A method and system for fast transmission of time-sensitive messages

Abstract

This invention discloses a method and system for rapid transmission of time-sensitive messages, belonging to the field of power system control. The method includes: when a message is sent from a device, it carries a priority tag and is forwarded into a network transmission device; when the message is sent from the network transmission device, the priority order is determined according to the priority tag carried by the message; when a low-priority message is being transmitted, a high-priority message is intercepted and transmitted via a high-priority port; when a higher-priority message needs to be transmitted, the intercepted port transmits the higher-priority message; the receiving side buffers the truncated message after receiving it, and forwards it after subsequent messages are received completely, thus achieving rapid transmission of time-sensitive messages. This invention reduces the network transmission time of time-sensitive messages and ensures their priority transmission.

Description

Technical Field

[0001] This invention belongs to the field of power system control, and specifically relates to a method for fast transmission of time-sensitive messages. Background Technology

[0002] With the widespread application of smart substations, new energy sources, and intelligent power distribution, the scope of rapid business applications in power systems is becoming increasingly broad. Substation protection, control, and intelligent operation and maintenance services, such as drone and robot inspections, all require rapid response and necessitate quick information transmission. Reducing information interaction time will improve response speed. Currently, these services are transmitted via traditional Ethernet and wireless networks, sharing network resources with control and monitoring services. Monitoring services, which have lower real-time requirements, compete with control services for network resources, leading to longer transmission times for control services with high real-time requirements, ultimately resulting in longer control response times.

[0003] Prior art document 1 (CN116668376B) discloses an Ethernet controller, but its shortcoming is that the truncation of messages is limited by the minimum frame length of Ethernet. The truncation of messages must insert 6 bytes of interframe gap, 1 byte of start delimiter and 1 byte of slice counter. Therefore, messages with high real-time requirements cannot be preempted for transmission as soon as they arrive. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for rapidly sending time-sensitive messages, thereby solving the technical problem that in current power system multi-service network transmission scenarios, messages with low real-time requirements occupy transmission channels, resulting in the inability to rapidly transmit messages with high real-time requirements.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution.

[0006] The first aspect of this invention provides a method for fast transmission of time-sensitive messages, comprising:

[0007] Step 1: The device's output port sends messages according to message priority. When a low-priority message is sent at the device's output port, a high-priority message will preempt the device's output port as a preemption message. The truncated portion of the low-priority message will be used as a truncated message, resulting in truncated messages and preemption messages.

[0008] Step 2: Add a truncation flag and a checksum at the truncation point of the truncated message in Step 1. Process the preempted message according to whether the message is sent without gaps, and obtain the processed truncated message and the processed preempted message.

[0009] Step 3: After determining that the truncated message processed in Step 2 is a truncated message at the ingress port, it is buffered at the ingress port. When the truncated part of the low-priority message arrives at the ingress port, it is concatenated with the buffered processed truncated message and buffered at the ingress port to obtain the buffered complete low-priority message.

[0010] Step 4: After the incoming port determines that the preemptive message processed in Step 2 is a complete message, it sends the complete high-priority message and the complete low-priority message buffered in Step 3 according to the priority from high to low, thus obtaining the complete high-priority message and the complete low-priority message, realizing the fast transmission of time-sensitive messages.

[0011] Preferably, in step 1, the message includes Ethernet frame format messages and custom format messages, and the message length is greater than 16 bytes.

[0012] Preferably, in step 1, the output port sends the same low-priority message frame. When a higher-priority message arrives at the output port, it will continue to preempt the output port.

[0013] Preferably, in step 2, the high-priority message is set to whether it is sent without gaps with the truncated message. The high-priority message is a preemptive message. When it is set to be sent without gaps, the information between the preemptive message and the truncated message does not include the frame gap and preamble; when it is set to be sent with gaps, the information between the preemptive message and the truncated message includes the frame gap and preamble.

[0014] Preferably, when the preemption message is sent without gaps, a two-byte gapless transmission identifier is inserted before the preemption message is sent to identify the preemption message. The contents of the two bytes are 0xEB and 0xEB respectively.

[0015] Preferably, in step 2, the truncation identifier is represented by 4 bytes, the contents of which are 0xEB, 0xEB, 0xBE and the truncation sequence number, respectively.

[0016] The checksum consists of 2 bytes and performs a CRC check on all bytes of the message content sent before this truncation.

[0017] Preferably, the truncation sequence number is represented by 1 byte, with the high 4 bits representing the number of the truncated message, up to a maximum of 16 truncated messages; and the low 4 bits representing the number of times the truncated message has been truncated, with a maximum of 16 times per frame.

[0018] Preferably, in step 3, after the ingress port receives the truncated portion, it concatenates it with the received portion and then buffers it. When the ingress port receives the truncated second-highest priority message, it buffers the message in the ingress buffer. The buffer queue depth of each port is set according to the number of priority queues. When the ingress port receives the truncated message, it performs truncation check and reception identification.

[0019] Preferably, in step 3, a buffer queue is set for each ingress port, and the length of each buffer entry is the maximum allowed message length. The length of the buffer entry is adjusted by setting.

[0020] A second aspect of the present invention provides a system for fast transmission of time-sensitive messages, used to run the aforementioned method for fast transmission of time-sensitive messages, comprising:

[0021] The message transmission truncation module is used to send messages according to message priority at the device's output port. When a low-priority message is sent at the device's output port, a high-priority message will preempt the device's output port as a preemption message. The truncated portion of the low-priority message will be used as a truncated message, thus obtaining the truncated message and the preemption message.

[0022] The message preemption transmission module is used to add a truncation mark and a check code at the truncation point of the truncated message, and to process the preempted message according to whether the message is sent without gaps, so as to obtain the processed truncated message and the processed preempted message.

[0023] The message caching module is used to cache the processed truncated message received at the ingress port after determining that it is a truncated message. When the truncated part of a low-priority message arrives at the ingress port, it is concatenated with the cached processed truncated message and cached at the ingress port to obtain the cached complete low-priority message.

[0024] The message splicing and sending module is used to determine whether the received preemptive message after processing is a complete message at the ingress port. Then, it sends the complete high-priority message and the buffered complete low-priority message in descending order of priority, thus obtaining the complete high-priority message and the complete low-priority message, enabling the fast sending of time-sensitive messages.

[0025] The beneficial effects of this invention are compared with those of the prior art:

[0026] This invention provides a method for fast transmission of time-sensitive messages, which solves the problem that in the scenario of power business shared network transmission, low real-time requirement service messages occupy network transmission resources, resulting in high real-time requirement service messages, thus causing high real-time services to be unable to respond quickly;

[0027] This invention provides two preemption methods: one is a preemption method with frame gaps, but it is not limited by the minimum Ethernet packet length, allowing packets with high real-time requirements to be sent faster and reducing transmission time; the other is a preemption method without frame gaps, which inserts a two-byte preemption indication between the preempted frame and the preempted frame to isolate the two frames, making the preempted frame sent faster and further reducing transmission time. By preempting the transmission of non-real-time service packets and prioritizing the transmission of real-time service packets to the receiving side, the transmission time of real-time services on the network is reduced, thereby improving the response speed of real-time services and thus improving the overall reliability of the power system. The packets have high real-time performance, and two different preemption methods can be selected according to different scenarios.

[0028] The method of this invention is applicable to scenarios where real-time and non-real-time services are transmitted on the same network. It reduces the number of networks while ensuring the real-time transmission of real-time services and has strong applicability. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the message preemption mechanism transmission process in this invention.

[0030] Figure 2 This is a schematic diagram of the message structure before and after message transmission truncation in this invention.

[0031] Figure 3 This is a schematic diagram of seamless transmission of truncated and preempted messages in this invention.

[0032] Figure 4 This is a schematic diagram illustrating the intermittent transmission of truncated and preemptive messages in this invention.

[0033] Figure 5 This is another schematic diagram illustrating the intermittent transmission of truncated and preemptive messages in this invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0035] The embodiments described in this application are merely some, not all, embodiments of the present invention. Based on the spirit of the present invention, other embodiments obtained by those skilled in the art without inventive effort are all within the protection scope of the present invention.

[0036] Example 1

[0037] Embodiment 1 of this invention proposes a method for fast transmission of time-sensitive messages. This method is also applicable to wireless network scenarios and specifically includes:

[0038] Step 1: The device's output port sends messages according to message priority. When a low-priority message is sent at the device's output port, a high-priority message will preempt the device's output port as a preemption message. The truncated portion of the low-priority message will be used as a truncated message, resulting in truncated messages and preemption messages.

[0039] The message can be in Ethernet frame format or a custom format, and the message length should be greater than 16 bytes.

[0040] Step 2: Add a truncation flag and a checksum at the truncation point of the truncated message in Step 1. Process the preempted message according to whether the message is sent without gaps, and obtain the processed truncated message and the processed preempted message.

[0041] At the device's output port, messages are sent according to message priority. When a message is being sent, a message with a higher priority than the current message will preempt the output port and be sent.

[0042] The same low-priority message frame sent from the output port can be preempted multiple times. After receiving the truncated message, the input port concatenates it with the already received portion and buffers it. When a higher-priority message arrives at the output port, it will continue to preempt the output port.

[0043] The truncation identifier is represented by 4 bytes, the contents of which are {0xEB, 0xEB, 0xBE, truncation number} in sequence. It can be set according to the use case, and the meaning of each byte can also be redefined through settings.

[0044] The checksum consists of 2 bytes. It is a checksum of the sent message portion plus the truncation identifier portion. It performs CRC check on all bytes of the sent message content before this truncation. The checksum algorithm of the outgoing port is consistent with the checksum algorithm of the incoming port.

[0045] High-priority messages can be configured to be sent without gaps or truncated frames. When set to send without gaps, there is no gap, preamble, or other information between the preemptive message and the truncated message. A two-byte gapless transmission identifier is inserted before the preemptive message is sent to identify it. The contents of these two bytes are {0xEB, 0xEB}, which can be set according to the usage scenario. The meaning of each byte can also be redefined through settings. Before sending the preemptive frame, the preamble and delimiter are sent to distinguish between the two frames.

[0046] When set to non-immediately following transmission, there is a true gap and preamble between the look-ahead message and the truncated message. Before sending the preemptive frame, sufficient frame gap is ensured between the two frames, and then the preamble, delimiter, message content, etc. are sent sequentially.

[0047] Step 3: After determining that the truncated message processed in Step 2 is a truncated message at the ingress port, it is buffered at the ingress port. When the truncated part of the low-priority message arrives at the ingress port, it is concatenated with the buffered processed truncated message and buffered at the ingress port to obtain the buffered complete low-priority message.

[0048] When the incoming port of the device receives a truncated message, the message is buffered in the incoming port buffer. When the truncated part of the message arrives, the message is reassembled. The buffered messages are buffered in order according to the frame sequence number provided in the truncation identifier.

[0049] When an incoming port receives a truncated second-highest priority packet, it buffers the packet in the incoming port buffer. The buffer queue depth for each port is set according to the number of priority queues. A truncated packet has a truncation identifier and a checksum added at the truncation point. The incoming port performs truncation checks and reception correctness verification upon receiving the truncated packet.

[0050] Set up a buffer queue for each ingress port. The length of each buffer entry is the maximum allowed packet length. The buffer entry length can be adjusted through settings.

[0051] The truncation sequence number is represented by 1 byte. The high 4 bits represent the number of the truncated message, and a maximum of 16 truncated messages can be numbered. The low 4 bits represent the number of times the message has been truncated, and a single message frame can be truncated a maximum of 16 times.

[0052] Step 4: After the incoming port determines that the preemptive message processed in Step 2 is a complete message, it sends the complete high-priority message and the complete low-priority message buffered in Step 3 according to the priority from high to low, thus obtaining the complete high-priority message and the complete low-priority message, realizing the fast transmission of time-sensitive messages.

[0053] When the ingress port receives a complete message, it calculates the checksum and compares the results. If they match, the message is forwarded directly; otherwise, it is discarded.

[0054] After receiving a truncated packet, the ingress port identifies the truncated packet identifier, recalculates the packet's checksum, and compares it with the checksum at the end of the packet. If they are different, the packet is discarded. If they are the same, the port continues to identify the truncation sequence number in the truncation identifier. If the number is 1, it indicates the first truncation. This part of the packet is cached in the ingress port queue, and the truncation count is recorded. If the number is not 1, the port searches for a preceding packet in the corresponding position in the cache queue. If there is a preceding packet and they are consecutive, this part of the packet is cached in the ingress port queue. The port then determines whether the packet is complete. If it is complete, it enters the forwarding stage. If it is incomplete, it continues to be cached, waiting for the remaining part of the packet.

[0055] The beneficial effect of the present invention is that it provides a method for fast transmission of time-sensitive messages, which solves the problem that in the scenario of power business co-network transmission, the low real-time requirement of service messages occupies network transmission resources, resulting in the high real-time requirement of service messages and thus the inability of high real-time services to respond quickly.

[0056] This invention provides two preemption methods: one is a preemption method with frame gaps, but it is not limited by the minimum Ethernet packet length, allowing packets with high real-time requirements to be sent faster and reducing transmission time; the other is a preemption method without frame gaps, which inserts a two-byte preemption indication between the preempted frame and the preempted frame to isolate the two frames, making the preempted frame sent faster and further reducing transmission time. By preempting the transmission of non-real-time service packets and prioritizing the transmission of real-time service packets to the receiving side, the transmission time of real-time services on the network is reduced, thereby improving the response speed of real-time services and thus improving the overall reliability of the power system. The packets have high real-time performance, and two different preemption methods can be selected according to different scenarios.

[0057] The method of this invention is applicable to scenarios where real-time and non-real-time services are transmitted on the same network. It reduces the number of networks while ensuring the real-time transmission of real-time services and has strong applicability.

[0058] Example 2

[0059] Embodiment 2 of the present invention provides a system for fast transmission of time-sensitive messages, used to run the method for fast transmission of time-sensitive messages described in Embodiment 1, comprising:

[0060] The message transmission truncation module is used to send messages according to message priority at the device's output port. When a low-priority message is sent at the device's output port, a high-priority message will preempt the device's output port as a preemption message. The truncated portion of the low-priority message will be used as a truncated message, thus obtaining the truncated message and the preemption message.

[0061] The message preemption transmission module is used to add a truncation mark and a check code at the truncation point of the truncated message, and to process the preempted message according to whether the message is sent without gaps, so as to obtain the processed truncated message and the processed preempted message.

[0062] The message caching module is used to cache the processed truncated message received at the ingress port after determining that it is a truncated message. When the truncated part of a low-priority message arrives at the ingress port, it is concatenated with the cached processed truncated message and cached at the ingress port to obtain the cached complete low-priority message.

[0063] The message splicing and sending module is used to determine whether the received preemptive message after processing is a complete message at the ingress port. Then, it sends the complete high-priority message and the buffered complete low-priority message in descending order of priority, thus obtaining the complete high-priority message and the complete low-priority message, enabling the fast sending of time-sensitive messages.

[0064] Example 3

[0065] Figure 1 Example 3 of a message preemption sending process is given. First, three frames of messages with different priorities arrive at the output port of device 1. The priority relationship of the three frames of messages is: priority 1 < priority 2 < priority 3. The message with priority 1 enters the sending stage because it arrives earliest and there are no other messages to send at the output port.

[0066] Device 1 output port:

[0067] During the transmission of a priority 1 message, a priority 2 message arrives at the output port of device 1. Because the priority 2 message has a higher priority than the priority 1 message, the transmission of the priority 1 message is preempted. The remaining portion of the priority 1 message will continue to be buffered at the output port, waiting to be transmitted. A truncation identifier and checksum field are appended to the end of the already transmitted portion of the priority 1 message. Since this is the first frame and the first time it is truncated, the truncation identifier is filled with {0xEB, 0xEB, 0xBE, 0x11}. The checksum is the message checksum of the transmitted portion plus the truncation identifier, which can be calculated using the CRC method. The checksum algorithm on the sending side is consistent with the checksum algorithm on the receiving side.

[0068] Before and after message truncation Figure 2 As shown, before truncation, the message is complete, containing a preamble + delimiter, destination address, source address, data, and checksum. After truncation, unsent messages contain the source address, data, and checksum, while truncated messages contain a preamble + delimiter, destination address, truncation identifier, and checksum. Priority 2 messages can be sent after the truncation identifier and checksum inserted by priority 1 messages have been sent. Figures 3-5 The method of message preemption and transmission is used. Figure 3 and Figure 4 This indicates that there is no gap between the transmission of two frames. Figure 5 This indicates that there is a gap between the transmission of two frames.

[0069] Figure 3 The method is that priority 2 messages are sent immediately after priority 1 messages. A 2-byte preemption frame identifier is inserted before the message is sent to distinguish the two frames. The identifier value is {0xEB, 0xEB}. At this time, the preemption message contains the identifier and the destination address, and the truncated message contains the checksum.

[0070] Figure 4The method is to send a priority 2 message without immediately following a priority 1 message. Before sending the preemptive frame, a preamble and a delimiter are sent first, and the preamble and delimiter are used to distinguish between the two frames.

[0071] Figure 5 The method is to send priority 2 messages without immediately following priority 1 messages. Before sending the preemptive frame, ensure that there is enough frame gap between the two frames, and then send the preamble, delimiter, message content, etc. in sequence.

[0072] During the transmission of a priority 2 message, a priority 3 message arrives at the output port of device 1, and the priority 2 message will be truncated. Similarly, the remaining portion of the priority 2 message will continue to be buffered at the output port. A truncation identifier and checksum field are appended to the end of the already transmitted portion of the priority 2 message. Because this is the first truncation of the second frame, the truncation identifier is filled with {0xEB, 0xEB, 0xBE, 0x21}. The checksum is the message checksum of the transmitted portion plus the truncation identifier portion. The priority 3 message can proceed after the truncation identifier and checksum inserted in the priority 2 message have been transmitted. Figures 3-5 The method of message preemption and transmission is used.

[0073] After the priority 3 message is sent, the priority 2 message enters the outgoing port for transmission, and then the priority 1 message enters the outgoing port for transmission, thus completing the transmission of the three priority messages.

[0074] Device 2 Ingress Ports:

[0075] After receiving a truncated message with priority 1 at the ingress port of device 2, the device identifies the truncated message identifier, recalculates the message's checksum, and compares it with the checksum included at the end of the message. If they are different, the message is discarded; if they are the same, the device continues to identify the truncation sequence number in the truncation identifier. If the sequence number is 1, it indicates the first truncation. This portion of the message is then buffered in the ingress port queue, and the truncation count is recorded. Buffered messages are buffered sequentially according to the frame sequence number provided in the truncation identifier. The truncated message with priority 1 will be buffered as the first entry in the ingress port buffer queue.

[0076] After receiving a truncated message of priority 2 at the ingress port of device 2, the device identifies the truncated message identifier, recalculates the message's checksum, and compares it with the checksum included at the end of the message. If they are different, the message is discarded; if they are the same, the device continues to identify the truncation sequence number in the truncation identifier. If the sequence number is 1, it indicates the first truncation. This portion of the message is then buffered in the ingress port queue, and the truncation count is recorded. Buffered messages are buffered sequentially according to the frame sequence number provided in the truncation identifier. The truncated message of priority 2 will be buffered in the second entry of the ingress port buffer queue.

[0077] When device 2 receives a complete message with priority 3 at its ingress port, it calculates the checksum and compares the results. If they match, the message is forwarded directly; otherwise, it is discarded.

[0078] After receiving a truncated packet with priority 2 at the ingress port of device 2, the device identifies the truncated packet identifier, recalculates the packet's checksum, and compares it with the checksum at the end of the packet. If they are different, the packet is discarded. If they are the same, the device continues to identify the truncation sequence number in the truncation identifier. If the number is not 1, the device searches for a preceding packet in the corresponding position of the buffer queue. If a preceding packet is found and is consecutive, the packet is buffered in the ingress port queue. The device then determines whether the packet is complete. If it is complete, the packet proceeds to the forwarding stage. If it is incomplete, the device continues to buffer the packet and waits for the remaining packets.

[0079] After receiving a truncated packet with priority 1 at the ingress port of device 2, the device identifies the truncated packet identifier, recalculates the packet's checksum, and compares it with the checksum at the end of the packet. If they are different, the packet is discarded. If they are the same, the device continues to identify the truncated sequence number in the truncated identifier. If the number is not 1, the device searches for a preceding packet in the corresponding position of the buffer queue. If a preceding packet is found and is consecutive, the packet is buffered in the ingress port queue. The device then determines whether the packet is complete. If it is complete, the packet proceeds to the forwarding stage. If it is incomplete, the device continues to buffer the packet and waits for the remaining packets.

[0080] The device's input port 2 completes the reception of 3 frames of messages with different priorities.

[0081] Device 2 output port:

[0082] Device 2 sends priority 3 packets, priority 2 packets, and priority 1 packets in sequence from its output port.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for fast transmission of time-sensitive messages, characterized in that, Includes the following steps: The device's output port sends messages according to message priority. When a low-priority message is sent at the device's output port, a high-priority message arrives at the output port and preempts the device's output port as a preemption message. The truncated portion of the low-priority message is then used as a truncated message, resulting in truncated messages and preemption messages. A truncation identifier and a checksum are added at the truncation point of the truncated message. The truncation identifier is represented by 4 bytes, with the contents of 0xEB, 0xEB, 0xBE and the truncation sequence number in sequence. The truncation sequence number is represented by 1 byte, with the high 4 bits representing the truncated message number, up to a maximum of 16 truncated messages; the low 4 bits represent the number of times the truncated message has been truncated, with a maximum of 16 truncated messages per frame. The checksum contains 2 bytes, performing a CRC check on all bytes of the message content sent before this truncation. Preemptive messages are processed according to whether gapless message transmission is enabled, resulting in processed truncated messages and processed preemptive messages. When preemptive messages are transmitted gaplessly, a 2-byte gapless transmission identifier is inserted before the preemptive message is sent to identify the preemptive message; the contents of the two bytes are 0xEB and 0xEB in sequence. After determining that the received processed truncated message is a truncated message at the ingress port, it is buffered at the ingress port. When the truncated part of a low-priority message arrives at the ingress port, it is concatenated with the buffered processed truncated message and buffered at the ingress port to obtain the buffered complete low-priority message. After receiving the truncated part, the ingress port concatenates it with the received part and buffers it. When the ingress port receives the truncated second-highest priority message, it buffers the message in the ingress buffer. The buffer queue depth of each port is set according to the number of priority queues. The ingress port performs truncation checks and reception identification when it receives a truncated message. After the ingress port determines that the received preemptive message is a complete message, it sends the complete high-priority message and the buffered complete low-priority message in descending order of priority, thus obtaining the complete high-priority message and the complete low-priority message, and realizing the fast transmission of time-sensitive messages.

2. The method for fast transmission of time-sensitive messages according to claim 1, characterized in that: The message contains Ethernet frame format messages and custom format messages, and the message length is greater than 16 bytes.

3. The method for fast transmission of time-sensitive messages according to claim 1, characterized in that: When a low-priority packet is sent to the output port, the port will continue to preempt the higher-priority packet when it arrives.

4. The method for fast transmission of time-sensitive messages according to claim 1, characterized in that: The setting for high-priority messages determines whether they are sent without gaps with truncated messages. High-priority messages are preemptive messages. When set to be sent without gaps, the information between the preemptive message and the truncated message does not include the frame gap and preamble. When set to be sent with gaps, the information between the preemptive message and the truncated message includes the frame gap and preamble.

5. The method for fast transmission of time-sensitive messages according to claim 1, characterized in that: Set up a buffer queue for each ingress port. The length of each buffer entry is the maximum allowed packet length. The buffer entry length can be adjusted by setting it.

6. A system for rapidly sending time-sensitive messages, used to run a method for rapidly sending time-sensitive messages according to any one of claims 1-5, characterized in that, include: The message transmission truncation module is used to send messages according to message priority at the device's output port. When a low-priority message is sent at the device's output port, a high-priority message will preempt the device's output port as a preemption message. The truncated portion of the low-priority message will be used as a truncated message, thus obtaining the truncated message and the preemption message. The message preemption transmission module is used to add a truncation mark and a check code at the truncation point of the truncated message, and to process the preempted message according to whether the message is sent without gaps, so as to obtain the processed truncated message and the processed preempted message. The message caching module is used to cache the processed truncated message received at the ingress port after determining that it is a truncated message. When the truncated part of a low-priority message arrives at the ingress port, it is concatenated with the cached processed truncated message and cached at the ingress port to obtain the cached complete low-priority message. The message splicing and sending module is used to determine whether the received preemptive message after processing is a complete message at the ingress port. Then, it sends the complete high-priority message and the buffered complete low-priority message in descending order of priority, thus obtaining the complete high-priority message and the complete low-priority message, enabling the fast sending of time-sensitive messages.

Images