High-reliability low-flow data backhaul protocol method

Abstract

The invention discloses a high-reliability low-flow data backhaul protocol method, and relates to the technical field of communication protocol and transmission strategy optimization. Comprising the following steps: S1, data coding and segmentation; s2, packaging and sending a micro package; s3, receiving and confirming data; s4, a retransmission mechanism; and S5, performing differential compression and uploading. According to the method, the redundancy rate is dynamically calculated, and the characteristics of fountain codes are combined, so that accurate packet loss compensation is realized, and the retransmission probability is remarkably reduced; based on a timeout retransmission mechanism, the anti-interference capability is further enhanced, and the data integrity in an extreme environment is ensured; through time-frequency and two-dimensional random sending strategies, co-frequency collision and channel congestion are effectively avoided, and the transmission success rate in a low signal-to-noise ratio scene is improved.

Description

Technical Field

[0001] This invention relates to the field of communication protocol and transmission strategy optimization technology, and in particular to a highly reliable low-traffic data backhaul protocol method. Background Technology

[0002] In today's era of rapid digital and information technology development, various types of sensor networks are widely used in numerous fields, such as environmental monitoring, geological exploration, intelligent agricultural management, and industrial automation control. These sensor nodes are responsible for collecting critical data from the field and transmitting this data to remote processing centers or data centers for analysis and decision-making. However, due to the specificity and complexity of application scenarios, traditional data transmission protocols and methods often face many challenges and struggle to meet actual needs, especially in ensuring high reliability of data return and adapting to low-traffic transmission scenarios.

[0003] In many practical applications, sensor nodes are typically deployed in complex geographical environments, such as remote mountainous areas, forests, deserts, or underground mines. Wireless communication conditions in these areas are extremely harsh, and signals are easily affected by terrain, obstacles, and electromagnetic interference. For example, in some valleys, the signal strength between the ground base station and the sensor node is significantly weakened due to mountain obstruction; in industrial sites with complex electromagnetic environments, the operation of other equipment generates a large amount of electromagnetic noise, interfering with normal data transmission. When a sensor node detects that the signal strength indication (RSSI) of the ground base station is below a certain threshold (e.g., -100dBm) or the round-trip time (RTT) is too long (exceeding 2 seconds), it means that the quality of the current communication link is already very poor. Traditional transmission methods can easily lead to data loss or errors, thus affecting the integrity and accuracy of the data.

[0004] Many existing data encoding schemes employ fixed encoding strategies, failing to dynamically adjust redundancy based on network conditions. In actual transmission, network packet loss rates are constantly changing. Fixed encoding methods may not provide sufficient redundancy to cope with packet loss in some situations, leading to decoding failures; while in other situations, they may be excessively redundant, wasting valuable bandwidth resources. For example, encoding with high redundancy during periods of good network conditions increases unnecessary transmission burden; conversely, during periods of network congestion and severe packet loss, insufficient redundancy prevents the recovery of the original data.

[0005] Traditional data transmission often uses a single frequency or fixed time slot, making it susceptible to interference from other nodes on the same frequency. In a network environment with multiple nodes, collisions can occur when multiple nodes transmit data simultaneously on the same frequency, leading to data loss. Furthermore, this fixed transmission pattern is easily predicted and interfered with by malicious attackers, further reducing the security and reliability of data transmission.

[0006] While some existing protocols possess certain error detection and correction capabilities, they lack robust retransmission mechanisms when faced with persistent transmission failures. If data packets are lost and retransmission is not performed promptly or if the retransmission strategy is inadequate, data may fail to reach the receiving end, impacting the overall system operation. For example, some simple stop-and-wait protocols simply retransmit the previously sent data packet after a timeout, without considering changes in network conditions or accumulated errors, resulting in inefficient retransmission.

[0007] Specifically, the existing MQTT / CoAP protocol requires maintaining long connections, with satellite links taking more than 800ms round trip, and frequent reconnections resulting in more than 30% of traffic wastage; the traditional packet retransmission mechanism is drastically less efficient in scenarios with 20% packet loss; satellites are billed based on traffic, with a single 100B message costing 0.01 yuan, and the daily cost for 10,000 sensors reaching tens of thousands of yuan, which is unsustainable. Summary of the Invention

[0008] The purpose of this invention is to address the shortcomings of existing technologies by proposing a highly reliable low-traffic data backhaul protocol method.

[0009] To achieve the above objectives, the present invention adopts the following technical solution: A highly reliable low-traffic data backhaul protocol method includes the following steps: S1: Data encoding and segmentation; S11: Data detection and segmentation: When the sensor node detects that the RSSI of the ground base station is lower than -100dBm or the RTT exceeds the set time, it automatically segments the original data packet M into K source symbols; S12: Fountain code generation: N≥K encoded symbols are generated using LubyTransform, and the redundancy rate ρ is dynamically calculated based on the packet loss rate p. S2: Micropacket encapsulation and transmission; S3: Data reception and confirmation; S31: Data Decoding: When the mobile aggregation node receives any K' symbols, K'≥K+δ,δ=3, it performs decoding. S32: Return ACK: After successful decoding, 4BACK_ID is returned via BeiDou short message or BGAN; the sensor clears its local cache immediately after receiving the ACK; S4: Retransmission mechanism; S41: Timeout Retransmission: If an ACK is not received within the set time, the sensor automatically increases the redundancy rate and retransmits; S5: Differential compression upload.

[0010] Preferably, in step S2, the micro-packet encapsulation and transmission are as follows: S21: Encapsulation encoding: Each encoded symbol is appended with a 2B symbol ID and a 2B CRC16 checksum to form a UDP micropacket with a total size not exceeding 60B; S22: Two-dimensional random transmission: adopts time-frequency, two-dimensional random transmission mode; time slot random backoff 0–511ms; frequency hopping 20 times in the range of 400–410MHz to reduce the probability of co-frequency collision.

[0011] Preferably, in step S5, the differential compression upload is performed as follows: S51: Differential Compression: The aggregation node uses differential compression for uploading, and only transmits back ΔS=S_t–S_{t-1}; S52: Further compression: The data is compressed using the DEFLATE algorithm to a compression rate of no less than 60%, and the compressed data is then uploaded to the destination via a satellite link.

[0012] Preferably, in S11, data detection and segmentation specifically include: S111: Signal strength detection. The sensor node continuously monitors the signal strength of the ground base station and obtains the current signal strength value through RSSI. The RSSI value is compared with a preset threshold. If it is less than the preset threshold, proceed to the next step; otherwise, maintain the existing connection status. S112: Round-trip time detection. The sensor node sends a probe data packet to the ground base station and records the sending timestamp Tx. After receiving the probe data packet, the ground base station immediately returns an acknowledgment message and records the receiving timestamp Rx. The RTT is calculated as Rx-Tx. If the RTT is greater than the set time, proceed to the next step; otherwise, maintain the existing connection status. S113: Data segmentation, determine the size and content of the data packet M to be transmitted; according to the preset K value, segment the data packet M into K source symbols S_1, S_2, ..., S_K; each source symbol S_i contains a part of the original data, ensuring that the sum of all source symbols is equal to the original data packet M.

[0013] Preferably, in step S12, the fountain code generation specifically involves: S121: Select encoding parameters, dynamically calculate the redundancy rate ρ=1 / (1-p)-1+0.1 based on the current packet loss rate p; set the total number of encoded symbols N such that N≥K; S122: Generate encoded symbols, initialize an empty list of encoded symbols; for each source symbol S_i, use the LubyTransform algorithm to generate several encoded symbols; ensure that the total number of generated encoded symbols reaches N; S123: Add a checksum, appending a 2B symbol ID and a 2B CRC16 checksum to each encoded symbol; forming the final encoded symbol, with each encoded symbol not exceeding 60B in size.

[0014] Preferably, in step S21, the encapsulation encoding is specifically as follows: S211: Create a UDP micropacket, create a UDP datagram structure; add each encoded symbol to the UDP datagram; ensure the total size of the UDP datagram does not exceed 60 bytes; S212: Add header information; add metadata to the UDP datagram header; ensure that all UDP micro-packets are formatted correctly and consistently.

[0015] Preferably, in step S22, the two-dimensional random transmission specifically involves: S221: Random time backoff, generating a random time slot for each UDP micro-packet, ranging from 0 to 511 ms; UDP micro-packets are sent sequentially according to the generated time slots. S222: Frequency hopping transmission, set the initial frequency to 400MHz; for each UDP micro-packet sent, the frequency increases by 1MHz until it reaches 410MHz; repeat the above process, for a total of 20 frequency hopping cycles.

[0016] Preferably, in step S31, data decoding specifically involves: S311: Collect coded symbols; the mobile aggregation node receives UDP micro-packets from the sensor nodes; and stores the received coded symbols in a buffer. S312: Check if the number of encoded symbols in the buffer has reached K'; use the LT code decoding algorithm to decode the collected encoded symbols; restore the original data packet M.

[0017] Preferably, in step S32, returning ACK specifically means: S321: Generate an ACK message. After the aggregation node successfully decodes the message, it generates a 4-byte ACK_ID. The ACK_ID is generated based on the first 4 bytes of the symbol ID's CRC32, with a collision probability of 1 / 2^32. S322: Send an ACK message, and send the ACK_ID back to the sensor node via BeiDou short message or BGAN; ensure that the ACK message is accurately delivered to the sensor node.

[0018] Preferably, in S41, the timeout retransmission specifically refers to: S411: Set a timer. After sending data, the sensor node starts a 30-minute timer; during the timer period, it continuously listens for ACK messages from the aggregation node. S412: Timeout handling. If the ACK message is not received before the timer expires, the retransmission process is initiated; the redundancy rate ρ is automatically increased to ρ+0.2; the data segmentation, fountain code generation, and encapsulation encoding symbol steps are re-executed, with a maximum of 3 retry rounds.

[0019] The beneficial effects of this invention are as follows: 1. This invention achieves accurate packet loss compensation by dynamically calculating redundancy rate and combining the characteristics of fountain codes, significantly reducing the probability of retransmission; it further enhances anti-interference capability based on timeout retransmission mechanism to ensure data integrity in extreme environments; and it effectively avoids co-frequency collisions and channel congestion through time-frequency, two-dimensional random transmission strategy, improving the transmission success rate in low signal-to-noise ratio scenarios.

[0020] 2. The micro-packet encapsulation of this invention strictly limits the total size to ≤60B, which is suitable for narrowband communication requirements; differential compression combined with the DEFLATE algorithm greatly reduces the amount of data in the satellite link and lowers the transmission cost; the data segmentation mechanism based on RSSI / RTT avoids blind encoding and only activates redundancy protection when necessary, balancing reliability and overhead.

[0021] 3. The mobile aggregation node of this invention only needs to receive K'≥K+δ (δ=3) symbols to decode, without waiting for all data packets to arrive, thus shortening the end-to-end delay; after ACK confirmation, the local cache is cleared immediately to release storage resources; the BeiDou short message / BGAN hybrid confirmation mechanism takes into account both timeliness and coverage, and is suitable for deployment in remote areas or complex terrain.

[0022] 4. This invention provides basic error detection capability based on CRC16 checksum, while the natural error correction characteristics of fountain code can tolerate partial symbol loss; based on the dual guarantee mechanism of coding redundancy and physical layer collision avoidance, the system maintains stable performance in strong attenuation channels. Attached Figure Description

[0023] Figure 1 This is a flowchart of a highly reliable low-traffic data backhaul protocol method proposed in this invention. Detailed Implementation

[0024] The technical solution of the present invention will be further described in detail below with reference to specific embodiments. Example

[0025] A highly reliable low-traffic data backhaul protocol method includes the following steps: S1: Data encoding and segmentation; S11: Data detection and segmentation: When the sensor node detects that the signal strength indication (RSSI) of the ground base station is lower than -100dBm or the round-trip time (RTT) exceeds 2 seconds, it automatically segments the original data packet M into K source symbols; S12: Fountain code generation: N≥K encoded symbols are generated using LubyTransform (LT code). The redundancy rate ρ is dynamically calculated based on the packet loss rate p, and the formula is ρ=1 / (1-p)-1+0.1. S2: Micropacket encapsulation and transmission; S21: Encapsulation encoding: Each encoded symbol is appended with a 2B symbol ID and a 2B CRC16 checksum to form a UDP micropacket with a total size not exceeding 60B; S22: Two-dimensional random transmission: adopts a "time-frequency" two-dimensional random transmission method; time slot random backoff 0–511ms; frequency hopping 20 times in the range of 400–410MHz to reduce the probability of co-frequency collision; S3: Data reception and confirmation; S31: Data Decoding: The mobile aggregation node (with satellite modem) can decode any K' (≥K+δ,δ=3) symbols it receives; S32: Return ACK: After successful decoding, 4BACK_ID is returned via BeiDou short message or BGAN; the sensor clears its local cache immediately after receiving the ACK; S4: Retransmission mechanism; S41: Timeout retransmission: If no ACK is received within 30 minutes, the sensor automatically increases the redundancy rate ρ→ρ+0.2 and retransmits, with a maximum of 3 rounds of retransmission; S5: Differential compression upload; S51: Differential Compression: The aggregation node uses differential compression for uploading, and only transmits back ΔS=S_t–S_{t-1}; S52: Further compression: The data is compressed using the DEFLATE algorithm to a compression rate of no less than 60%, and the compressed data is then uploaded to the destination via a satellite link.

[0026] Specifically, in S11, data detection and segmentation are as follows: S111: Signal strength detection. The sensor node continuously monitors the signal strength of the ground base station and obtains the current signal strength value through RSSI (Received Signal Strength Indicator). The RSSI value is compared with a preset threshold (-100dBm). If RSSI < -100dBm, proceed to the next step; otherwise, maintain the existing connection status. S112: Round-trip time detection. The sensor node sends a probe data packet to the ground base station and records the sending timestamp Tx. After receiving the probe data packet, the ground base station immediately returns an acknowledgment message and records the receiving timestamp Rx. The RTT (round-trip time) is calculated as Rx-Tx. If RTT > 2 seconds, proceed to the next step; otherwise, maintain the existing connection state. S113: Data segmentation, determine the size and content of the data packet M to be transmitted; according to the preset K value (e.g., K=10), segment the data packet M into K source symbols S_1, S_2, ..., S_K; each source symbol S_i contains a part of the original data, ensuring that the sum of all source symbols is equal to the original data packet M.

[0027] Specifically, in S12, the fountain code generation is as follows: S121: Select encoding parameters, dynamically calculate the redundancy rate ρ=1 / (1-p)-1+0.1 based on the current packet loss rate p; set the total number of encoded symbols N such that N≥K; S122: Generate encoded symbols, initialize an empty list of encoded symbols; for each source symbol S_i, use the LubyTransform (LT code) algorithm to generate several encoded symbols; ensure that the total number of generated encoded symbols reaches N; S123: Add a checksum, appending a 2B symbol ID and a 2B CRC16 checksum to each encoded symbol; forming the final encoded symbol, with each encoded symbol not exceeding 60B in size.

[0028] Specifically, in S21, the encapsulation encoding is as follows: S211: Create a UDP micropacket, create a UDP datagram structure; add each encoded symbol to the UDP datagram; ensure the total size of the UDP datagram does not exceed 60 bytes; S212: Add header information. Add necessary metadata to the UDP datagram header, such as source address, destination address, etc.; ensure that all UDP micro-packets are formatted correctly and consistently.

[0029] Specifically, in S22, the two-dimensional random transmission is as follows: S221: Random time backoff, generating a random time slot for each UDP micro-packet, ranging from 0 to 511 ms; UDP micro-packets are sent sequentially according to the generated time slots. S222: Frequency hopping transmission, set the initial frequency to 400MHz; for each UDP micro-packet sent, the frequency increases by 1MHz until it reaches 410MHz; repeat the above process, for a total of 20 frequency hopping cycles.

[0030] Specifically, in S31, data decoding involves: S311: Collect coded symbols; the mobile aggregation node receives UDP micro-packets from the sensor nodes; and stores the received coded symbols in a buffer. S312: Check if the number of encoded symbols in the buffer reaches K' (≥K+δ,δ=3); use the LT code decoding algorithm to decode the collected encoded symbols; restore the original data packet M.

[0031] Specifically, in S32, returning ACK means: S321: Generate an ACK message. After the aggregation node successfully decodes the message, it generates a 4-byte ACK_ID. The ACK_ID is generated based on the first 4 bytes of the symbol ID's CRC32, with a collision probability of 1 / 2^32. S322: Send an ACK message, and send the ACK_ID back to the sensor node via BeiDou short message or BGAN; ensure that the ACK message is accurately delivered to the sensor node.

[0032] Specifically, in S41, the timeout retransmission is as follows: S411: Set a timer. After sending data, the sensor node starts a 30-minute timer; during the timer period, it continuously listens for ACK messages from the aggregation node. S412: Timeout handling. If the ACK message is not received before the timer expires, the retransmission process is initiated; the redundancy rate ρ→ρ+0.2 is automatically increased; the data segmentation, fountain code generation, and encapsulation of encoded symbols are re-executed, with a maximum of 3 retry rounds.

[0033] 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 highly reliable low-traffic data backhaul protocol method, characterized in that, Includes the following steps: S1: Data encoding and segmentation; S11: Data detection and segmentation: When the sensor node detects that the RSSI of the ground base station is lower than the set value or the RTT exceeds the set time, it automatically segments the original data packet M into K source symbols; S12: Fountain code generation: N≥K encoded symbols are generated using LubyTransform, and the redundancy rate ρ is dynamically calculated based on the packet loss rate p. S2: Micropacket encapsulation and transmission; S3: Data reception and confirmation; S31: Data Decoding: When the mobile aggregation node receives any K' symbols that meet the set standard, it will perform decoding; S32: Return ACK: After successful decoding, 4BACK_ID is returned via BeiDou short message or BGAN; the sensor clears its local cache immediately after receiving the ACK; S4: Retransmission mechanism; S41: Timeout Retransmission: If no ACK is received within the set time, the sensor automatically increases the redundancy rate and retransmits; S5: Differential compression upload.

2. The method for a highly reliable low-traffic data backhaul protocol according to claim 1, characterized in that, In step S2, the micro-packet encapsulation and transmission are specifically as follows: S21: Encapsulation Encoding: Each encoded symbol is appended with a symbol ID and a checksum to form a UDP micropacket whose total size does not exceed a set value; S22: Two-dimensional random transmission: Using time-frequency, two-dimensional random transmission method.

3. The highly reliable low-traffic data backhaul protocol method according to claim 2, characterized in that, In step S5, the differential compression upload is performed as follows: S51: Differential Compression: The aggregation node uses differential compression for uploading, and only transmits back ΔS=S_t–S_{t-1}; S52: Further Compression: The DEFLATE algorithm is used to compress the data, and the compressed data is then uploaded to the destination via a satellite link.

4. The method for a highly reliable low-traffic data backhaul protocol according to claim 1, characterized in that, In step S11, data detection and segmentation specifically involve: S111: Signal strength detection. The sensor node continuously monitors the signal strength of the ground base station and obtains the current signal strength value through RSSI; the RSSI value is compared with a preset threshold. If the value is less than the preset threshold, proceed to the next step; otherwise, maintain the existing connection status. S112: Round-trip time detection. The sensor node sends a probe data packet to the ground base station and records the sending timestamp Tx. After receiving the probe data packet, the ground base station immediately returns an acknowledgment message and records the receiving timestamp Rx. The RTT is calculated as Rx-Tx. If the RTT is greater than the set time, proceed to the next step; otherwise, maintain the existing connection status. S113: Data segmentation, determine the size and content of the data packet M to be transmitted; according to the preset K value, segment the data packet M into K source symbols S_1, S_2, ..., S_K; each source symbol S_i contains a part of the original data, ensuring that the sum of all source symbols is equal to the original data packet M.

5. The highly reliable low-traffic data backhaul protocol method according to claim 4, characterized in that, In step S12, the fountain code generation is specifically as follows: S121: Select encoding parameters, dynamically calculate the redundancy rate ρ=1 / (1-p)-1+0.1 based on the current packet loss rate p; set the total number of encoded symbols N such that N≥K; S122: Generate encoded symbols, initialize an empty list of encoded symbols; for each source symbol S_i, use the LubyTransform algorithm to generate several encoded symbols; ensure that the total number of generated encoded symbols reaches N; S123: Add a check code, attaching a symbol ID and a check code to each encoded symbol; forming the final encoded symbol.

6. The method for a highly reliable low-traffic data backhaul protocol according to claim 2, characterized in that, In step S21, the encapsulation encoding is specifically as follows: S211: Create a UDP micropacket, create a UDP datagram structure; add each encoded symbol to the UDP datagram; S212: Add header information, add metadata to the header of UDP datagrams; Ensure that all UDP micro-packets are formatted correctly and consistently.

7. The highly reliable low-traffic data backhaul protocol method according to claim 6, characterized in that, In step S22, the two-dimensional random transmission specifically refers to: S221: Random time backoff, generating a random time slot for each UDP micro-packet; UDP micro-packets are sent sequentially according to the generated time slot order; S222: Frequency hopping transmission, set the initial frequency; for each UDP micro-packet sent, the frequency increases by a fixed value until the preset standard is reached; repeat the above process until the total number of frequency hopping reaches the preset number.

8. The method for a highly reliable low-traffic data backhaul protocol according to claim 1, characterized in that, In step S31, data decoding specifically involves: S311: Collect coded symbols; the mobile aggregation node receives UDP micro-packets from the sensor nodes; and stores the received coded symbols in a buffer. S312: Check if the number of encoded symbols in the buffer has reached K'; use the LT code decoding algorithm to decode the collected encoded symbols; Restore the original data packet M.

9. A highly reliable low-traffic data backhaul protocol method according to claim 8, characterized in that, In S32, the return of ACK specifically means: S321: Generate an ACK message. After the aggregation node successfully decodes the message, it generates an ACK_ID. The ACK_ID is generated based on the CRC32 front-end of the symbol ID. S322: Send an ACK message, and send the ACK_ID back to the sensor node via BeiDou short message or BGAN; Ensure that ACK messages are delivered to the sensor nodes accurately.

10. A highly reliable low-traffic data backhaul protocol method according to claim 1, characterized in that, In S41, the timeout retransmission specifically refers to: S411: Set a timer. The sensor node starts the timer after sending data; during the timer period, it continuously listens for ACK messages from the aggregation node. S412: Timeout handling. If the ACK message is not received before the timer expires, the retransmission process is initiated; the redundancy rate ρ is automatically increased to ρ+0.2; the data segmentation, fountain code generation, and encapsulation encoding symbol steps are re-executed, with a maximum of 3 retry rounds.

Images