A reliable communication method based on master-slave polling and data grading

Abstract

The application discloses a reliable communication method based on master-slave polling and data grading, and belongs to the technical field of communication. In the existing half-duplex communication system, the system is unstable or even collapses due to the transceiver conflict and blind retransmission in the wireless relay high packet loss environment. The method comprises the following steps: dividing the communication data into unacceptable discarded data and acceptable discarded data, and only performing retransmission management with a serial number for the unacceptable discarded data; a master device sends a polling message according to time slots, and carries the latest message serial number of a slave device received; the slave device responds after receiving the polling message, and carries the latest message serial number of the master device received; both sides judge the message delivery state according to the serial number fed back by the opposite end, and decide whether to remove or resend the message from the sending queue. Through the data grading and master-slave polling mechanism, the half-duplex communication conflict is effectively avoided, important data can be reliably delivered in the extreme environment with a packet loss rate of more than 70%, and the safety and stability of the system are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of industrial control communication technology. More specifically, it relates to a reliable communication method based on master-slave polling and data hierarchy. Background Technology

[0002] Serial communication buses play a crucial role in industrial control, building automation, medical equipment, and energy management. The RS-485 bus, with its differential signal transmission at the physical layer, possesses strong common-mode interference immunity and supports long-distance communication and multi-node networking, making it one of the mainstream communication interfaces in the fieldbus layer. However, as application scenarios evolve towards complex electromagnetic environments, especially in hybrid link scenarios involving wireless relays (such as Bluetooth and Wi-Fi pass-through), the unreliability of communication links increases significantly. In systems with extremely high real-time and safety requirements, such as high-pressure injectors for medical devices, ensuring reliable transmission of critical commands in a half-duplex communication environment with high packet loss rates has become a core problem urgently needing to be solved by those skilled in the art.

[0003] In existing technologies, a common solution is a communication method based on application-layer acknowledgment and timeout retransmission mechanisms. This approach typically builds an application-layer protocol on top of the standard RS-485 hardware layer. After sending a data frame, the sending end starts a timer to wait for an acknowledgment frame from the receiving end. If no acknowledgment frame is received within a preset timeout period, the sending end triggers retransmission logic to retransmit the data frame. A maximum number of retransmissions is usually set to prevent infinite loops. If the maximum number of retransmissions is exceeded without success, the link is considered interrupted or the communication connection is reset. This approach performs well in light-load, low-interference laboratory environments and can solve occasional random packet loss problems. However, this approach has the following technical drawbacks: First, it uses a stop-and-wait protocol, meaning the sending end cannot send new data while waiting for an acknowledgment frame, resulting in extremely low channel utilization. Second, this approach adopts a uniform reliable transmission strategy for all application data, without classifying data importance. In scenarios with severe interference, when a large amount of non-critical periodic status data is lost and triggers retransmission, it quickly consumes limited bandwidth resources, causing communication channel congestion. At this point, the truly critical emergency control commands cannot be issued in a timely manner due to queuing, leading to system response delays or even loss of control.

[0004] Another technical solution is a half-duplex collision avoidance scheme based on Carrier Sense Multiple Access (CSMA) or Token Ring mechanisms. This scheme attempts to avoid data collisions on the RS-485 bus through software-level carrier sensing. Specifically, the device listens to the bus level before sending data; if the bus is busy, it randomly delays for a period of time before attempting to send. Some improved schemes also introduce a simple polling mechanism, where the master node polls the slave nodes, and the slave nodes only reply with data when they receive a polling command. While these schemes alleviate bus collision problems to some extent, they still have significant shortcomings. On the one hand, a simple polling mechanism usually only solves the problem of channel access rights allocation, but does not solve the problem of data transmission reliability. If the polling frame or response frame is damaged due to interference during transmission, the receiving end often cannot identify lost or duplicate packets due to the lack of an effective sequence number acknowledgment mechanism, resulting in compromised data integrity. On the other hand, in order to improve reliability, such solutions often need to introduce independent acknowledgment frame interactions. In half-duplex communication mode, frequent send and receive switching increases bus switching latency, and the independent acknowledgment frames themselves are also at risk of being lost, which can easily lead to repeated transmission due to lost acknowledgments, further exacerbating bandwidth waste.

[0005] Therefore, designing a communication method that can ensure the reliable delivery of critical data, maximize the utilization of bandwidth resources, and has a low-overhead acknowledgment mechanism in complex environments with half-duplex communication and high packet loss rates is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0006] The purpose of this invention is to provide a reliable communication method based on master-slave polling and data hierarchy. This method combines data hierarchy with an implicit acknowledgment mechanism of message sequence numbers, which significantly improves bandwidth utilization and system response speed while ensuring the reliable delivery of critical data.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: A reliable communication method based on master-slave polling and data hierarchy is applied to a communication system including a master device and a slave device, wherein the master device and the slave device are connected via a half-duplex communication bus; the method includes the following steps: Step 1: Divide the communication data to be sent between the master device and the slave device into unacceptable discardable data and acceptable discardable data. The unacceptable discardable data located in the master device's sending queue carries a first message sequence number, and the unacceptable discardable data located in the slave device's sending queue carries a second message sequence number. Step 2: The master device sends the first message or polling message in the unacceptable discard message queue to the slave device according to a predetermined time slot. The first message in the unacceptable discard message queue carries a first message sequence number, and the polling message carries a second message sequence number of the unacceptable discard data that the master device has successfully received from the slave device. Step 3: After receiving the polling message, the slave device sends the first message in the unacceptable discard message queue or the polling response to the master device. The first message in the unacceptable discard message queue carries a second message sequence number, and the polling response carries the first message sequence number of the unacceptable discard data that the slave device has successfully received from the master device. Step 4: The master device determines whether the slave device has successfully received the latest unacceptable discarded data sent by the master device based on the first message sequence number carried in the received polling response. If the determination is yes, the corresponding unacceptable discarded data is removed from the master device's sending queue. If the determination is no, the corresponding unacceptable discarded data is resent. Step 5: The slave device determines whether the master device has successfully received the latest unacceptable discarded data sent by the slave device based on the second message sequence number carried in the received polling message. If the determination is yes, the corresponding unacceptable discarded data is removed from the slave device's sending queue. If the determination is no, the corresponding unacceptable discarded data is resent.

[0008] Furthermore, the communication data is divided into unacceptable discardable data and acceptable discardable data, including: identifying the service type of the communication data; marking data whose service type is control command or operation response as unacceptable discardable data; marking data whose service type is status reporting request and status reporting as acceptable discardable data; for the acceptable discardable data, the master and slave devices send it directly without storing it in the master and slave device sending queue, and without performing retransmission processing.

[0009] Furthermore, before the step of the master device sending polling messages to the slave device according to a predetermined time slot, the method further includes: the master device dividing the communication time slot to prevent half-duplex communication transmission and reception conflicts; the master device sending a status reporting request in a fixed time slot (the first 0.1s) within a complete polling cycle (e.g., 1s); and in other time slots of a complete polling cycle, if there is unacceptable discarded data in the master device's sending queue, then the unacceptable discarded data is sent; if the master device's sending queue is empty, then the polling message is sent.

[0010] Furthermore, the polling message / status reporting request belongs to acceptable discardable data; after receiving the status reporting request, the slave device immediately sends a status report to the master device; after receiving the polling message, the slave device checks whether there is unacceptable discardable data in the slave device's sending queue; if there is, the unacceptable discardable data is sent to the master device as a response to the polling message; if there is no such data, a preset polling response is sent to the master device, and the polling response carries the first message sequence number of the master device that has been successfully received.

[0011] Furthermore, the first message sequence number and the second message sequence number carried by the unacceptable discarded data are numbered according to an auto-incrementing or auto-decrementing pattern.

[0012] Compared with the prior art, the present invention has the following beneficial effects: (1) Because the present invention adopts a time slot allocation and polling mechanism dominated by the master device, the transmission right in the half-duplex communication system is strictly controlled. Compared with the existing CSMA / CD competition mechanism or simple timeout retransmission mechanism, the present invention completely eliminates the bus conflict and blocking problem caused by simultaneous transmission and reception of devices, and ensures the stability of the communication link from the physical level.

[0013] (2) By setting up a data classification mechanism, communication data is divided into unacceptable discardable data and acceptable discardable data, and an unacceptable discardable data (such as status reporting) is sent directly without retransmission. This feature ensures that in extreme environments with high packet loss rates, the system will not exhaust bandwidth resources due to the retransmission of non-critical data, thereby avoiding channel congestion and ensuring that the system can maintain basic communication capabilities even with a packet loss rate of 50% or even 70%.

[0014] (3) Because the present invention adopts a message sequence number-based confirmation mechanism, it achieves accurate closed-loop confirmation of the critical data transmission status by carrying the latest message sequence number received by the peer in the polling message and polling response. Compared with the simple ACK confirmation in the prior art, the present invention can accurately determine whether the peer has received the latest message, avoid misjudgment caused by ACK loss, ensure that unacceptable discarded data arrives in order without duplication or omission, and greatly improve the security of critical scenarios such as medical equipment.

[0015] (4) By setting the polling message to accept discarded data and using the polling frame to carry the receiving status sequence number, the present invention achieves the multiplexing of link keep-alive and data acknowledgment. Even when there is no service data to be sent, the master device can detect the link status through periodic polling and synchronize the receiving status, which reduces the overhead of the communication protocol and ensures the real-time synchronization of the system status.

[0016] (5) It has the characteristics of high cohesion and low coupling, which makes it easy to be ported and deployed in existing embedded medical devices or industrial control terminals, and has wide applicability and engineering application value. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall process of a reliable communication method based on master-slave polling and data hierarchy.

[0018] Figure 2 A timing diagram showing the polling and response sequence number interaction between the master and slave devices. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0020] This invention provides a reliable communication method, apparatus, and medium based on master-slave polling and data hierarchy. Its core concept is to address the high packet loss and collision issues caused by half-duplex communication (such as RS-485) combined with wireless relay links by employing a master-device-led time-slot scheduling strategy to physically eliminate transmit / receive conflicts. Simultaneously, a data hierarchy mechanism based on service importance is introduced, ensuring reliable transmission management with sequence numbers only for critical control data, while implementing "best-effort" transmission for non-critical status data. This guarantees zero loss of core control commands and system stability even in extremely harsh network environments.

[0021] like Figure 1 and Figure 2 As shown, this embodiment provides a reliable communication method based on master-slave polling and data hierarchy, including data hierarchy processing, master-slave polling scheduling, sequence number synchronization, and queue management. First, in the data hierarchy processing stage, the master and slave devices assess the importance of the data to be transmitted, dividing it into unacceptable discardable data (critical data) and acceptable discardable data (non-critical data). Second, in the master-slave polling scheduling stage, the master device, acting as the communication control core, divides the timeline into fixed communication time slots and actively initiates communication in each time slot. The slave device is only allowed to send a response when it receives a polling message or data frame from the master device, thus fundamentally avoiding bus conflicts in half-duplex communication. Third, in the sequence number synchronization stage, both communicating parties achieve real-time confirmation of the bidirectional reception status by carrying the sequence number of the latest successfully received unacceptable discardable data in the polling frame and response frame. Finally, in the queue management stage, the sending end determines whether the data has been delivered based on the sequence number fed back by the receiving end, and decides whether to remove the data from the sending queue or trigger a retransmission, ensuring the reliable delivery of critical data.

[0022] In this embodiment, both the master and slave devices are equipped with data classification functions to categorize and label communication data sent from the application layer. Specifically, when a piece of data to be sent is generated, the function code field in the frame header of the data is parsed. If the function code field indicates that the data is a "control command" (such as start injection, stop injection, parameter setting) or an "operation response" (such as command execution success, command execution failure), the data is marked as "unacceptable discard data". This type of data is directly related to the safe operation and functional implementation of the device and must be delivered to the other end accurately. Therefore, it needs to enter the transmission queue for reliability management. If the function code field indicates that the data is a "polling message" or a "status report" (such as current flow rate, remaining volume, pressure information, etc.), the data is marked as "acceptable discard data". This type of data is usually periodically transmitted at high frequency. Even if it is occasionally lost, subsequent new data will overwrite the old data and will not affect the eventual consistency of the system. Therefore, it does not need to enter the retransmission queue to save communication bandwidth.

[0023] This embodiment employs a strict time-division multiplexing mechanism. The master device is internally configured with a time-slot timer, which generates an interrupt at a fixed time interval (e.g., 100ms) to trigger the master device's communication task. At the beginning of each time slot, the master device first checks its transmission queue for unacceptable discarded data. If such data exists, the master device prioritizes transmitting this unacceptable discarded data, and the data frame carries the latest message sequence number generated by the master device. If the transmission queue is empty, the master device sends a preset polling message, which carries the latest message sequence number successfully received from the slave device. The polling message contains acceptable discarded data and does not require retransmission; its main function is to trigger the slave device's response and carry the peer's reception acknowledgment sequence number. When the slave device has not received any messages (including polling messages or data frames) from the master device, its communication interface is in a silent receiving state, and it is prohibited from actively sending any signals to the bus. Only when the slave device correctly parses the frame header sent by the master device and confirms that the frame is intended for itself is the slave device allowed to transmit data within a preset response time window. This mechanism ensures that only one device is transmitting signals on the bus at any given time, solving the common transmit / receive blocking problem in RS-485 half-duplex communication. Especially in scenarios where wireless repeaters (such as Bluetooth or Wi-Fi modules) are introduced, the latency and uncertainty of the wireless link itself are relatively large. If contention mechanisms such as CSMA / CA are used, the packet loss rate can easily soar due to hidden nodes or latency conflicts. However, the master-slave polling mechanism of this invention completely avoids such risks.

[0024] This embodiment defines an acknowledgment and retransmission protocol. For unacceptable discarded data, the sending end assigns a unique, monotonically increasing (decreasing) message sequence number when generating the data. The sequence number can be an 8-bit unsigned integer cyclically counted, ranging from 0 to 255. When the sending end sends unacceptable discarded data, the data is retained in the transmission queue until acknowledgment is received from the peer. The acknowledgment mechanism is implemented through a "piggybacking" method of polling messages and polling response messages: when the master device sends polling message 0x07, it fills the acknowledgment field of the frame structure with the sequence number of the latest successfully received slave device data; similarly, when the slave device sends polling response 0x07, it also fills the acknowledgment field of the response frame with the sequence number of the latest successfully received master device data. For example, if the master device sends a command with sequence number 1, and the slave device receives it correctly, it will set the acknowledgment number to 1 in the next polling response 0x07. Upon receiving polling response 0x07, the master device parses the acknowledgment number. If acknowledgment number 1 matches the latest sequence number sent in its transmission queue, it determines that the data has been successfully received and removes it from the transmission queue. If the master device does not receive acknowledgment number 1 in polling response 0x07, it determines that the data is lost, triggers the retransmission mechanism, and retransmits the command with sequence number 1. The same principle and process apply to discarding unacceptable data on the slave device.

[0025] In time slot T1, if the master device's sending queue is not empty, it sends the first message in that queue with sequence number K. If the sending queue is empty, it sends a polling message P1, carrying an acknowledgment sequence number N, indicating that the master device has acknowledged receiving the data with sequence number N from the slave device. After receiving the message with sequence number K, the slave device adds the message response to its sending queue. After receiving P1, the slave device checks its own sending queue. If the first message in its local sending queue has sequence number N, it removes that message from the queue. If the remaining queue is not empty, it sends a data frame D_M (with sequence number M). If the remaining queue is empty, it sends a polling response message carrying sequence number K, indicating that it has received the data with sequence number K from the master device. After receiving the polling response from the slave device, if the first message in its local sending queue has sequence number K, the master device removes that message from the queue and continues acknowledgment. This clearly demonstrates how bidirectional data transmission and state synchronization are achieved through a question-and-answer pattern.

[0026] During the initialization phase, the master device resets the sequence number counter to zero. When new unacceptable discardable data is generated, the master device enqueues it, assigns it a current sequence number, and then increments (or decrements) the sequence number counter. In each time slot, the queue is checked; if the queue is not empty, the data at the head of the queue is retrieved and sent. Simultaneously, an acknowledgment timer corresponding to that data is started. Upon receiving a response from the slave device, the master device extracts the acknowledgment field and compares it with the sequence number of the data in the sending queue. If a match is found, the data is cleared; otherwise, it is retransmitted.

[0027] The working process of this embodiment will be explained in detail below with reference to specific application scenarios.

[0028] Control command issuance for the high-pressure injector system: In the application scenario of medical high-pressure injectors, the host computer (master device) is located in the doctor's control room, and the slave computer (slave device) is located in the angiography room. The two are connected via RS-485 bus and wirelessly transmitted via Bluetooth relay. Suppose the host computer needs to send a "start injection" control command to the slave computer.

[0029] First, the doctor clicks the "Start" button on the host computer interface, and the host computer generates a control command data. This data is parsed as a "control command" by the function code parser, marked as unacceptable, and discarded. It is assigned sequence number 1, encapsulated into a data frame, and stored in the transmission queue.

[0030] Next, the timer slot triggers an interrupt, checks the transmit queue, and finds data with sequence number 1. A transmit frame is constructed, with the frame body containing 1 and command data payload, and sent to the RS-485 bus via the UART interface, and then transmitted to the lower-level machine via Bluetooth relay.

[0031] The lower-level device receives the frame via Bluetooth relay. First, it performs a CRC check; if the check passes, it parses out a 1. This 1 is recorded locally, indicating successful reception of the start command with sequence number 1 from the upper-level device. The lower-level device places the start response data at the tail of its send queue, then retrieves the first data item from the send queue and sends it to the upper-level device. Assuming the lower-level device's previous send sequence was empty, it directly sends the start response to the upper-level device; this start response sequence number is 1 (this is the lower-level device's send sequence number, unrelated to the received start command sequence number 1 from the upper-level device). When the lower-level device receives a polling message from the upper-level device, if its send queue is empty, it sends a polling response message carrying sequence number 1, which is the sequence number of the upper-level start command.

[0032] When the host computer receives the polling response from the slave computer, it parses out the latest message sequence number 1 received from the host computer by the slave computer, compares 1 with the 1 in the host computer's sending queue, confirms a match, and removes the data with sequence number 1 from the sending queue, completing one reliable communication cycle. If packet loss occurs during transmission and the polling response message does not carry sequence number 1, the data with sequence number 1 will be retransmitted in the next time slot until the polling response message carries sequence number 1.

[0033] The host computer receives the start response from the slave computer and parses out the sequence number 1. In the next polling message, it carries sequence number 1 from the start response and sends it to the slave computer. The slave computer parses the polling message, finds the sequence number 1, compares it with the 1 in the slave computer's sending queue, confirms a match, and removes the start response message with sequence number 1 from the sending queue, completing one reliable communication cycle. If packet loss occurs during transmission and the polling message does not carry sequence number 1, the slave computer retransmits the start response with sequence number 1 in the next time slot until it receives a polling message with sequence number 1.

[0034] Status data reporting of the high-pressure injector system: The host computer allocates a fixed time slot (e.g., the first 0.1s) in a polling cycle (e.g., 1s) for the status reporting process. In this time slot, the host computer sends a status reporting request. After receiving the status reporting request, the slave computer reports the current status (e.g., real-time speed, real-time dose, pressure, etc.). Neither the status reporting request nor the status report enters the sending queue of unacceptable discarded data of the host computer nor occupies the sequence number resource of the corresponding sending queue.

[0035] When the master device's polling message arrives at the slave device, the slave device checks its transmission queue and finds no unacceptable discarded data. The slave device then sends a polling response, carrying the stress data payload in the response frame. If this response frame is lost during wireless transmission, the master device will not receive it. Since this data did not enter the slave device's transmission queue, the slave device will not retransmit it. The master device will receive new stress data again in the next polling cycle. This mechanism ensures that even if status data is lost, it will not consume retransmission bandwidth, ensuring the smooth operation of the control command channel.

[0036] Communication assurance under extremely high packet loss rates: Assume extremely poor wireless link quality with a packet loss rate as high as 70%. If a traditional ACK retransmission mechanism is used, due to the mixing of status data and control data, a large number of retransmission requests for status data will quickly clog the channel, causing control commands to fail and the system to crash.

[0037] In this embodiment, status data (which accounts for the majority of communication traffic) is marked as disposable and does not consume retransmission resources. Channel bandwidth is primarily used for retransmission of control commands. Although the packet loss rate is as high as 70%, the data volume of control commands is small and they have the highest priority. Through repeated polling and retransmission, the master device statistically guarantees that the control commands will eventually reach the other end. Experiments show that even with a 70% packet loss rate, this invention can still ensure that control commands are successfully delivered within an acceptable delay (e.g., 3-5 polling cycles), and the system remains secure and controllable.

[0038] Additional notes on parameter settings: In this embodiment, the time slot length should be set based on the round-trip delay of the communication link and the data frame length. For RS-485 with Bluetooth repeater, the round-trip delay may be between 20 and 100 ms. Assuming a data frame length of 50 bytes and a baud rate of 9600 bps, sending one frame of data will take approximately 50 ms. It is recommended to set the time slot length to 200-500 ms to allow sufficient transmission and response time.

[0039] Regarding alternative embodiments: In the above embodiments, the master device is a host computer, and the slave device is a slave computer. In another alternative embodiment, the master and slave roles can be dynamically assigned according to the network topology. For example, in a multi-machine communication system, one device can be designated as a permanent master device, and the other devices as slave devices; alternatively, a token ring approach can be used, with the device holding the token acting as a temporary master device for polling. Furthermore, the wireless repeater module is not limited to Bluetooth; it can also be a wireless communication module such as Wi-Fi, ZigBee, or LoRa, or even a wired fiber optic repeater, as long as the communication link has a risk of packet loss and is half-duplex. The method of this embodiment is applicable to all such devices.

[0040] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A reliable communication method based on master-slave polling and data hierarchy, applied to a communication system including a master device and a slave device, wherein the master device and the slave device are connected via a half-duplex communication bus; characterized in that, The method includes the following steps: Step 1: Divide the communication data to be sent between the master device and the slave device into unacceptable discardable data and acceptable discardable data. The unacceptable discardable data located in the master device's sending queue carries a first message sequence number, and the unacceptable discardable data located in the slave device's sending queue carries a second message sequence number. Step 2: The master device sends a polling message to the slave device according to a predetermined time slot. The polling message carries a second message sequence number of the unacceptable discarded data that the master device has successfully received from the slave device. Step 3: After receiving the polling message, the slave device sends a polling response to the master device. The polling response carries the first message sequence number of the unacceptable discarded data that the slave device has successfully received from the master device. Step 4: The master device determines whether the slave device has successfully received the latest unacceptable discarded data sent by the master device based on the first message sequence number carried in the received polling response. If the determination is yes, the corresponding unacceptable discarded data is removed from the master device's sending queue. If the determination is no, the corresponding unacceptable discarded data is resent. Step 5: The slave device determines whether the master device has successfully received the latest unacceptable discarded data sent by the slave device based on the second message sequence number carried in the received polling message. If the determination is yes, the corresponding unacceptable discarded data is removed from the slave device's sending queue. If the determination is no, the corresponding unacceptable discarded data is resent.

2. The reliable communication method based on master-slave polling and data hierarchy according to claim 1, characterized in that, Communication data is divided into unacceptable discardable data and acceptable discardable data, including: identifying the service type of the communication data; marking important data such as configuration data, configuration response, control command or operation response as unacceptable discardable data; marking data with service types such as polling, status reporting request, and status reporting as acceptable discardable data; for the acceptable discardable data, the master device sends it directly without storing it in the master device's sending queue and does not perform retransmission processing.

3. The reliable communication method based on master-slave polling and data hierarchy according to claim 1, characterized in that, Before the step of the master device sending polling messages to the slave device according to a predetermined time slot, the method further includes: the master device dividing the communication time slot to prevent half-duplex communication transmission and reception conflicts; if there is unacceptable discarded data in the master device's sending queue in the current time slot, the master device sends the unacceptable discarded data; if the master device's sending queue is empty, the master device sends the polling message or status reporting request.

4. The reliable communication method based on master-slave polling and data hierarchy according to claim 1, characterized in that, The polling message or status reporting request belongs to acceptable discardable data; after receiving the polling message, the slave device checks whether there is unacceptable discardable data in the slave device's sending queue; If it exists, the unacceptable discarded data is sent to the master device as a response to the polling message; if it does not exist, a preset polling response is sent to the master device, and the first message sequence number of the master device that has been successfully received is carried in the polling response.

5. A reliable communication method based on master-slave polling and data hierarchy according to claim 1, characterized in that, The master device status reporting request is allocated a fixed time slot within the polling cycle. Upon receiving the status reporting request, the slave device directly responds with a status report. Furthermore, the status reporting request is not entered into the master device's sending queue and is not assigned the first message sequence number; the status report is not entered into the slave device's sending queue and is not assigned the second message sequence number.

6. A reliable communication method based on master-slave polling and data hierarchy according to claim 1, characterized in that, The first and second message sequence numbers carried by the unacceptable discarded data are numbered according to an auto-incrementing, auto-decrementing, or agreed-upon pattern.

Images