JP6026918B2 - Time synchronization control method and control apparatus in wired LAN - Google Patents

Description

  The present invention relates to a time synchronization control method and control apparatus in a wired LAN, and more specifically, in a wired LAN, the time synchronization control in the wired LAN that improves the accuracy of time synchronization by focusing on the gap between frames. The present invention relates to a method and a control device.

  When performing time synchronization using Ethernet (registered trademark) (hereinafter referred to as “network standard”) in a wired LAN, a synchronization frame may be waited in a hub (layer 2 switch) due to a frame generated by another communication. In that case, incorrect timing information may be transmitted. The side receiving the synchronization frame usually cannot distinguish whether or not the frame was kept waiting in the hub, so that the time synchronization error increases as a result of correcting the time according to the arrival timing of the frame. The problem occurs.

  In order to prevent such an increase in time synchronization error, it is considered effective to discard an undesirable synchronization frame. Based on this idea, a proposal has been made to reduce the error due to the waiting time by controlling the time that the synchronization frame stays inside the hub (Japanese Patent Laid-Open No. 2010-213101). The apparatus according to this proposal will prevent synchronization errors due to asymmetry of communication delay by discarding frames that cannot maintain symmetry of communication delay in transmission and reception, in other words, delayed frames. It is what. However, this device is complex and expensive and is not suitable for general use in inexpensive systems.

  If there is no hardware support for reducing synchronization error, it is necessary to detect and remove a frame delayed by software. However, for this purpose, synchronous packet filtering based on statistical processing is required, which complicates the calculation, and a sufficient amount of time stamps have already been observed to enable statistical processing. Therefore, there is a problem that it is difficult to detect a delayed frame in the initial stage of synchronization.

JP 2010-213101 A

  As described above, when performing time synchronization in a wired LAN, it is effective to discard undesirable synchronization frames in order to prevent an increase in time synchronization errors caused by waiting for synchronization frames in the hub. However, the conventionally proposed method for discarding a synchronization frame has a problem that it is not suitable for general use because an apparatus used to implement the method becomes complicated and very expensive. In the case of a method using software, there is a problem that calculation is complicated because statistical processing is required, and it is difficult to detect a delayed frame in the initial stage of synchronization.

  The present invention has been made to solve such a problem in the above-described prior art, and reduces the influence of the waiting time of the synchronization frame caused by the hub. It is an object of the present invention to provide a time synchronization control method and control device in a wired LAN that can be provided at a simple and relatively low price, and that can prevent a decrease in accuracy.

  In the network standard in the wired LAN, in order to secure a preparation time for the receiving circuit to receive the next frame, a certain inter-frame gap must be provided between the frames to be transmitted. Yes. For example, in the case of 100BASE-TX communication, a gap of 24 clocks is required when viewed at the MII level. When the hub transmits a frame, the inter-frame gap is set as small as possible in order to maximize the communication band. Therefore, it can be considered that a frame transmitted following a small inter-frame gap has a high possibility that transmission is awaited by the previous frame in the hub.

  This inventor completed this invention paying attention to this point. That is, the invention according to claim 1 for solving the above-mentioned problem compares the value of the inter-frame gap immediately before the received synchronization frame with the specified value of the inter-frame gap specified in the network standard, and based on the comparison result A time synchronization control method in a wired LAN, wherein received synchronization frames are selected and time synchronization is performed using the selected synchronization frames.

  Further, the invention according to claim 2 for solving the above problem is to compare the value of the interframe gap immediately before the received synchronization frame with the specified value of the interframe gap specified in the network standard, and to determine the interframe gap. Time synchronization control method in a wired LAN, in which the time synchronization accuracy is improved by discarding the synchronization frame on the assumption that there is a high possibility of waiting in the hub. It is.

  According to a third aspect of the present invention for solving the above problem, a frame analysis unit, an IFG monitoring unit for monitoring an inter-frame gap, and a frame destruction unit are provided. Is analyzed to determine whether the frame is a synchronous frame, and the information is supplied to the IFG monitoring unit. The IFG monitoring unit performs an interframe large enough immediately before the currently received frame. It is monitored whether a gap exists, a gap between frames having a sufficiently large size cannot be detected, and when a synchronization frame detection signal is received from the frame analysis unit, the frame destruction unit The frame destruction request signal is sent, and the frame destruction unit that has obtained the frame destruction request signal adds noise to the currently passing frame signal and Frame is time synchronization control device in a wired LAN, characterized in that processing to cause errors.

  Further, the invention according to claim 4 for solving the above-mentioned problem comprises a PHY, a MAC, an IFG monitoring unit, and a CPU, and the IFG monitoring unit is arranged between the PHY and the CPU and is separated from the PHY. The inter-frame gap of the MII sent to the MAC is monitored and the size of the inter-frame gap is supplied to the CPU, and the CPU supplies the size of the inter-frame gap supplied from the IFG monitoring unit for the received synchronization frame. When the time is insufficient, the received frame is discarded, and this is a time synchronization control device in a wired LAN.

  The time synchronization control method and control device in the wired LAN according to the present invention is as described above, and the inter-frame gap immediately before the synchronization frame is inspected, and this inter-frame gap becomes the specified value of the inter-frame gap specified in the network standard. The synchronization frame is selected based on whether it is close or not, time synchronization is performed using only the selected synchronization frame, and the synchronization frame that was not selected is discarded. Is transmitted, and a control device therefor can be provided at a relatively low cost.

  Further, in the invention according to claim 4, there is no increase in data on the network, and since the relay device does not accumulate frames, the communication delay time does not fluctuate. It has the effect of not having an effect.

It is a block diagram which shows one structural example of the time synchronous control apparatus in wired LAN which concerns on this invention. It is a block diagram which shows another structural example of the time synchronous control apparatus in wired LAN which concerns on this invention. It is a figure which shows an example of the network incorporating the time synchronous control apparatus in wired LAN which concerns on this invention.

  A mode for carrying out the present invention will be described with reference to the accompanying drawings. The time synchronization control method in the wired LAN according to the present invention compares the value of the interframe gap (hereinafter referred to as IFG) immediately before the received synchronization frame with the specified value of the IFG specified in the network standard, and based on the comparison result The received synchronization frame is selected and time synchronization is performed using the selected synchronization frame.

  In another aspect, the time synchronization control method in the wired LAN according to the present invention compares the IFG value immediately before the received synchronization frame with the IFG specified value specified in the network standard, and the received IFG is If it is close to the specified value, it is considered that there is a high possibility of waiting in the hub, and the synchronization frame is discarded, thereby preventing an erroneous timing from being transmitted and improving time synchronization accuracy. It is characterized by this.

  As described above, in the network standard, an IFG of a certain level or more is provided between frames to be transmitted. This IFG is set as small as possible to maximize the communication band. Therefore, when the IFG is small, that is, when the IFG is close to the specified value of the IFG specified in the network standard, the subsequent frame sent may be waited for transmission by the previous frame inside the hub. It can be considered that the nature is high. Therefore, in the method according to the present invention, the synchronization frame in that case is to be discarded.

  If this method is used, synchronization frames will be discarded with a certain degree of probability, so it may be necessary to increase the transmission frequency of the synchronization frames to the extent that it can be compensated. Compared to the situation in which the timepiece is controlled based on the above, the possibility of causing a large error is reduced.

  There are two frame discarding methods: a method of discarding by a device that relays the frame, and a method of discarding the frame by the node that has received the frame. For convenience, in the following description, a method of discarding by a device that relays a frame is referred to as a relay type, and a method in which a node that receives a frame discards itself is referred to as an embedded type.

  First, a relay-type frame discarding apparatus and method will be described with reference to FIG. In the case of the network standard, a checksum exists at the end of the frame, and even if it is a part of the frame, if it contains data different from the original data, it is determined as an error frame. In general, the hub has a function of discarding error frames. In the present invention, this feature of the hub is used to cause a frame error by adding electrical or logical noise to the frame while the frame to be removed passes through the repeater. To. Since the hub that has received the frame including the error discards the frame, the frame is eventually removed from the network.

  This method has the advantage that the data on the network does not increase, the communication delay time does not change because the relay device does not accumulate frames, and the frames other than the frames to be deleted are not affected at all. .

  As shown in FIG. 1, the relay-type frame discarding device includes a frame analysis unit 1, an IFG monitoring unit 2, and a frame destruction unit 3. In the case of this structure, the frame analysis unit 1 analyzes the header portion of the frame that has entered from the communication path, and determines whether or not it is a synchronization frame. The IFG monitoring unit 2 monitors whether a sufficiently large IFG exists immediately before the currently received frame.

  If the IFG monitoring unit 2 cannot detect a sufficiently large IFG and a synchronization frame detection signal is received from the frame analysis unit 1, the frame destruction request signal is sent to the frame destruction unit 3. Put out. When a frame destruction request signal is obtained, the frame destruction unit 3 adds noise to the currently passing frame signal so as to process the frame to cause an error. When there is no frame destruction request, the frame is passed as it is.

  In the case of the built-in type, as shown in FIG. 2, a frame removal device (function) is added inside the device that finally receives the frame. In FIG. 2, the area between PHY 4 and MAC 5 is expressed as MII, but this portion varies depending on the type of network standard to be used. For example, in the case of 100BASE-TX, this part is usually MII.

  Normally, the MII signal is connected only to the MAC 5, but it is also connected to the IFG monitoring unit 6 to measure the size of the IFG. If the IFG monitoring unit 6 cannot detect a sufficiently large IFG, it notifies the CPU 7 of it. The CPU 7 confirms whether or not the received frame is a synchronization frame, and further obtains the size of the IFG from the IFG monitoring unit 6. If the received frame is a synchronization frame and the IFG size is insufficient, the received frame is discarded.

<Example>
As an embodiment of the present invention, FIG. 3 shows a configuration example in the case of performing time synchronization in a master / slave type (IEEE 1588 or the like). Here, a network composed of two-stage hubs 8 and 9, one master 10, and two slaves 11 and 12 is assumed. The relay type device 13 is installed between the hub 8 and the hub 9 or between the slave 12 and the hub 9 that do not have a built-in type discard function. The built-in device 14 is installed inside the master 10 and the slave 11. The slave 11 in which the built-in device 14 is incorporated is directly connected to the hub 9.

  In the case of using the FPGA-based MII time stamper, the method according to the present invention can be carried out by adding only one counter for counting IFGs, so that the mounting cost is considerably low. In addition, since it is possible to transmit information indicating that the IFG size is small on the network only by adding a 1-bit flag to the communication packet for synchronization, software implementation is also simple. However, this method can be distinguished only when the hub has a single stage. However, when the hub is connected in multiple stages, the configuration described in the above embodiment may be used.

  Although the invention has been described in some detail with its most preferred embodiments, it is evident that a wide variety of embodiments can be constructed without departing from the spirit and scope of the invention. Accordingly, the invention is not limited to that particular embodiment, except as limited in the appended claims.

1 Frame analysis unit 2 IFG monitoring unit 3 Frame destruction unit 4 PHY
5 MAC
6 IFG monitoring unit 7 CPU
8 Hub 9 Hub 10 Master 11 Slave 12 Slave 13 Relay type device 14 Embedded type device

Claims (4)

  The value of the interframe gap immediately before the received sync frame is compared with the specified value of the interframe gap specified in the network standard, and the received sync frame is selected according to the comparison result, and the selected sync frame is used. A time synchronization control method in a wired LAN, characterized by performing time synchronization.   When the inter-frame gap value immediately before the received synchronization frame is compared with the inter-frame gap specified value in the network standard, and the inter-frame gap is close to the specified value in the network standard, A time synchronization control method in a wired LAN, characterized in that time synchronization accuracy is improved by discarding the synchronization frame as having a high possibility of waiting.   It consists of a frame analysis unit, an IFG monitoring unit that monitors the gap between frames, and a frame destruction unit. The frame analysis unit analyzes the header part of the frame that has entered from the communication path to determine whether it is a synchronous frame. , Supplying the information to the IFG monitoring unit, which monitors whether there is a sufficiently large interframe gap immediately before the currently received frame, When an inter-gap cannot be detected, and when a synchronous frame detection signal is received from the frame analysis unit, a frame destruction request signal is sent to the frame destruction unit, and the frame destruction request signal is obtained. The frame destruction unit adds noise to the currently passing frame signal and processes the frame to cause an error. Time synchronization control apparatus in a wired LAN to.   PHY, MAC, IFG monitoring unit, and CPU, the IFG monitoring unit is arranged between the PHY and the CPU and monitors the inter-frame gap of MII sent from the PHY to the MAC. The gap size is supplied to the CPU, and the CPU discards the received frame when the size of the gap between frames supplied from the IFG monitoring unit for the received synchronization frame is insufficient. A time synchronization control device in a wired LAN.

Images