Method for determining relative time error in a synchronisation network

Abstract

A method for determining a relative time error between a first node 210 and a second node 220 of a synchronisation network. A measurement device is configured to record a time at which a message traverses a timeplane 280 between the first node and the second node. A first relative time error between the first node and the measurement device at the timeplane is determined by calculating the delay D between the timeplane and the first node. D = (r(Tc – Tb) – (T5 – T4)) / 2 where r is the Neighbour Rate Ratio (NRR). Thus, the first relative time error is T1 – (Ta – D). A second relative time error between the second node and the measurement device at the timeplane is determined in the same manner. The relative time error between the first and second nodes is thus the difference between the two relative time errors. The method may use Precision Time Protocol (PTP) messaging and may be employed in automotive applications where Ethernet is replacing legacy technologies. The timeplane may be part of a measurement device situated in-line between the two nodes or the timeplane may be part of a network tap operatively coupled to the measurement device.

Description

FIELD The present disclosure relates to a method of determining relative time errors between nodes of a synchronisation network. In particular, the disclosure relates to the use of a measurement device to timestamp messages transmitted and received between nodes in a synchronisation network. Also disclosed is a measurement device configured to perform the disclosed method and a synchronisation network comprising the measurement device. BACKGROUND Precision Time Protocol (PTP) is an example of a message-based time transfer protocol that is used for transferring time (phase) and / or frequency across a packet based network. It ensures various points in the network are precisely synchronized to the reference clock so that the network meets specific performance limits according to the network’s application. PTP timing messages are carried within the packet payload. The precise time a packet passes an ingress or egress point of a PTP-aware device is recorded using a timestamp. Assessing the Time Error introduced by these devices is critical to determining suitability of equipment, and demonstrating network timing compliance in many industries e.g. power networks, industrial networks, factory automation, automotive. For example, in the automotive industry, there is a technology transition from devices transmitting time error using Controller Area Network (CAN) bus to those using Ethernet. Higher time accuracies are therefore becoming more important with the increase in the number of communicating electronic devices in automotive vehicles. Many of these devices lack a ‘one pulse per second’ (1PPS) output and / or multiple ports for time error measurement. Verification of time error in these types of devices without isolating or dismantling the network is increasingly important to improve the performance of the networks to which they are connected. It is therefore desirable to provide an improved method of determining relative time error between adjacent nodes in synchronisation networks and networks which employ such methods to accurately record relative time errors, and as such maintain synchronicity. It is therefore an aim of at least one embodiment of at least one aspect of the present disclosure to obviate or at least mitigate at least one of the above identified shortcomings of the prior art. SUMMARY Various aspects of the present invention are defined in the independent claims. Some preferred features are defined in the dependent claims. According to a first aspect of the present disclosure there is provided a method for determining a relative time error between a first node and a second node of a synchronisation network, the method comprising: providing a measurement device configured to record a time at which a message traverses a timeplane between the first node and the second node; determining a first relative time error between the first node and the measurement device at the timeplane; and determining a second relative time error between the second node and the measurement device at the timeplane; wherein the relative time error between the first node and the second node is a difference between the first relative time error and the second relative time error. Advantageously by providing a measurement device coupled between the first node and a second node, a relative time error measurement may be made between the first node and the second node without requiring the first or second node, or the measurement device, to be supplied with, or otherwise coupled to an external primary reference clock. Only the messages transmitted between the first and second node may be necessary to make this measurement, with the required timings recovered from the messages exchanged between the first and second nodes. This means that synchronisation, in the form of relative time error, may be measured between two nodes without having to invasively connect to the devices and may be performed using messaging which is already available on the network. This may provide the ability to test devices with reduced cost and increased efficiency. Furthermore, using an ‘intermediate’ timeplane as reference (otherwise referred to herein as a “reference timeplane”) to measure relative time error to / from connected nodes may not only enable the measurement of relative time error, but beneficially may also provide the ability to observe two adjacent nodes ‘following’ each other. For example, if the first relative time error and the second relative time error are drifting in the same manner the relative time error between the first and second node will be determined to be zero, but the network may still be unstable. If the first node subsequently begins to drift relative to the second node then it may be due to the drift in the time source provided by the first node. This diagnosis may only be possible when the reference timeplane is independent of the internal network time local to each of the the first and second nodes. It will be appreciated that either or both of the first and second nodes may be known in the art as a time transmitter, which may be used as an alternative to archaic terms such as master, master node or master device or alternatively as a time receiver, which may be used as an alternative to archaic terms such as slave, slave node or slave device. These terms may be used interchangeably within the present disclosure. Furthermore, the term “clock” used in the present disclosure may refer to the clock of a time transmitter, time receiver or any other time aware device which may be used to record a time or timestamp and may also be known in the art as a master clock or a slave clock. The clock used by a device to timestamp events on a timeplane comprised within that device may be referred to within the present disclosure as the "local clock” of said device. The term “timeplane” used in the present disclosure may refer to a virtual plane in a clock where any timestamps taken on that plane at exactly the same time would give exactly the same value according to a local clock measuring at that timeplane. The travel time between one timeplane and another is typically called a timeplane offset. The term “measurement timeplane” used in the present disclosure may refer to a timeplane at which a message might be timestamped, it may represent a physical location in or on the network, for example the input or output of a device in or on the network. The term “reference timeplane” used in the present disclosure may refer to a timeplane that may or may not refer to a physical location in or on the network but may be used as a common timeplane to which timestamps can be adjusted or referenced. Accordingly, it will be understood that depending on, for example, the configuration of the measurement device in relation to the reference timeplane, the reference timeplane may correspond to (e.g. be physically collocated with) the measurement timeplane for a particular timestamp. This may allow for simpler and clearer calculations to be made based on the measured timestamps. The term “message” used in the present disclosure may refer to at least a unit of data. In a non-limiting example, a message may typically include control information relating to the source and destination node, and a payload. The payload may contain data relating to the transmission time of the message or previously transmitted messages such that the first and second path delay can be determined. The term “Relative Time Error” used in the present disclosure may refer to the Time Error between any two clocks. In other words, the difference in time that two clocks may measure at the same global, reference or “real” time instant. The “real” time may refer to an idealised concept of a universal reference clock. The measurement of such a time error may have any physical separation which may result in a measurement delay between the clocks correctly accounted for. The term “path delay” used in the present disclosure may refer to a propagation delay, or in other words a transit or travel time of a message transmitted from a first node and received at a second node. The transit time may be the difference between the “real” time the message was transmitted by the first node and the “real” time the message was received at the second node. The path delay may not include any “Relative Time Error” caused by differences in the local time as determined by the local clocks of the first and second node. Therefore, it will be understood that the path delay between two nodes may not be measured directly as the difference in timestamps taken by the respective local clock of the first and second node. It will also be understood that the terms “first” and “second” are simply used in the present disclosure to label the relevant elements for the ease of description, and do not necessarily imply any limitations to the sequence or the total number of the relevant elements. This method may be implemented using standard network protocols in the field provided that a synchronisation message is passed along with a transmit timestamp in each transmission direction e.g. from a first node to a second node and from the second node to the first node. For example, IEEE1588, ITU-T or the like may be used in a network employing a two-way timing protocol e.g. precision time protocol (PTP) or network time protocol (NTP). In this way relative time error between nodes in the network may be determined without having to directly inspect the local clock of the nodes in the network and / or without requiring proprietary protocols to be deployed into the network environment. Determining the first relative time error may comprise determining a difference between: a first recorded time at which a first message is transmitted from the first node; and a difference between a second recorded time at which the first message traverses the timeplane, and a first path delay between the first node and the timeplane. The first path delay is a reported time taken for the first message transmitted by the first node to traverse the timeplane. Including the path delay between the first node and the timeplane in the determination of relative time error between the first node and the measurement device may allow the value to be adjusted to account for any delays that may occur in the transmission of the time at which the first message traverses the timeplane increasing the accuracy of the relative time error measurement. When the clock of the first node and the measurement device are not sufficiently accurate and stable the first path delay may be further adjusted by a factor which represents the ratio of the frequency of the clock of the measurement device and the clock of the first node. This factor is known in the art as the Neighbour Rate Ratio (NRR) and may further increase the accuracy of the relative time error measurement. Determining the second relative time error may comprise determining a difference between: a third recorded time at which a second message is transmitted from the second node; and the difference between a fourth recorded time at which the second message traverses the timeplane, and a second path delay between the second node and the timeplane. The second path delay may be a predetermined or reported time taken for the second message transmitted by the second node to traverse the timeplane. By using a predetermined (e.g. fixed) delay value provided by a user e.g. a network user or engineer before, during or following disposition of the measurement device between the first and second nodes, the relative time error may be determined in networks which may not support determination of path delay from extant messages (e.g. from so called reverse sync messages). The reported time for the second message transmitted by the second node to traverse the timeplane may be based on a message which may be transmitted by the second node to the first node. The message transmitted by the second node may be known in the art as a “Reverse Sync” message. By using the reported times within the reverse sync messages the second path delay may be determined based on real time measurements which may result in a more accurate and reliable relative time error determination. Furthermore, this increased performance may be achieved even if other nodes in the overall network do not support e.g. do not transmit and / or retransmit reverse-sync messages as long as the node supporting reverse-sync messaging is connected to (i.e. is a neighbour of) the measurement device. The measurement device may comprise the timeplane. The measurement device may comprise a clock which may comprise the timeplane. Recording a time a message traverses the timeplane (which may be a / the reference timeplane) may comprise determining a total of a recorded time at which the message is received at an input of the measurement device, a measurement timeplane of the measurement device or other physical or reference location and a path delay between that location and the timeplane. In this way the measurement device may be disposed in an ‘inline’ configuration. If the device is arranged in this way such that messages flowing to / from the first and second nodes in different ‘directions’ through the measurement device are timestamped on different measurement timeplanes (corresponding to different measurement device inputs), and the offset between said measurement timeplanes is known (e.g. predetermined, calculated, modelled or the like) a single common reference timeplane may be used for the relative time error measurement. Disposing the measurement device inline on the network between the first and second node may enable analysis and network tests to be performed in addition to determining relative time errors. For example, such an inline configuration may provide the ability to perform impairments (e.g. the addition of delays, errors, and / or corruption) to the messages, in order to determine how the first and / or second node(s) respond to specific conditions. For example, the inline configuration may allow network traffic between nodes to be manipulated to simulate network issues and observe how one or more nodes in the network under test responds. A network tap may be operatively coupled between the first and second node. The network tap may comprise the timeplane. The measurement device may be operatively coupled to the network tap. The measurement device may comprise the timeplane. The measurement device may comprise a clock which may comprise the timeplane. Recording a time a message traverses the timeplane (which may be a / the reference timeplane) may comprise determining the path delay between each of the inputs of the tap and the measurement timeplane of the measurement device. The path delay may be a difference between a time at which the message enters the network tap and a time at which the message is received at an input of the measurement device, a measurement timeplane of the measurement device or other physical or reference location. The path delay may be predetermined, calculated, modelled or the like prior to, or during, installation of the measurement device. As with the inline configuration, with the measurement device disposed in a ‘tapped’ configuration it may be possible to calculate a reference timeplane such that all message timestamps taken by the measurement device are referenced to a single common reference timeplane which may be used for the relative time error measurement. By using a network tap to couple the measurement device between the first and second nodes, the required messages and times may be recorded whilst minimising / eliminating any disturbance to the transmission path. This provides a more realistic and accurate measurement of relative time delay between the first and second nodes. As the messages do not directly pass through the measurement instrument there may be fewer delays to account for, providing for a less complex method of determining the relative time error. The term “network tap” used in the present disclosure may refer to a device which allows data transmitted and / or received to and / or from nodes in a network e.g. messages, to be recorded without interrupting the flow of data between the respective nodes. Recording the data may comprise recording a time a message traverses a timeplane which is defined by the physical position of the network tap in the network and recording the contents of the message which may include the recorded time the message was transmitted from its respective network node. At least one of the first and second nodes may comprise full or partial timing support as defined by ITU-T G.8260. Networks which have full or partial timing support as defined by an industry regulated standard may implement the method without any additional specialised or proprietary hardware or software on the network itself. It would be appreciated that any suitable standard comprising a two-way timing protocol may be used provided a suitable timestamped message in the reverse direction can be measured. As will be understood by those skilled in the art, any asymmetry between nodes in the network may increase the relative time error between said nodes if it is not known and accounted for. Accordingly, any unknown asymmetry between nodes should be minimised or eliminated in order to provide the most accurate measurement of relative time error. The measurement device may be external to and / or independent from (e.g. operationally independent from or the like) the synchronisation network. The expression “external to and / or independent from” used in the present disclosure may encompass physical separation, the use of independent clock sources not directly and / or indirectly ‘linked’ to, or dependent on each other, independent operational control and / or the like. Advantageously, external and / or independent test equipment may provide a measurement of relative time error using an independent clock source not directly and / or indirectly ‘linked’ to, or dependent on, the local clock of any node of the synchronisation network, additionally providing an independent verification of network performance. A measurement using an independent clock source may not be limited by the hardware and / or processing capabilities of nodes on the network. In particular, the accuracy of measurements taken by the measurement device may not be limited to the hardware capabilities of the network node, the use of the independent clock to take the measurements may not interfere or delay the normal operation of the network and / or the measurement may not rely on the precision or stability of a clock on a network node that may be the same or worse as any other clock in or on the network, therefore making it difficult to determine the source of any noise in the relative time error measurements. Furthermore, a network linked clock (directly or indirectly) may not have visibility of its own internal inaccuracies and therefore may itself be a source of time-error which may be beneficially avoided by using an independent clock in an external measurement device to carry out the measurement. For example, corrections in the delay of forward and reverse sync messages which have been calculated based on a clock internal to the network may cancel each other out, which may effectively ‘mask’ or ‘hide’ a relative time error between nodes in the network. Furthermore, by using an external measurement device, relative time error measurement testing may beneficially be ‘standardised’ such that the device and / or method of carrying out the measurement is not ‘bespoke’ to each system architecture to which the measurement device is coupled, which may reduce the cost and time required to measure relative time errors across different networks or at different positions within a single network. This may also reduce or eliminate the requirement for additional network configuration prior to testing providing a so called “plug and play” (PnP) measurement device. According to a second aspect of the present disclosure there is provided a computer program product comprising instructions that, when implemented on a processing system, causes the controller or processing system to control the measurement device to perform the method of the first aspect. The computer program product may be provided on a carrier-medium, such as a non-transient and / or tangible carrier medium. The computer program product may be programmed or programmable into a processor and / or provided on a memory or storage, such as a RAM, ROM, on a hard drive, on a memory card, USB memory storage, a flash drive or card, and / or the like. According to a third aspect of the present disclosure there is provided a measurement device comprising a processing system configured to implement the computer program product of the second aspect. The processing system may comprise at least one data processing module, such as at least one processor; which may include one or more different types of processor such as one or more of a central processing unit (CPU), Graphics Processing Unit (GPU), maths co-processor, a tensor processing unit, neural processing unit or other type of artificial intelligence (Al) accelerator, a physics processing unit, a field programmable gate array (FPGA), application specific integrated circuit (ASIC), a digital signal processor (DSP), and / or the like. The measurement device may be operatively couplable to a synchronisation network. The synchronisation network may comprise a plurality of nodes. One or more network taps may be operatively coupled between two nodes of the plurality of nodes. The measurement device may be couplable to at least one of the one or more network taps. According to a fourth aspect of the present disclosure there is provided a synchronisation network comprising: a plurality of nodes; and a measurement device configured to perform the method of the first aspect such as the measurement device of the third aspect. The synchronisation network may further comprise at least one network tap operatively coupled between two nodes of the plurality of nodes. The measurement device may be operatively coupled to the network tap. The above summary is intended to be merely exemplary and non-limiting. The disclosure includes one or more corresponding aspects, embodiments or features in isolation or in various combinations whether or not specifically stated (including claimed) in that combination or in isolation. It should be understood that features defined above in accordance with any aspect of the present disclosure or below relating to any specific embodiment of the disclosure may be utilized, either alone or in combination with any other defined feature, in any other aspect or embodiment or to form a further aspect or embodiment of the disclosure. These and other aspects of the present disclosure will now be described, by way of example only, with reference to the accompanying Figures, in which: Figure 1 is an illustration of a known method of network synchronisation using Precision Time Protocol (PTP). Figure 2 is an illustration of the use of PTP in a synchronisation network comprising a reference timeplane according to an example embodiment of the present disclosure. Figure 3 is an illustration of the use of PTP including reverse sync messaging in a synchronisation network comprising a reference timeplane according to an example embodiment of the present disclosure. Figure 4 is a schematic illustration of a measurement device disposed in an in-line configuration within a two node synchronisation network. Figure 5 is an illustration of a method of network synchronisation in the network of Figure 4. Figure 6 is a schematic illustration of a measurement device disposed in a tapped configuration within a two node synchronisation network. Figure 7 is an illustration of a method of network synchronisation in the network of Figure 6. Figure 8 is a flowchart of a method for determining relative time in a synchronisation network according to an example embodiment of the present disclosure. In the Figures, like parts are denoted by like reference numerals. It will be appreciated that the drawings are for illustration purposes only and are not drawn to scale. DETAILED DESCRIPTION OF THE DRAWINGS Figure 1 shows an example configuration 100 of Precision Time Protocol (PTP) peer-to-peer delay messaging as standardised by, for example, IEEE 1588. PTP messaging is typically used, inter alia, to measure and maintain synchronicity between clocks of a first node 110 (e.g. a master node) and a second node 120 (e.g. a slave node) in a synchronisation network. Peer-to-peer messaging is used between ports in the synchronisation network which are directly connected to each other. In a fully synchronised network the local clock of the second node 120 should be synchronised with the local clock of the first node 110 i.e. the time error between the local clock of the first node 110 and the local clock of the second node 120 should be minimised or eliminated. At a first time T1, as determined by the clock of the first node 110, a “Sync” message 130 is transmitted from the first node 110 to the second node 120. In a network comprising a plurality of second nodes 120 the sync message 130 may be transmitted to all of the second nodes 120 simultaneously or sequentially. The second node 120 receiving the message, determines a second time T2 which is the time the sync message 130 was received by the second node 120, as determined by the clock of the second node 120. The sync message 130 may include the value of the first time T1. Alternatively, if the first node 110 is only able to determine the first time T1 after the sync message 130 has been transmitted, and therefore unable to add the first time T1 to the sync message 130, the first node 110 may send a “Follow_Up” message 140 to the second node 120 including the value of the first time T1. At a third time T3, as determined by the clock of the second node 120, a “Peer Delay_Req” message 150 is transmitted from the second node 120 to the first node 110. The first node 110 determines a fourth time T4, which is the time the Peer Delay_Req message 150 is received by the first node 110, as determined by the clock of the first node 110. The first node 110 responds to the Peer Delay_Req message 150 with a “Peer Delay_Resp” message 160 including the value of the fourth time T4 and the value of the fifth time T5, or the difference between the fifth time T5 and the fourth time T4. Similar to the transmission of the Sync message 130, if the first node 110 is only able to determine the fifth time T5 after the Peer Delay_Resp message 160 has been transmitted, a second “Peer Delay Response Follow_up” message 170 may be transmitted by the first node 110 to the second node 120 including the value of the fifth time T5. After the Peer Delay Response message 160 is transmitted by the first node 110, the second node 120 determines a sixth time T6, which is the time the Peer Delay Response message 160 is received by the second node 120, as determined by the clock of the second node 120. In synchronisation networks comprising one or more second nodes 120 in which reverse sync messaging as standardised by, for example, IEEE 1588, has been implemented and enabled, the example messaging shown in Figure 1 can be reversed. In such a case the second node 120 transmits a “Reverse Sync” message (corresponding to sync message 130) and optional follow-up message 140, to the first node 110. The first node 110 responds with a Peer Delay_Req message 150 and the second node then responds with a Peer Delay_Resp message 160 and optional Peer Delay Response Follow_up message 170. Through this exchange of messages, the second node 120 may obtain the value of the first to sixth times T1, T2, T3, T4, T5, and T6 and optionally (with reverse sync messaging) the first node 110 obtains the value of the corresponding reverse first to sixth times. These time values allow, as described in more detail below, the determination of the relative Time Error between the first node 110 and the second node 120 while also optionally accounting for any path delay between the nodes. The sum of the relative time error and path delay of a message transmitted from the first node 110 to the second node 120 as measured by the clock local to the respective node is then the difference between the second and first times, T2 - T1. Assuming path symmetry between the first node 110 and the second node 120 (i.e. the path delays are equal), and that the exchange of messages as described above is completed in a short enough time such that it can be considered no change in synchronisation or path delay symmetry occurs over the message exchange period, the mean path delay (MPD) is then given by: Mean Path Delay = [(T6-T3) - (T5-T4)] - 2 Figure 2 shows an example configuration 200 of PTP messaging which is generally as described in Figure 1, with the additional definition of a reference timeplane 280 between the first node 210 and the second node 220. The value of times Ta, Tb and Tc are measured using a measurement device, such as the measurement device 1000 of figures 4 and 6 (not shown), operatively coupled to the transmission line at the reference timeplane 280. This configuration 200 of PTP messaging can be used to determine a first relative time error between a time T(t)(2W) as generated by the clock of the first node 210 and a time T(t)(woo) as generated by the clock of the measurement device 1000 operatively coupled at reference timeplane 280. As described above with reference to Figure 1, the values T1, T4 and T5 are timestamps which are determined by the clock of the first node 210, passed inside the PTP messages, and received by the first node 210 and / or the second node 220 (depending on network configuration). These timestamps may be available in the Sync message 230, Peer Delay_Resp message 260, and / or follow-up messages 240, 270 depending on the network and / or node configuration. Similarly, values Ta, Tb and Tc are timestamps determined by the local clock of the measurement device 1000, which are recorded by the measurement device as the corresponding messages traverse (or in other words pass through) the reference timeplane 280. A first Relative Time Error (TER(t)(2io,iooo)) between a time T(t)(2io) as generated by the clock of the first node 210, which in this example is a time transmitter, and a time T(t)(iooo) as generated by the clock of the measurement device is then given by: TEr(£)(210,1000) = ^(0(210)- ^(0(1000) Equationl The value of T1 provides T(t)(2io> but for T(t)(iooo)the value of Ta may need to be adjusted to account for any delay D that occurred in transmission of T1 from the first node 210 to the reference timeplane 280. Therefore, substituting in these values to Equation 1 the Relative Time Error (TER(t)(2io,iooo)) is given by: TEr (0(210,1000) = T1 — (Ta — D) Equation 2 where D is the delay between the first node 210 and the reference timeplane 280. As will be appreciated by those skilled in the art, delay D is commonly referred to as the mean link delay and is given by: r(Tc - Tb) - (T5 - T4) D =------------------- Equation 3 where the value of factor r (which is known in the art as the Neighbour Rate Ratio (NRR)) represents the ratio of the frequency of the clock on the measurement device 1000 and the frequency of the local clock of the first node 210 (or in other words the frequency offset between the respective clocks). The NRR is a further source of error which can affect the accuracy of the relative time error calculation and may be given by: T5 (n) — T5 (n — x) NRR(n) = ——--------- Equation 4 Tc(n) - Tc(n - x) The value of r (or NRR) at sample n is calculated with the ratio of the elapsed time on the first node 210 over x samples to the elapsed time on the measurement device 1000 over x samples. As will be appreciated by those skilled in the art, in some implementations x is always equal to n such that the ratio is always calculated with respect to the first sample. Other methods of calculating the value of NRR will be apparent to those skilled in the art. In cases where the clock of the first node 210 and the clock of the measurement device 1000 both comprise a sufficiently accurate and stable oscillator, or where there is very little difference in time between Tc and Tb, it may not be necessary to account for r as the effects of rate offset will be insignificant compared to the link delay. Figure 3 shows an example configuration 300 of PTP messaging which is generally as described in Figure 2. In the configuration 300 shown in Figure 3, reverse sync messaging is used to determine a second relative time error between a time T(t)(22o> as generated by the clock of the second node 220 and a time T(t)(iooo) as generated by the clock of the measurement device 1000 operatively coupled at reference timeplane 280. As in the configuration 200 of Figure 2, the value of times Ta, Tb and Tc in configuration 300 are measured using the measurement device operatively coupled to the transmission line at the reference timeplane 280. Different to the configuration 200, in the configuration 300 the values T1, T4 and T5 are timestamps which are determined by the clock of the second node 220 (as opposed to the first node 210), passed inside the PTP messages, and received by the first node 210 and / or the second node 220 (depending on network configuration). In configuration 300, the Sync message 230 is replaced by a “Reverse Sync” message 330 i.e. a Sync message transmitted from the second node 220 and received by the first node 210. Equations 1 to 4 detailed above can then be used to determine a second relative time error (TER(t)(22o,iooo)) between a time T(t)(22o> as generated by the clock of the second node 220, which in this example is a time receiver, and a time T(t)(woo) as generated by the clock of the measurement device. Once the first relative time error and second relative time error have been determined as detailed above (or by any other suitable method) a relative time error (TER(t)(2w,220)) between a time T(t)(2io) as generated by the clock of the first node 210 and a time T(t)(220) as generated by the clock of the second node 220 is given by: TER(t\2w,22Q) = ^r(O(21o, looo) — (0 (220,1000) Equation 5 In examples where the first relative time error and the second relative time error (calculated as detailed above) are observed to ‘drift’ in unison over time (e.g. at the same rate in the same direction), the determined relative time error between the first node 210 and the second node 220 will be substantially zero. However, this may indicate an instability in the network and may provide an opportunity for the instability to be highlighted and / or for appropriate action to be taken to rectify the instability. Advantageously, as Peer Delay messages (e.g. Peer Delay_Req messages 250 and Peer Delay Resp messages 260) are only exchanged between two nodes (i.e. do not traverse further throughout the network) the use of Peer Delay messages may enable increased visibility of the ‘source’ of the determined time error. They may provide a measurement of the time that a node takes to respond to a Peer Delay_Req message 250 which, as would be appreciated by those skilled in the art, may be integral to its performance. They may also allow the accuracy of the calculated NRR value (see equation 4 above) to be determined. In an alternative configuration, in accordance with a further embodiment of the invention, in the case where one or both of the first node 210 and the second node 220 do not respond to Peer Delay_Req messages 250 (e.g. they are not enabled or available) the value of delay D in Equation 2 is provided by a user instead of being determined in real time by elements of the synchronisation network (for example as per Equation 3 above). This ‘fixed delay’ may be determined by directly measuring the length of the transmission path between the measurement device and the respective node or by using an oscilloscope, or other suitable test equipment, to test the time it takes a signal to traverse between the measurement device and the respective node. One or both of the first and second relative time errors as denoted in Equation 5 can be determined using this method. This may advantageously allow the relative time error to be determined with a single timestamped message i.e. without requiring the determination of time values Tb, Tc, T4 and T5 or the corresponding reverse time values. In order to increase the accuracy of the relative time error measurement further, a time delay between a data message being received at an input of the measurement device and the data message traversing the reference timeplane can be accounted for. This ensures that the times recorded by the measurement device are all recorded at as close to the same timeplane as possible. The value of the time delay which may be accounted for depends on the configuration of the measurement device relative to the reference timeplane e.g. whether the measurement device is disposed in an ‘inline’ or ‘tapped’ configuration as described in more detail below. Figure 4 shows an illustration of a portion of a synchronisation network 2000 in accordance with an example embodiment of the invention. The network 2000 comprises a first node 2100 operatively coupled to a second node 2200. A measurement device 1000 comprising first input 1100 and second input 1200 is disposed in an ‘inline’ configuration with the first node 2100 operatively coupled to the first input 1100 and the second node 2200 operatively coupled to the second input 1200. If the measurement device 1000 records the time (or timestamp), as determined by the clock of the measurement device, when a message is received at the first input 1100 and the second input 1200, a message received by the first input 1100 will be timestamped on a different ‘timeplane’ to a message received by the second input 1200. By accounting for this difference, or ‘through delay’, an accuracy of any determination (e.g. relative time error) which is based on these timestamps may be significantly improved. Figure 5 illustrates an example method of accounting for the through delay in a synchronisation network comprising an inline measurement device, such as the inline measurement device 1000 of Figure 4, to measure relative time error using, for example, box A of the configuration 200 of Figure 2. When sync message 230 and Peer Delay_Resp message 260 are received at the first input 1100 of measurement device 1000, they are timestamped (at time Ta and Tc) on a first timeplane 282. When Peer Delay_Req message 250 is received at the second input 1200 of measurement device 1000, it is timestamped (at time Tb) on a second timeplane 284 which is separated, in time, from the first timeplane 282 by through delay T. By adjusting timestamps Ta and Tc, or Tb by through delay T, all timestamps are recorded at the same timeplane by measurement device 1000. Figure 6 shows an illustration of a portion of a synchronisation network 3000 in accordance with an example embodiment of the invention. The network 3000 comprises a first node 3100 operatively coupled to a second node 3200. A network tap 4000 comprising tap point 4100 is disposed in a ‘tapped’ configuration between the first node 3100 and the second node 3200. The tap point 4100 is operatively coupled to a first input 1100 of the measurement device 1000. The timeplane of the tap point 4100 for messages transmitted from the first node 3100 and received by the second node 3200 (i.e. downstream messages) may not be the same as the timeplane of the tap point 4100 for messages transmitted from the second node 3200 and received by the first node 3100 (i.e. upstream messages). Therefore, if the measurement device 1000 records the time (or timestamp), as determined by the clock of the measurement device when a message is received at the first input 1100, downstream messages will be timestamped on a timeplane which is different from upstream messages. By accounting for the time taken for a message to travel from the tap point 4100 to the first input 1100 of the measurement device 1000, or ‘tap to measurement delay’, the accuracy of any determination (e.g. relative time error) which is based on these timestamps may be significantly improved. Figure 7 illustrates an example method of accounting for the tap to measurement delay in a synchronisation network comprising a measurement device coupled to a network tap, such as the measurement device 1000 coupled to the network tap 4000 of Figure 6, to measure relative time error using, for example, box A of the configuration 200 of Figure 2. When downstream sync message 230 and Peer Delay_Resp message 260 traverse the tap point 4100 of network tap 4000 and are received at the first input 1100 of the measurement device 1000, they are timestamped (at time Ta and Tc) on a first timeplane 284. When upstream Peer Delay_Req message 250 traverses the tap point 4100 of network tap 4000 and is received at the first input 1100 of the measurement device 1000, it is timestamped (at time Tb) on a second timeplane 282 which is separated, in time, from the first timeplane 284 by a delay from the tap to the measurement device timeplane. All messages will appear, with reference to the ‘real’ time, to be timestamped on a timeplane that is later in time due to this delay. Delay P represents twice the delay from the network tap point 4100 to the first input 1100. By adjusting timestamps Ta and Tc, or Tb by delay P, all timestamps may be recorded at the same timeplane by measurement device 1000. Figure 8 is a flowchart summarizing a method 4000 for determining a relative time error between a first node and a second node of a synchronisation network as described throughout this disclosure. The method comprises at step S4100 providing a measurement device configured to record a time at which a message traverses a timeplane between the first node and the second node. The measurement device may be provided in an in-line configuration, tapped configuration or any other suitable configuration which enables the required times to be recorded. At step S4200 a first relative time error between the first node and the measurement device at the timeplane is determined. At step S4300 a second relative time error between the second node and the measurement device at the timeplane is determined. Determining the first and / or second time errors may include accounting for the path delay between the timeplane and the input(s) of the measurement device, depending on the measurement device configuration. The first and / or second time errors may be based on timestamps recorded by the measurement device and / or contained within messages transmitted by the first and / or second nodes. The first and / or second time errors may be based on path delays between the measurement devices and the first and / or second nodes determined from recorded times and / or provided by a user. At step S4400 a relative time error between the first node and the second node is determined. By determining the relative time error between the first node and the measurement device and second node and the measurement device the difference between these time errors with respect to the common reference (in this case the measurement device) allows the relative time error between the first node and the second node to be determined. Although the disclosure has been described in terms of preferred embodiments as set forth above, it should be understood that these embodiments are illustrative only and that the claims are not limited to those embodiments. Those skilled in the art will be able to make modifications and alternatives in view of the disclosure which are contemplated as falling within the scope of the appended claims. Each feature disclosed or illustrated in the present specification may be incorporated in the disclosure, whether alone or in any appropriate combination with any other feature disclosed or illustrated herein.

Claims

1. A method for determining a relative time error between a first node and a second node of a synchronisation network, the method comprising:providing a measurement device configured to record a time at which a message traverses a timeplane between the first node and the second node;determining a first relative time error between the first node and the measurement device at the timeplane; anddetermining a second relative time error between the second node and the measurement device at the timeplane;wherein the relative time error between the first node and the second node is a difference between the first relative time error and the second relative time error.

2. The method of claim 1 wherein determining the first relative time error comprises determining a difference between:a first recorded time at which a first message is transmitted from the first node; anda difference between a second recorded time at which the first message traverses the timeplane, and a first path delay between the first node and the timeplane.

3. The method of claim 2 wherein the first path delay is a reported time taken for the first message transmitted by the first node to traverse the timeplane.

4. The method of any preceding claim wherein determining the second relative time error comprises determining a difference between:a third recorded time at which a second message is transmitted from the second node; andthe difference between a fourth recorded time at which the second message traverses the timeplane, and a second path delay between the second node and the timeplane.

5. The method of claim 4 wherein the second path delay is a predetermined or reported time taken for the second message transmitted by the second node to traverse the timeplane.

6. The method of claim 5 wherein the reported time for the second message transmitted by the second node to traverse the timeplane is based on a reverse sync message transmitted by the second node.

7. The method of any preceding claim wherein:the measurement device comprises the timeplane; andrecording a time a message traverses the timeplane comprises determining a total of a recorded time at which the message is received at an input of the measurement device and a reported time taken for the message to traverse the timeplane after being received at the input.

8. The method of any one of claims 1 to 6 wherein:a network tap is operatively coupled between the first and second node, the network tap comprising the timeplane;the measurement device is operatively coupled to the network tap; andrecording a time a message traverses the timeplane comprises determining a difference between a recorded time at which the message is received at an input of the measurement device and a time taken for the message to be received at the input after traversing the timeplane.

9. The method of any preceding claim wherein at least one of the first and second nodes comprises full or partial timing support as defined by ITU-T G.8260.

10. The method of any preceding claim wherein the measurement device is external to, and / or independent from, the synchronisation network.

11. A computer program product comprising instructions that, when implemented on a processing system, causes the controller or processing system to control the measurement device to perform the method of any of claims 1 to 10.

12. A measurement device comprising a processing system configured to implement the computer program product of claim 11.

13. The measurement device of claim 12, wherein the measurement device is 5 operatively couplable to a synchronisation network, the synchronisation networkcomprising a plurality of nodes.

14. The measurement device of claim 13, wherein the measurement device is operatively couplable to at least one network tap operatively coupled between 10 two nodes of the plurality of nodes of the synchronisation network.

15. A synchronisation network comprising: a plurality of nodes; and the measurement device of claim 12.1516. The synchronisation network of claim 15 further comprising at least one network tap operatively coupled between two nodes of the plurality of nodes and wherein the measurement device is operatively coupled to the network tap.

Images