the non-time-synchronized data stream to the time-synchronized data stream during the non-common network transmission control process.

[0066] The system terminates the non-shared network transmission control according to a pre-configured setting before executing shared network transmission control. Shared network transmission control refers to the shared transmission of non-time-synchronized data streams and time-synchronized data streams on a time-synchronized path within a non-TSN domain. The pre-configuration can be a command issued by the coordinating controller or a pre-configuration of N (N is a natural number greater than 1) time synchronization cycles by the non-TSN forwarding domain controller. Thus, the non-TSN forwarding domain controller can terminate the non-shared network transmission control upon receiving a command from the coordinating controller and then execute shared network transmission control; alternatively, the non-TSN forwarding domain controller can proactively terminate the non-shared network transmission control after N time synchronization cycles and then execute shared network transmission control.

[0067] Thus, in this embodiment of the application, for non-TSN forwarding domains with complex and diverse network device configurations, special transmission of time synchronization data streams is achieved through non-shared network transmission control on the intra-domain time synchronization path of the non-TSN forwarding domain. That is, during the non-shared network transmission control process, the non-TSN forwarding domain controller ensures that there is no interference from other service traffic on its intra-domain time synchronization path. In this way, the cooperative controller can measure the parameters (i.e., uplink and downlink fixed delays) used for error compensation algorithms during the non-shared network transmission control process. After measuring the uplink and downlink fixed delays, normal forwarding of other service traffic is restored. And under shared network transmission control, the subsequent time synchronization data stream is forwarded. That is, during the shared network transmission control process, the time synchronization data stream and other service traffic can be transmitted on the same network within the intra-domain time synchronization path of the non-TSN domain.

[0068] In the transmission control schemes of the above two data flows executed by the non-TSN forwarding domain controller, the source TSN domain controller and the destination TSN domain controller will collect the corresponding timestamp information. That is, the timestamp information collected by the source TSN domain controller and the destination TSN domain controller includes the first type of timestamp information reported by the source TSN domain and the destination TSN domain during the non-common network transmission control process, and also includes the second type of timestamp information reported by the source TSN domain and the destination TSN domain during the common network transmission control process.

[0069] In this way, the collaborative controller can establish a timestamp database, add the two types of timestamp information reported by the source TSN domain controller and the destination TSN domain controller to the timestamp database, and the collaborative controller can calculate the time error compensation value from the clock based on the data in the timestamp database.

[0070] In some application scenarios, the collaborative controller obtains the time error compensation value of the slave clock based on the timestamp information, specifically including:

[0071] The collaborative controller obtains the uplink and downlink fixed delays corresponding to the time synchronization path based on the first type of timestamp information, and obtains the end-to-end phase offset value based on the second type of timestamp information;

[0072] The uplink and downlink fixed delays and the end-to-end phase offset value are used as parameters of a preset combined error compensation algorithm, and the time error compensation value of the slave clock is obtained through the preset combined error compensation algorithm.

[0073] It can be seen that the uplink and downlink fixed delays in this embodiment are measured without the influence of other service data flows, and their accuracy is high. However, considering random factors such as clock drift on network element nodes, in some application scenarios of this embodiment, the non-common network transmission control includes multiple time synchronization cycles. The cooperative controller obtains the uplink and downlink fixed delays corresponding to the time synchronization path based on the first type of timestamp information, specifically including:

[0074] The collaborative controller obtains the uplink and downlink delays corresponding to each time synchronization period based on the first type of timestamp information within each time synchronization period, and uses the minimum uplink and downlink delay as the uplink and downlink fixed delay.

[0075] Thus, by using the minimum uplink and downlink delay among multiple time synchronization cycles as the uplink and downlink fixed delay in this embodiment, the accuracy of the uplink and downlink fixed delay can be further improved.

[0076] In some embodiments of this application, the cooperative controller obtains the uplink and downlink latency corresponding to each time synchronization period based on the first type of timestamp information within each time synchronization period, specifically including:

[0077] The sum of the uplink node dwell time is obtained based on the uplink node entry timestamp and the uplink node exit timestamp, and the sum of the downlink node dwell time is obtained based on the downlink node entry timestamp and the downlink node exit timestamp;

[0078] The sum of uplink delays is obtained based on the uplink transmission medium length and the message transmission rate of the link transmission medium, and the sum of downlink delays is obtained based on the downlink transmission medium length and the message transmission rate of the link transmission medium.

[0079] The sum of the uplink node dwell times and the sum of the uplink delays are used as the uplink delay for each time synchronization period, and the sum of the downlink node dwell times and the sum of the downlink delays are used as the downlink delay for each time synchronization period.

[0080] According to the calculation steps for uplink and downlink latency in this embodiment, this application uses the minimum latency between the egress bridge node of the source TSN domain and the ingress bridge node of the destination TSN domain as the fixed uplink and downlink latency corresponding to the time synchronization path. When calculating the fixed uplink and downlink latency in this embodiment, this application does not utilize information provided by network devices in non-TSN forwarding domains, but instead utilizes timestamp information collected by the destination TSN domain and the source TSN domain. Therefore, this solution is not affected by the network configuration of non-TSN forwarding domains and can be applied to various non-TSN forwarding domains mentioned above. Furthermore, the fixed uplink and downlink latency in this embodiment can be understood as the fixed end-to-end transmission latency portion formed when the time synchronization data stream is transmitted via the time synchronization path without the influence of other service data streams (specifically, non-time synchronization data streams). The cooperative controller can then calculate the time error compensation value for the slave clock based on the fixed uplink and downlink latency. This allows the slave clock to eliminate interference from non-time synchronization data streams when performing time synchronization correction based on the time error compensation value, thereby improving the time synchronization accuracy of the slave clock.

[0081] The uplink and downlink fixed delays calculated through the above embodiments can be used as parameters for a preset combined error compensation algorithm. For example, when the preset combined error compensation algorithm combines a phase offset estimation algorithm and a phase offset control algorithm, the uplink and downlink fixed delays can be used as parameters for the phase offset estimation algorithm to compensate for path asymmetry and eliminate fixed-time errors. In practical applications, the phase offset estimation algorithm can be a traditional phase offset estimation method (such as mean, median, etc.) or a more effective minimax estimation method.

[0082] The end-to-end phase offset value obtained from the second type of timestamp information can be used as a parameter for the phase offset control algorithm to compensate for random noise and improve dynamic time error. In practical applications, the phase offset control algorithm can be a feedback-based control method such as proportional-integral-derivative (PID) control or Kalman filtering, or it can be an intelligent control method, such as PID control based on backpropagation neural networks. PID control based on backpropagation neural networks can adapt to the dynamic changes of the synchronization network by introducing the backpropagation neural network as the adjustment algorithm for the PID controller parameters, thereby obtaining the optimal control parameter tuning and solving nonlinear and complex control problems.

[0083] As can be seen from the above embodiments of this application, the cross-domain time synchronization method in this application has at least the following advantages:

[0084] First, the cross-domain time synchronization scheme in this application is an extension of the function on the basis of standardized technology. It is not targeted at a single network scenario. For example, it can be used in time synchronization systems in various scenarios such as IT / OT converged scenarios, 5G-TSN converged scenarios, and wide area interconnection, and has good versatility.

[0085] Second, the cross-domain time synchronization scheme in this application embodiment is an extension of TSN function by the control plane based on the global network view. When cross-domain network connection is achieved, the reliability and accuracy of inter-domain time synchronization can be guaranteed.

[0086] Third, the cross-domain time synchronization scheme in this application separates time transmission and error compensation in the time synchronization process, and achieves high-precision inter-domain time synchronization and intra-domain time synchronization through the coordinated control of the cooperative controller and the domain controller.

[0087] Fourth, the cooperative controller in this application uses the timestamp information collected by the source TSN domain controller and the destination TSN domain controller to calculate the parameters required for the combined error compensation algorithm. Specifically, the calculated uplink and downlink fixed delays are used as parameters for the phase offset estimation algorithm. Compared with the prior art, which obtains the relevant parameters of the phase offset estimation algorithm based on empirical or prior values, this application embodiment has higher compensation accuracy for path asymmetry and can effectively improve the accuracy of time synchronization.

[0088] Fifth, the cross-domain time synchronization scheme in this application is a combined error compensation algorithm designed for fixed time errors and dynamic time errors introduced by noise accumulation during the step-by-step synchronization process, which can further improve the accuracy and stability of time synchronization.

[0089] This application also provides a cross-domain time synchronization system, such as... Figure 2 As shown, a schematic diagram of the system architecture of a cross-domain time synchronization system according to an embodiment of this application is provided. The cross-domain time synchronization system includes a control plane and a data plane. The control plane includes a cooperative controller and a domain controller, which are used to execute the cross-domain time synchronization method in the foregoing embodiments. Those skilled in the art can refer to the relevant embodiments above for the specific functions of the control plane, and this embodiment will not be described in detail here.

[0090] The data plane in this application embodiment includes TSN domains located at both ends and non-TSN domains connecting the TSN domains at both ends.

[0091] Specifically, the data plane consists of a large number of network devices and clock devices, such as network element nodes including gateways and switches, terminal stations connected to network element nodes (such as sensors, actuators, control systems, etc.), and clock resources (such as master clocks, slave clocks, etc.). The data plane is mainly responsible for the efficient forwarding of network data.

[0092] It should be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0093] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0094] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A cross-domain time synchronization method, characterized in that, The cross-domain time synchronization method is executed by a composite control plane, which includes a cooperative controller and a domain controller. The cross-domain time synchronization method includes: The collaborative controller obtains a synchronization path query request, which is used to obtain the time synchronization path from the slave clock to the ancestor clock. The networks where the slave clock and the ancestor clock are located are both TSN networks. The TSN networks where the slave clock is located and the TSN networks where the ancestor clock is located are connected through a non-TSN network. The cooperative controller determines the time synchronization path between the slave clock and the ancestor clock according to the synchronization path query request, and sends the time synchronization path to the target domain controller. The target domain controller includes the source TSN domain controller, the destination TSN domain controller, and the non-TSN forwarding domain controller. The target domain controller performs time synchronization control according to the time synchronization path, wherein the non-TSN forwarding domain controller uses a flow control strategy to forward the time synchronization data stream, and the source TSN domain controller and the destination TSN domain controller respectively collect timestamp information within their respective domains and report it to the coordinating controller. The collaborative controller obtains the time error compensation value of the slave clock based on the timestamp information, and performs time synchronization correction on the slave clock based on the time error compensation value; The flow control strategy includes non-shared network transmission control and shared network transmission control. The non-TSN forwarding domain controller uses the flow control strategy to forward time synchronization data streams, including: When the non-TSN forwarding domain controller plans its intra-domain time synchronization path according to the time synchronization path, it first executes non-common network transmission control. Non-common network transmission refers to adjusting the forwarding strategy of non-time-synchronized data streams to avoid interference between non-time-synchronized data streams and time-synchronized data streams during the non-common network transmission control process. And after the non-common network transmission control is completed according to the pre-configuration, the common network transmission control is executed, which refers to the common network transmission of non-time-synchronized data streams and time-synchronized data streams on the time-synchronized path within the domain; The timestamp information includes a first type of timestamp information collected by the source TSN domain controller and the destination TSN domain controller during the non-common network transmission control process, and a second type of timestamp information collected by the source TSN domain controller and the destination TSN domain controller during the common network transmission control process. The cooperative controller obtains the time error compensation value of the slave clock based on the timestamp information, including: The collaborative controller obtains the uplink and downlink fixed delays corresponding to the time synchronization path based on the first type of timestamp information, and obtains the end-to-end phase offset value based on the second type of timestamp information; The uplink and downlink fixed delays and the end-to-end phase offset value are used as parameters of a preset combined error compensation algorithm, and the time error compensation value of the slave clock is obtained through the preset combined error compensation algorithm.

2. The cross-domain time synchronization method according to claim 1, characterized in that, The preset combined error compensation algorithm is a combination of a phase offset estimation algorithm and a phase offset control algorithm.

3. The cross-domain time synchronization method according to claim 1, characterized in that, The non-common network transmission control includes multiple time synchronization cycles. The cooperative controller obtains the uplink and downlink fixed delays corresponding to the time synchronization path based on the first type of timestamp information, including: The collaborative controller obtains the uplink and downlink delays corresponding to each time synchronization period based on the first type of timestamp information within each time synchronization period, and uses the minimum uplink and downlink delay as the uplink and downlink fixed delay.

4. The cross-domain time synchronization method according to claim 3, characterized in that, The collaborative controller obtains the uplink and downlink delays corresponding to each time synchronization period based on the first type of timestamp information within each time synchronization period, including: The sum of the uplink node dwell time is obtained based on the uplink node entry timestamp and the uplink node exit timestamp, and the sum of the downlink node dwell time is obtained based on the downlink node entry timestamp and the downlink node exit timestamp; The sum of uplink delays is obtained based on the uplink transmission medium length and the message transmission rate of the link transmission medium, and the sum of downlink delays is obtained based on the downlink transmission medium length and the message transmission rate of the link transmission medium. The sum of the uplink node dwell times and the sum of the uplink delays are used as the uplink delay for each time synchronization period, and the sum of the downlink node dwell times and the sum of the downlink delays are used as the downlink delay for each time synchronization period.

5. The cross-domain time synchronization method according to claim 1, characterized in that, The collaborative controller determines the time synchronization path between the slave clock and the ancestor clock based on the synchronization path query request, including: The collaborative controller allocates a ancestor clock to the destination TSN domain controller based on the global network view, and calculates the time synchronization path according to a preset routing algorithm.

6. The cross-domain time synchronization method according to claim 5, characterized in that, The step of calculating the time synchronization path according to a preset routing algorithm includes: The time synchronization path within the TSN domain is calculated using the shortest path bridging algorithm, and the time synchronization path outside the TSN domain is calculated using the shortest path algorithm.

7. The cross-domain time synchronization method according to claim 6, characterized in that, The global network view is obtained through the following steps: The coordinating controller sends an intra-domain topology query request to the domain controller, enabling the domain controller to query the network topology within its domain based on the intra-domain topology query request and report it to the coordinating controller. Based on the network topology within each domain reported by each domain controller, construct a global network view across domains.

8. A cross-domain time synchronization system, characterized in that, The cross-domain time synchronization system includes a control plane and a data plane. The control plane is used to execute the cross-domain time synchronization method according to any one of claims 1 to 7. The control plane includes a cooperative controller and a domain controller. The data plane includes a TSN domain located at both ends and a non-TSN domain connecting the TSN domains at both ends.

Images