A method and apparatus for transmitting and receiving multiple messages

Abstract

This application discloses a transmission method and apparatus for multiple-transmission and selective-reception of messages. The method includes: receiving an original Ethernet message from a southbound interface; determining the number of copies and the sending interface according to path rules; generating a custom protocol header for each copy, containing a device identifier and a data frame sequence number, wherein the combination of the device identifier and the data frame sequence number is used to uniquely identify the original message; encapsulating the custom protocol header and the original message into a User Datagram Protocol (UDP) message; and sending multiple copies through a designated interface. The receiving end performs deduplication based on the combined identifier, and only decapsulates the first arriving message copy before forwarding it to the upper layer. This application achieves highly reliable, low-latency transparent data transmission through controllable replication at the sending end and precise selective reception at the receiving end, and can dynamically optimize redundancy based on link quality, saving network resources.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a transmission method and apparatus for multiple message sending and receiving. Background Technology

[0002] In modern communication systems, especially in wireless networks or hybrid network environments composed of various types of network links, the reliability and real-time performance of data transmission face severe challenges. A single communication link is often affected by factors such as signal interference, network congestion, and device switching, leading to problems such as packet loss, out-of-order delivery, and latency jitter. These issues can severely impact the performance of upper-layer business applications. In scenarios such as remote control, video surveillance, and drone swarm communication, data transmission failures or delays can have serious consequences.

[0003] To address these challenges, existing technologies have proposed several solutions. Transmission Control Protocols (TCP) ensure data reliability through acknowledgment and retransmission mechanisms, but the retransmission process introduces significant transmission delays, making it unsuitable for services with high real-time requirements. Another solution is link aggregation, which bundles multiple physical links into a single logical link to increase bandwidth and reliability. However, traditional link aggregation schemes are typically implemented at lower levels of the network protocol stack, resulting in complex configurations and a lack of transparency to upper-layer applications, potentially requiring modifications to the network configuration of existing service devices. Furthermore, many simple redundant transmission schemes employ static configurations, failing to dynamically adjust redundancy based on the real-time quality of network links. This can lead to unnecessary bandwidth waste when network conditions are good, while under deteriorating network conditions, it may not provide sufficient redundancy to guarantee communication quality.

[0004] Therefore, how to design a transmission method that can transparently access existing services, effectively combat the negative impact of unstable links, and intelligently optimize resource utilization based on real-time network conditions is a technical problem that urgently needs to be solved in the current technology field. Summary of the Invention

[0005] The purpose of this invention is to provide a method and apparatus for transmitting multiple messages and selectively receiving them, so as to solve the problems of low data transmission reliability, poor real-time performance and low resource utilization in mixed or unstable network environments mentioned in the background art.

[0006] To achieve the above objectives, the first aspect of the present invention provides a transmission method for multiple transmission and selective reception of messages. This method is executed on a sending device and is characterized by comprising: receiving an original Ethernet message sent by a service device from a southbound interface; determining, according to a preset path rule, the target number of copies of the original Ethernet message and the northbound network interface corresponding to each copy; generating a custom protocol header for each copy of the original Ethernet message, the custom protocol header containing a device identifier of the sending end and an incrementing data frame sequence number for the original Ethernet message, the combination of the device identifier and the data frame sequence number being used to uniquely identify the original Ethernet message; encapsulating the custom protocol header and the original Ethernet message as payload together into a User Datagram Protocol (UDP) message to form an encapsulated message; and sending multiple copies of the generated encapsulated message to one or more networks through the northbound network interface determined by the path rule.

[0007] A second aspect of the present invention provides a transmission method for selective message reception, which is executed on a receiving device. The method includes: receiving multiple encapsulated messages from one or more network interfaces, wherein the encapsulated messages are User Datagram Protocol (UDP) messages containing original Ethernet messages; parsing the encapsulated messages, extracting a custom protocol header, and obtaining the sending device identifier and data frame sequence number from the custom protocol header; identifying the received encapsulated messages based on the combination of the device identifier and the data frame sequence number; maintaining a set of identifier records for successfully received messages; when an encapsulated message is received, checking whether the corresponding combination of the device identifier and the data frame sequence number exists in the identifier record set; if it does not exist, adding the combination to the identifier record set, decapsulating the encapsulated message, extracting the original Ethernet message, and forwarding it to an upper-layer application; if it already exists, directly discarding the currently received encapsulated message.

[0008] The present invention also provides a message transmission apparatus and system for implementing the above method.

[0009] The technical solution provided in this application has the following beneficial effects: By copying the same original message multiple times and transmitting them concurrently through different network links, and employing the "fastest arrives" selection logic at the receiving end, it can effectively combat sudden packet loss and instantaneous congestion on a single link. Even if some links are interrupted or severely degraded, data can still be successfully delivered through other normal links, thereby improving the reliability of data transmission. Simultaneously, since the system always selects the earliest arriving message copy, the end-to-end transmission delay can approach the optimal delay among all currently available links. This application adopts an encapsulation method above Layer 2, transmitting the complete original Ethernet message as the payload. This mechanism is completely transparent to the user's original business data and network protocol stack; business devices can access the system without any software or hardware modifications or network configuration changes, exhibiting strong compatibility and ease of deployment. By introducing a dynamic redundancy scheduling mechanism, the sending end can intelligently adjust the number of message copies and the transmission path based on real-time link quality information fed back from the receiving end and the current service quality level of the service. When network conditions are good, the system can reduce unnecessary redundant copies, conserving bandwidth resources; when network quality deteriorates, it increases redundancy to ensure communication. This adaptive capability can significantly reduce operating costs on pay-per-traffic networks. Based on real-time link race feedback from the receiver, the system can achieve millisecond-level perception and path adjustment of dynamic changes in the network environment. When a link experiences increased latency due to network switching, the system can quickly detect a decrease in its "first-to-reach probability" and automatically switch high-priority data streams to a better link, thereby ensuring service continuity and quality of service. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a structural block diagram of the message transmission device provided in an embodiment of the present invention.

[0012] Figure 2 This is a flowchart of a message multiple transmission and selective reception method provided in an embodiment of the present invention.

[0013] Figure 3 This is an overall system block diagram in one implementation scenario of the present invention.

[0014] Figure 4 This is a schematic diagram of the registration message format in an embodiment of the present invention.

[0015] Figure 5 This is a schematic diagram of the format of the client-encapsulated message in an embodiment of the present invention.

[0016] Figure 6 This is a schematic diagram of the format of the controller encapsulating the message in an embodiment of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below with reference to specific embodiments.

[0019] Example 1

[0020] This embodiment provides a message multiplexing and selective receiving transmission system, which is used to achieve transparent data interoperability in multiple network environments. For example... Figure 3 As shown, the system can consist of one or more multi-transmit / receive controllers, multiple multi-transmit / receive clients, and multiple networks to which they are connected. The multi-transmit / receive clients connect to various service devices through their southbound interfaces and to multiple networks through their northbound interfaces. These networks can be any physical network capable of transmitting Ethernet packets, such as 5G networks, wireless networks, wired Ethernet, or satellite networks, or logical networks partitioned using technologies such as Virtual LANs. The system aims to achieve highly reliable, low-latency, and transparent data transmission between any two service devices.

[0021] The message transmission method of this embodiment can be executed in a message transmission device. For example... Figure 1 As shown, this device can serve as a multi-transmitter / receiver client or controller in the aforementioned system. The device may include a processor 101, a memory 102, and a communication interface 103 connected to the processor 101 and the memory 102. The memory 102 stores program instructions, and the processor 101 executes the program instructions stored in the memory 102 to implement the various steps in the following methods. The communication interface 103 includes a southbound interface for connecting service devices and a northbound network interface for connecting multiple networks.

[0022] In a specific implementation, the device can employ a heterogeneous processor system architecture. For example, processor 101 can be a system-on-a-chip that integrates a central processing unit (CPU) and a field-programmable gate array (FPGA). The CPU is responsible for performing complex control plane tasks, such as managing path rules, signaling interactions with other devices in the network, and calculating dynamic scheduling logic. The FPGA, utilizing its parallel processing capabilities, is responsible for performing high-speed data plane processing, such as packet encapsulation and decapsulation, and hardware-based fast lookup and deduplication. Data exchange between the two occurs via a high-speed bus. The system also includes a power supply module to power the entire device, non-volatile memory storing firmware and configuration information, and main memory providing temporary data storage space for system operation. Each network interface is equipped with a corresponding physical layer transceiver to enable connectivity to the physical network.

[0023] The following example will describe in detail the message sending and receiving process using a multi-transmitter selective receiver client as the sender and a multi-transmitter selective receiver controller as the receiver.

[0024] Sender processing flow (taking the client as an example):

[0025] Combination Figure 2 The message multiple transmission and selective reception method provided in this embodiment may specifically include the following steps:

[0026] After the client device powers on, it needs to establish a connection with the controller in the network and be aware of the status of each network link. The client periodically sends registration messages to the controller through its various northbound network interfaces. For example... Figure 4 As shown, the registration message is a User Datagram Protocol (UDP) message. Its content includes not only the source / destination network address and port number used for routing, but also a specific registration message identifier, the client's own device identifier, and a network address-physical address mapping table and multicast table of the southbound service devices to which the client is connected. In this way, the controller can learn the entire network topology and device reachability information.

[0027] To achieve sensitive awareness of link status, the transmission frequency of registration messages is dynamically adjustable. In a preferred embodiment, when a northbound interface fails to receive a registration feedback message from the controller for an extended period, the system determines that the link may have a problem and will send registration messages at a higher frequency for probing. Once a registration feedback message is successfully received, confirming that the link has returned to normal, the transmission frequency will be reduced to a lower maintenance frequency to minimize unnecessary signaling overhead. If no feedback message is received from a specific network link within a preset time period, the client will update the status of that specific network link to "interrupted."

[0028] In a preferred implementation, the registration message may also include device information for network management, such as a device type field, firmware version number field, and bandwidth capacity indication field for each northbound network interface. This information helps the controller perform more granular network management and resource scheduling.

[0029] The client's southbound interface is responsible for receiving raw Ethernet packets from the service device. After receiving the packets, a cyclic redundancy check is performed to ensure packet integrity; packets that fail the check are discarded.

[0030] Passed messages are sent to the path selection module. This module determines the target number of copies of the original Ethernet message and which northbound network interface each copy should be sent through, based on preset path rules.

[0031] In a basic implementation, path rules are statically configured. When the path selection module detects that a link is interrupted, it switches the data stream that was originally destined for that link to a backup, functioning network link, according to preset switching logic. This mechanism ensures communication continuity.

[0032] In a more advanced implementation, path rules are dynamically generated. The sender receives link quality feedback from the receiver. This feedback indicates the real-time data transmission performance of each network link. Based on this feedback and the current service quality level of the transmission traffic, the sender dynamically calculates the required target replication count. For example, a larger baseline replication count can be set for control signaling with high real-time requirements, while a smaller baseline replication count can be set for non-critical log data.

[0033] Subsequently, the system calculates a comprehensive quality score based on link quality feedback. The final target replication quantity can be determined using a preset adjustment function. This function is designed to appropriately increase the target replication quantity to improve redundancy when the comprehensive quality score is low. After determining the replication quantity, the system selects links with higher transmission performance from all available network links as the transmission paths for the replicas. The sending end method refresh period can be set to once every 50 milliseconds to achieve rapid response to network changes.

[0034] Specifically, the sending end can refresh the lookup table of its distribution logic based on different service categories and link quality scores.

[0035] After determining the distribution logic, the data encapsulation module begins processing the original packets. For each copy to be sent, the module generates a custom protocol header. The core content of this header includes two fields: the sender's device identifier and a monotonically increasing sequence number. The device identifier uniquely identifies the sending device across the entire network, while the sequence number increments for each original packet. The combination of the device identifier and the sequence number forms a globally unique identifier used at the receiving end to identify and deduplicate multiple copies of the same original packet from different paths.

[0036] Next, the system encapsulates this custom protocol header and the original Ethernet packet as payload into a User Datagram Protocol (UDP) packet. For example... Figure 5 As shown, the final encapsulated message structure consists of: an external network protocol header, a user datagram protocol header, a custom protocol header, and the complete original Ethernet message.

[0037] After encapsulation, the system sends multiple copies of the encapsulated message to the network through their respective northbound network interfaces, according to the determined path rules.

[0038] In hardware-accelerated solutions, the high-speed data processing mentioned above, such as generating custom protocol headers and encapsulation, is all performed by the hardware logic of the Field-Programmable Gate Array (FPGA). An internal lookup table-driven distributor can be designed within the FPGA; this lookup table is updated in real-time by the central processing unit according to dynamic scheduling logic. Based on the lookup table results, the FPGA copies and distributes the original packet to the corresponding northbound interface's transmission queue, achieving low processing latency.

[0039] Through the above transmission process, an original message is converted into multiple encapsulated messages with unique identifiers, and sent to the receiving end in parallel through multiple links, laying the foundation for subsequent selective reception and reliability recovery.

[0040] Example 2

[0041] This embodiment describes the message receiving end processing flow, which can be executed on the message transmission device described in Embodiment 1, which acts as the receiving end in this scenario. The following explanation will use the controller as the receiving end and the client as the sending end as an example.

[0042] Receiver processing flow (taking the controller as an example):

[0043] The controller's northbound network interface receives data packets from one or more networks. It checks the physical address header and cyclic redundancy check (CRC) of the packets, discarding packets with mismatched destination addresses or CRC errors.

[0044] For messages that pass verification, the data decapsulation module will determine their type. If the message is a registration message sent by the client, it will be handled by the registration module.

[0045] If the message is determined to be an encapsulated message, the module will continue parsing to extract the custom protocol header and obtain the crucial sender device identifier and data frame sequence number. If the message is not an encapsulated message defined by this system, it will be discarded directly.

[0046] The receiving end maintains a set of identifier records for successfully received messages. In a preferred embodiment, this set employs a circular buffer structure (e.g., a circular record table with a depth of 1024). The system calculates a hash index using a combination of the device identifier and the data frame sequence number, and writes the reception status to the corresponding position in the buffer. As new data frame sequence numbers (FNs) are continuously added, the circular buffer automatically overwrites the oldest record, thereby achieving automatic record aging and memory space reuse without the need for additional clearing operations.

[0047] When a new encapsulated packet arrives and its combined identifier is parsed, the system checks if this combination exists in the identifier record set. If it doesn't exist, this indicates that it is the first arriving copy corresponding to the original packet. The system considers it the "fastest arriving" packet. At this point, the system performs two actions: adding this new combined identifier to the identifier record set; and decapsulating the encapsulated packet, i.e., stripping off the outer protocol header to restore the original Ethernet packet, and looking up the local address table based on the destination physical address in the original packet, forwarding it to the corresponding southbound interface. If it already exists, this indicates that one or more copies corresponding to the same original packet have arrived and been processed before the current packet. The current packet is a redundant duplicate packet, so the system discards it directly.

[0048] With this mechanism, no matter how many copies the sender sends, and no matter the order in which these copies arrive after being transmitted over the network, the receiver can always forward the first copy of the data to the upper layer, thus solving the problems of message duplication and out-of-order delivery and achieving reliable transmission.

[0049] In the hardware-accelerated solution, the deduplication and selection logic described above is implemented in hardware using a field-programmable gate array (FPGA). A hardware hash table can be constructed using the block memory within the FPGA. Whenever an encapsulated packet arrives, its combined identifier is hashed and used for table lookup. Both lookup and update operations can be performed at line speed in hardware, achieving high processing throughput.

[0050] To support dynamic redundancy scheduling at the transmitting end, the receiving end is also responsible for evaluating the performance of each receiving link and generating link quality feedback information.

[0051] In a preferred embodiment, the receiving end counts the number of times each received encapsulated message becomes a "first-delivery message" with a specific combination of identifiers within a preset sliding time window for each receiving network interface. For example, the hardware hash deduplication module can maintain a circular race record table. Whenever a new data frame sequence number arrives, the hardware logic records which physical interface received the first-delivery message.

[0052] At the end of each statistical cycle, the central processing unit reads this record table and calculates the first-to-delivery probability of a packet for each receiving network interface. This probability directly reflects the transmission speed advantage of the link.

[0053] To more comprehensively assess link quality, in addition to the first-to-reach probability, the system also measures and records the jitter value of the time interval between consecutive packets on each receiving network interface.

[0054] Finally, the system employs a pre-defined quality scoring algorithm to integrate multiple performance dimensions into a comprehensive quality score. For example, the link quality score can be calculated by weighting the proportion of times the link was the first-to-reach link out of the total number of counts, and the link's latency jitter measurement value. The first-to-reach proportion has a positive weight, while the jitter value has a negative impact on the score.

[0055] The calculated overall quality score for each link serves as the core content of the link quality feedback information, and is sent to the corresponding sender via a specific control message. The specific transmission method can be flexibly chosen: it can be encapsulated in a separate control message or embedded in the reverse service data message sent to that sender, thus reducing the bandwidth consumption of pure signaling messages. For example... Figure 6 As shown, in a preferred embodiment of the present invention, when the controller acts as a receiver and needs to send service data back to the sender, its encapsulated message structure sequentially includes: an external network protocol header, a User Datagram Protocol header (UDP header), a custom protocol header, a link quality feedback information field, and a complete original Ethernet message. The link quality feedback information field, located after the custom protocol header, carries the calculated comprehensive quality score for each link, thereby achieving dynamic redundancy scheduling at the sender without increasing the control link burden. This forms a closed-loop control system, enabling the redundancy method at the sender to adapt to network changes in real time. It is expected that this dynamic scheduling mechanism can reduce unnecessary redundant traffic by approximately 40% to 60% while ensuring high reliability.

[0056] In summary, this application constructs an efficient, transparent, and reliable cross-network transmission scheme by performing controllable message copying and encapsulation at the sending end and precise deduplication and selection based on unique identifiers at the receiving end. Furthermore, by introducing a closed-loop feedback mechanism for link quality from the receiving end to the sending end and a dynamic scheduling algorithm, the system can achieve intelligent and refined management of network resources while ensuring communication quality, thus possessing high practical value.

[0057] It should be noted that the various specific technical features described in the above embodiments can be combined with each other without contradiction. Furthermore, it will be apparent to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. Reference numerals in any of the drawings should not be construed as limiting the scope of the claims.

Claims

1. A transmission method for multiple sending and selective receiving of messages, characterized in that, include: Receive raw Ethernet packets sent by service devices from the southbound interface; Based on preset path rules, determine the target number of copies of the original Ethernet packet and the northbound network interface corresponding to each copy; A custom protocol header is generated for each copy of the original Ethernet packet. The custom protocol header contains the device identifier of the sending end and an incrementing data frame sequence number for the original Ethernet packet. The combination of the device identifier and the data frame sequence number is used to uniquely identify the original Ethernet packet. The custom protocol header and the original Ethernet packet as payload are encapsulated together into a User Datagram Protocol (UDP) packet to form an encapsulated packet. The generated multiple copies of the encapsulated message are sent to one or more networks through the northbound network interface determined by the path rules.

2. The method according to claim 1, characterized in that, The method further includes: The system periodically sends registration messages to the controller in the network through the northbound network interface and listens for the feedback messages from the controller to detect the connectivity status of the network links corresponding to each northbound network interface. If no feedback message is received from a specific network link within a preset time period, the status of the specific network link will be updated to interrupted. In the path rule, the data flow is switched from the specific network link that is in an interrupted state to a preset backup network link.

3. The method according to claim 2, characterized in that, The registration message contains device information for network management, including a device type field, a firmware version number field, and a bandwidth capacity indication field for each northbound network interface. The step of updating the status of the specific network link to interruption specifically includes: if no feedback message for the specific network link is received from the controller for multiple consecutive transmission cycles, then the specific network link is determined to be interrupted.

4. The method according to claim 1, characterized in that, The steps of determining the target number of copies of the original Ethernet packet and the northbound network interface corresponding to each copy specifically include: Receive link quality feedback information fed back by the receiving end, the link quality feedback information indicating the data transmission performance of each network link; The target number of copies is dynamically calculated based on the link quality feedback information and the service quality level of the current transmission service. From all available network links, select the network link with higher transmission performance as the transmission path for the northbound network interface corresponding to the replica.

5. The method according to claim 4, characterized in that, The step of dynamically calculating the target number of copies includes: Different baseline replication quantities are preset for different service quality levels, with services requiring high real-time performance corresponding to a larger baseline replication quantity. A comprehensive quality score is calculated based on the link quality feedback information. Based on the baseline replication count and the overall quality score, the final target replication count is determined by a preset adjustment function, wherein the adjustment function increases the target replication count when the overall quality score is low.

6. The method according to claim 1, characterized in that, The method is executed by a heterogeneous processor system, which includes a central processing unit and a field-programmable gate array; The hardware logic of the field-programmable gate array is responsible for performing high-speed data plane processing, including generating the custom protocol header for each copy and encapsulating the custom protocol header and the original Ethernet packet into the User Datagram Protocol (UDP) packet. The central processing unit is responsible for performing control plane processing, including managing the path rules and communicating with controllers in the network.

7. A method for selectively receiving messages during transmission, characterized in that, include: Receive multiple encapsulated messages from one or more network interfaces, wherein the encapsulated message is a User Datagram Protocol message containing the original Ethernet message; The encapsulated message is parsed, the custom protocol header is extracted, and the device identifier and data frame sequence number of the sending end are obtained from the custom protocol header; The received encapsulated message is identified based on the combination of the device identifier and the data frame sequence number; Maintain a set of identifier records for successfully received messages. When an encapsulated message is received, check whether the combination of the corresponding device identifier and the data frame sequence number exists in the set of identifier records. If it does not exist, the combination is added to the identifier record set, and the encapsulated message is decapsulated to extract the original Ethernet message and forward it to the upper layer application. If the encapsulated message already exists, discard the currently received message.

8. The method according to claim 7, characterized in that, The method further includes: Within a preset sliding time window, for each receiving network interface, the number of times the encapsulated message received by it becomes the first message with a specific device identifier and data frame sequence number combination is counted. Based on the number of times, calculate the first-pass probability of each receiving network interface, and generate link quality feedback information containing the first-pass probability of each interface. The link quality feedback information is sent to the corresponding sending end.

9. The method according to claim 8, characterized in that, The step of generating the link quality feedback information further includes: Measure and record the jitter value of the time interval between consecutive messages received by each receiving network interface; A preset quality scoring algorithm is used to calculate the comprehensive quality score of each interface. The calculation method of the quality scoring algorithm is to calculate the comprehensive quality score based on the weighted calculation of the first-to-reach probability and the time interval jitter value, wherein the weight of the first-to-reach probability is a positive value and the weight of the time interval jitter value is a non-positive value. The overall quality score is used as part of the link quality feedback information.

10. The method according to claim 7, characterized in that, The method is executed by a heterogeneous processor system, which includes a central processing unit and a field-programmable gate array; The hardware logic of the field-programmable gate array is responsible for executing high-speed data plane processing, including fast lookup and deduplication of the combination of device identifier and data frame sequence number based on hardware hash table, as well as racing statistics of first-delivery messages. The central processing unit is responsible for performing control plane processing, including generating link quality feedback information based on the racing statistics and sending it to the sending end.

11. A transmission device for multiple transmission and selective reception of messages, characterized in that, include: The receiving module is used to receive raw Ethernet packets sent by service devices from the southbound interface; The path determination module is used to determine the target number of copies of the original Ethernet packet and the northbound network interface corresponding to each copy according to preset path rules. The header information generation module is used to generate a custom protocol header for each copy of the original Ethernet packet. The custom protocol header includes the device identifier of the sending end and an incrementing data frame sequence number for the original Ethernet packet. The combination of the device identifier and the data frame sequence number is used to uniquely identify the original Ethernet packet. The encapsulation module is used to encapsulate the custom protocol header and the original Ethernet packet as payload together into a User Datagram Protocol (UDP) packet, forming an encapsulated packet. The sending module is used to send multiple copies of the encapsulated message to one or more networks via the northbound network interface determined by the path rules.

12. A message selection and reception transmission device, characterized in that, include: The receiving module is configured to receive multiple encapsulated messages from one or more network interfaces, wherein the encapsulated messages are User Datagram Protocol messages containing original Ethernet messages; The parsing module is used to parse the encapsulated message, extract the custom protocol header, and obtain the device identifier and data frame sequence number of the sending end from the custom protocol header; The identification module is used to identify the received encapsulated message based on the combination of the device identifier and the data frame sequence number; The recording module is used to maintain a set of identifier records for successfully received messages. When an encapsulated message is received, it checks whether the combination of the corresponding device identifier and the data frame sequence number exists in the identifier record set. The processing module is configured to add the combination to the identifier record set if it does not exist, and decapsulate the encapsulated message to extract the original Ethernet message and forward it to the upper-layer application; if it already exists, the currently received encapsulated message is directly discarded.

13. A message transmission device, comprising a memory and a processor, characterized in that, The memory is used to store program instructions, and the processor is used to execute the program instructions to implement the message multiplexing and selective receiving transmission method as described in any one of claims 1-10.

14. A computer-readable storage medium having program instructions stored thereon, characterized in that, When the program instructions are executed by the processor, they implement the message multiplexing and selective receiving transmission method as described in any one of claims 1-10.

Images