Satellite communication data transmission method for deep sea net cage

Abstract

The invention relates to the technical field of satellite communication, and discloses a satellite communication data transmission method for deep and far sea net cages. The method comprises the following steps: collecting net cage environment parameters, biological behavior data and equipment operation state data through a multi-modal sensor cluster deployed on a deep sea net cage, and generating an original multi-modal monitoring data set; performing adaptive data cleaning and format standardization processing on the original multi-modal monitoring data set to generate a standardized multi-modal data stream; by establishing the satellite channel state evaluation model, predicting the channel quality change in real time and dynamically selecting the optimal satellite communication protocol and channel coding scheme based on the prediction result, the problem of channel quality degradation can be sensed and avoided in real time, the continuity and reliability of data transmission are ensured, and the risk of data transmission interruption is reduced.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of satellite communication, and in particular to a satellite communication data transmission method for deep-sea net cages. BACKGROUND

[0002] Satellite communication is communication between radio communication stations on the earth using a satellite as a relay. A satellite communication system is composed of a satellite and an earth station. The characteristics of satellite communication are as follows: a large communication range, communication between any two points can be carried out as long as the points are within the range covered by the satellite's radio waves; not susceptible to land disasters; can be received at multiple locations, enabling economical realization of broadcasting and multi-address communication; and flexible circuit setting, which can disperse excessively concentrated traffic at any time.

[0003] At present, due to the complex environment of deep-sea net cages and unstable satellite communication transmission conditions, when monitoring data is remotely transmitted, the transmission channel state of the equipped satellite communication system changes dramatically, and the channel quality deterioration problem cannot be sensed and avoided in real time. When signal strength drops sharply and bit error rate rises sharply, data transmission is interrupted and integrity is damaged, and the reliability of monitoring data upload cannot be guaranteed.

[0004] Therefore, the satellite communication data transmission method for deep-sea net cages is proposed to solve the above problems. SUMMARY

[0005] In view of the deficiencies of the prior art, the satellite communication data transmission method for deep-sea net cages is provided to solve the problem of data transmission interruption and integrity damage in the background art, which cannot guarantee the reliability of monitoring data upload.

[0006] To achieve the above purpose, the application provides the following technical scheme: a satellite communication data transmission method for deep-sea net cages, the method comprising the following steps: S1, collecting net cage environment parameters, biological behavior data and equipment running state data through a multi-modal sensor cluster deployed on a deep-sea net cage to generate an original multi-modal monitoring data set; S2, performing adaptive data cleaning and format standardization processing on the original multi-modal monitoring data set to generate a standardized multi-modal data stream; S3, performing dynamic evaluation of transmission priority based on the standardized multi-modal data stream to generate a data transmission priority matrix; S4, selecting an optimal satellite communication protocol and channel coding scheme according to the data transmission priority matrix and real-time satellite channel state information; S5, performing distributed compression and encryption processing on the standardized multi-modal data stream based on the selected satellite communication protocol to generate an encrypted data packet set; S6, transmitting the encrypted data packet set by multi-satellite link aggregation technology, and implementing dynamic link switching and error correction control during transmission; S7, verifying the integrity of the received data packet and recombining it in the shore-based data center to generate a reconstructed multi-modal data set; S8, performing data validity analysis and anomaly detection based on the reconstructed multi-modal data set to generate a data quality evaluation report; S9, adjusting the data acquisition and transmission strategy of the deep-sea net cage according to the data quality evaluation report; S10, constructing a satellite communication transmission efficiency evaluation model to realize continuous optimization of the data transmission process.

[0007] Preferably, the collection of net cage environmental parameters, biological behavior data and equipment operating state data in S1 includes the following operations: S11, collecting water temperature, salinity, dissolved oxygen concentration and water flow velocity data through the underwater sensor array of the net cage to generate a hydrological environmental data set; S12, collecting fish activity frequency, feeding behavior and group distribution data through biological acoustic monitoring equipment to generate a biological behavior data set; S13, collecting netting tension, anchor chain stress and float displacement data through net cage structure monitoring sensors to generate a structure safety data set; S14, collecting solar panel output power, battery energy storage state and equipment energy consumption data through the energy management system to generate an energy state data set; S15, aligning the hydrological environmental data set, biological behavior data set, structure safety data set and energy state data set by time sequence to generate an original multi-modal monitoring data set.

[0008] Preferably, the adaptive data cleaning and format standardization processing in S2 includes the following operations: S21, detecting and removing outliers from the original multi-modal monitoring data set, using a sliding window-based Z-score algorithm to identify and replace data points outside the threshold range; Wherein, the Z-score standardization formula is: ; Where Z ij is the standard score of the jth sensor ith data point, X ij is the original data value, μ j is the mean of the jth sensor data in the sliding window, σ j is the standard deviation of the jth sensor data in the sliding window; When , it is determined as an outlier and replaced. S22, timestamp alignment and resampling processing of sensor data with different sampling frequencies, so that all data streams reach a unified time reference; S23, converting various data into standard JSON format and adding metadata descriptor to identify data source, collection time and sensor type; S24, semantic consistency verification of standardized data stream to ensure correct logical association of data; S25, output standardized multi-modal data stream with integrity mark.

[0009] Preferably, the transmission priority dynamic evaluation in S3 includes the following operations: S31, establish a data priority evaluation index system, including data real-time requirement, security level weight, business criticality index and data scarcity coefficient; S32, calculate the weight vector of each index based on the analytic hierarchy process, and generate a priority evaluation matrix: ; Wherein is the priority judgment matrix, is the importance degree of the pth priority index relative to the qth index, and n is the number of evaluation indexes; S33, dynamically adjust the evaluation matrix parameters according to the net cage operation scene, and increase the weight of safety class data in emergency state; S34, use fuzzy comprehensive evaluation method to score the priority of each data stream, and generate a data transmission priority matrix; S35, divide the data stream into three levels of emergency transmission, regular transmission and delay transmission according to the priority matrix.

[0010] Preferably, the selection of the optimal satellite communication protocol and channel coding scheme in S4 includes the following operations: S41, real-time monitoring of signal strength, bit error rate, bandwidth utilization and transmission delay parameters of available satellite channels; S42, establish a satellite channel state evaluation model to predict the communication quality of each channel in the future time window; S43, select the lowest power consumption communication protocol that meets the quality of service requirements according to the data transmission priority matrix; S44, automatically select an enhanced forward error correction coding scheme for high-priority data, and generate an optimized configuration scheme of satellite communication protocol and coding scheme accordingly.

[0011] Preferably, the distributed compression and encryption processing in S5 includes the following operations: S51, Select compression algorithm according to data characteristics: lossy compression for environmental data and lossless compression for equipment state data; S52, Use distributed compression technology to perform compression operations in parallel on the edge computing nodes of the net cage; S53, Generate dynamic encryption keys based on national encryption algorithms and perform layered encryption on the compressed data packets; S54, Add digital signatures and integrity check codes to each data packet; S55, Group and package the encrypted data packets according to transmission priority to generate an encrypted data packet set.

[0012] Preferably, the multi-satellite link aggregation and fragmented transmission in S6 includes the following operations: S61, Establish a multi-satellite link aggregation pool, and simultaneously connect more than three satellite communication systems in different orbits; S62, Perform adaptive fragmentation according to the priority characteristics of the encrypted data packet set, and use multiple copies for high-priority data; S63, Dynamically select the optimal transmission path and perform seamless switching through real-time link quality monitoring; S64, Implement a fountain code-based error correction control mechanism to adapt to the characteristics of high-delay and high-error-rate satellite channels; S65, Continuously optimize the fragmentation strategy and link allocation scheme during transmission.

[0013] Preferably, the integrity verification and recombination in S7 includes the following operations: S71, Perform digital signature verification and decryption processing on the received data packets; S72, Check the integrity of the data packets and request retransmission of lost and damaged data fragments; S73, Reconstruct the original data stream according to the fragment number and timestamp information; S74, Verify the timing consistency and logical integrity of the reconstructed data; S75, Output the reconstructed multi-modal data set with quality labels.

[0014] Preferably, the data validity analysis and anomaly detection in S8 include the following operations: S81, Perform outlier detection on the reconstructed multi-modal data set to identify data deviations caused by sensor failures and transmission anomalies; S82, Confirm data authenticity through multi-source data cross-validation technology; S83, Calculate data completeness, timeliness, and accuracy indicators; S84, Generate a data quality assessment report containing data quality scores and anomaly alerts; S85. Feedback the assessment report to the cage site control system and shore-based monitoring platform.

[0015] Preferably, the feedback adjustment of the data acquisition and transmission strategy in S9 includes the following operations: S91. Adjust the sensor sampling frequency and transmission trigger threshold according to the data quality assessment report; S92. Optimize satellite communication link selection strategy and transmission timing arrangement. Beneficial effects

[0016] Compared with existing technologies, this invention provides a satellite communication data transmission method for deep-sea cages, which has the following advantages: 1. In this invention, by establishing a satellite channel state assessment model, channel quality changes are predicted in real time, and the optimal satellite communication protocol and channel coding scheme are dynamically selected based on the prediction results. This enables real-time perception and avoidance of channel quality degradation problems, ensuring the continuity and reliability of data transmission and reducing the risk of data transmission interruption.

[0017] 2. In this invention, by constructing a data transmission priority matrix, dynamic evaluation is performed based on data real-time requirements, security level weights, business criticality index, and data scarcity coefficient. This enables the system to perform differentiated scheduling according to the value and urgency of data services, ensuring priority transmission of critical security data and preventing routine data from crowding out channel resources.

[0018] 3. In this invention, by constructing a transmission performance evaluation model and employing a reinforcement learning algorithm, the transmission parameter configuration is continuously optimized, and the transmission strategy is adaptively adjusted. This enables the system to learn and optimize autonomously according to environmental changes, thereby improving the long-term efficiency and stable reliability of satellite communication data transmission. Attached Figure Description

[0019] Figure 1 This is a flowchart of the satellite communication data transmission method for deep-sea cages according to the present invention. Detailed Implementation

[0020] 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.

[0021] For specific implementation examples, please refer to: Figure 1 A satellite communication data transmission method for deep-sea cages, comprising the following steps: S1. Collect environmental parameters, biological behavior data, and equipment operation status data of the cages by deploying a multimodal sensor cluster on the deep-sea cages, and generate the original multimodal monitoring dataset; S2. Perform adaptive data cleaning and format standardization on the original multimodal monitoring dataset to generate a standardized multimodal data stream; S3. Dynamically evaluate transmission priority based on standardized multimodal data streams to generate a data transmission priority matrix; S4. Select the optimal satellite communication protocol and channel coding scheme based on the data transmission priority matrix and real-time satellite channel status information; S5. Based on the selected satellite communication protocol, perform distributed compression and encryption processing on the standardized multimodal data stream to generate a set of encrypted data packets; S6. Encrypted data packets are fragmented and transmitted using multi-satellite link aggregation technology, and dynamic link switching and error correction control are implemented during transmission. S7. Perform integrity verification and reassembly on the received data packets at the shore-based data center to generate a reconstructed multimodal dataset; S8. Perform data validity analysis and anomaly detection based on the reconstructed multimodal dataset, and generate a data quality assessment report; S9. Adjust the data acquisition and transmission strategy for deep-sea cages based on the feedback from the data quality assessment report; S10. Construct a satellite communication transmission performance evaluation model to achieve continuous optimization of the data transmission process.

[0022] The following operations are included in S1 for collecting cage environmental parameters, biological behavior data, and equipment operating status data: S11. Collect water temperature, salinity, dissolved oxygen concentration, and water flow velocity data through an underwater sensor array in the net cage to generate a hydrological environment dataset. The specific deployment method of the underwater sensor array in the net cage includes: S111. Temperature sensor clusters are deployed at the four corners and center of the cage to form a five-point temperature monitoring network. S112. A salinity sensor is installed every 2 meters along the depth of the cage, for a total of 8 salinity monitoring points; S113. Install high-precision dissolved oxygen sensors at the inlet and outlet of the net cage, with a sampling frequency of once per minute; S114. Measure the water flow velocity parameters around the cage using acoustic flow measurement equipment, with the measurement range covering the entire water depth profile of the cage. S12. Collect data on fish activity frequency, feeding behavior and group distribution using bioacoustic monitoring equipment to generate a biological behavior dataset; S13. Collect data on netting tension, anchor chain stress, and floating frame displacement using netting structure monitoring sensors to generate a structural safety dataset. S14. Collect data on solar panel output power, battery energy storage status and equipment energy consumption through the energy management system, and generate an energy status dataset. S15. Align the hydrological environment dataset, biological behavior dataset, structural safety dataset, and energy status dataset according to time series to generate the original multimodal monitoring dataset.

[0023] The adaptive data cleaning and format standardization process in S2 includes the following operations: S21. Perform outlier detection and removal on the original multimodal monitoring dataset, and use the sliding window-based Z-score algorithm to identify and replace data points that exceed the threshold range; The Z-score standardization formula is as follows: ; Z ij Let X be the standard score of the i-th data point of the j-th sensor. ij The original data values, j Let σ be the mean of the data from the j-th sensor within the sliding window. j Let be the standard deviation of the j-th sensor data within the sliding window; when When an outlier occurs, it is identified as an anomaly and replaced. The specific implementation steps include: S211. Set the sliding window size to W. size =60 data points, window sliding step size is S step =1 data point; S212. Calculate the mean and standard deviation of the data within the window: ; ; Where X uv Let r be the v-th feature value of the u-th sample in the dataset, where u is the sample index and r is the index of the starting position of the sliding window; S213. Calculate the Z-score value for each data point within the window. When using the mean μ of the data within the window. v Replace the outlier value; S214. Record the location and replacement value of outlier values ​​and generate a data quality log. S22. Perform timestamp alignment and resampling on sensor data with different sampling frequencies to make all data streams reach a unified time reference. S23. Convert all types of data into standard JSON format and add metadata descriptors to identify the data source, acquisition time, and sensor type; S24. Perform semantic consistency verification on the standardized data stream to ensure the correctness of data logical associations. The specific operations of semantic consistency verification include: S241. Establish a sensor data logic constraint rule library, including: water temperature range, salinity range, and dissolved oxygen concentration; S242. Real-time detection of whether the data stream violates logical constraint rules; S243. Mark the data that violates the rules and start the re-collection procedure; S244. Generate a semantic consistency verification report, recording the verification pass rate and details of abnormal data; S25. Output a standardized multimodal data stream with integrity markers.

[0024] Dynamic evaluation of transmission priority in S3 includes the following operations: S31. Establish a data priority evaluation index system, including data real-time requirements, security level weight, business criticality index, and data scarcity coefficient; S32. Calculate the weight vectors of each indicator based on the analytic hierarchy process (AHP) to generate a priority evaluation matrix: ; in For priority judgment matrix, Let p represent the importance of the p-th priority indicator relative to the q-th indicator, and n be the number of evaluation indicators. The specific steps for calculating the weights of each indicator using the analytic hierarchy process include: S321. Construct the judgment matrix The 1-9 scale method was used to quantify the relative importance among the indicators; S322. Calculate the eigenvectors of the judgment matrix. and the largest eigenvalue :

[0025] ; in This is the priority weight vector. The highest priority eigenvalue; S323. Perform a consistency check and calculate the consistency ratio. : ; in The priority consistency ratio, As a priority consistency indicator, Priority random consistency index; in , where n is the number of evaluation indicators; S324, when If the weight allocation result is accepted, the judgment matrix is ​​readjusted; otherwise, the judgment matrix is ​​readjusted. S33. Dynamically adjust the evaluation matrix parameters according to the cage operation scenario, and increase the weight of safety-related data in emergency situations; S34. Employ the fuzzy comprehensive evaluation method to prioritize each data stream and generate a data transmission priority matrix. Specific implementation includes: S341. Establishing a factor set These correspond to real-time performance, security, criticality, and scarcity, respectively. S342. Establish a comment collection These correspond to high priority, medium priority, and low priority, respectively. S343. Constructing a fuzzy relation matrix ,in Indicator Factors Comments Membership degree; S344. Perform fuzzy transformation: ; in The vector of fuzzy comprehensive evaluation results. For fuzzy weight vectors, Let be the fuzzy relation matrix, and o be the fuzzy composition operator; S345. Determine the final priority score based on the principle of maximum membership. S35. Based on the priority matrix, the data stream is divided into three levels: urgent transmission, regular transmission, and delayed transmission.

[0026] Selecting the optimal satellite communication protocol and channel coding scheme in S4 includes the following operations: S41. Real-time monitoring of signal strength, bit error rate, bandwidth utilization, and transmission delay parameters of available satellite channels; S42. Establish a satellite channel state assessment model to predict the communication quality of each channel within a future time window. The methods for constructing the satellite channel state assessment model include: S421. Use the ARIMA time series model to predict the future state of the channel to obtain forward-looking channel quality information: ; Where B is the shift operator. and These are the autoregressive and moving average polynomials, respectively, where d is the degree of the difference. Let be the channel state observation at time t. Let be the random error term at time t; S422. Establish the channel quality scoring function: ; in The overall channel quality score includes RSSI (Received Signal Strength Indicator), BER (Bit Error Rate), and SNR (Signal-to-Noise Ratio). , , These are the weighting coefficients; S423. Based on the scoring results, the channel is divided into four levels: excellent, good, average, and poor. S43. Select the lowest power communication protocol that meets the quality of service requirements based on the data transmission priority matrix; S44. Automatically select an enhanced forward error correction coding scheme for high-priority data, and generate an optimized configuration scheme for satellite communication protocols and coding schemes accordingly.

[0027] Distributed compression and encryption processing in S5 includes the following operations: S51. Select compression algorithm based on data characteristics: use lossy compression for environmental data and lossless compression for equipment status data; S52. Distributed compression technology is adopted, and compression operations are performed in parallel on the computing nodes at the edge of the cage. S53. Generate a dynamic encryption key based on the national cryptographic algorithm, and perform layered encryption on the compressed data packet. Specific implementation includes: S531. Data encryption is performed using the SM4 block cipher algorithm with a key length of 128 bits. S532. Use the SM3 hash algorithm to generate a digital signature with a hash value length of 256 bits. S533. Key exchange and digital signature verification are performed using the SM2 elliptic curve cryptography algorithm; S534, the encryption key is updated every 24 hours, and the forward security protection period is more than 30 days; S54. Add a digital signature and integrity check code to each data packet; S55. Group and package the encrypted data packets according to transmission priority to generate a set of encrypted data packets.

[0028] Multi-satellite link aggregation and fragmented transmission in S6 includes the following operations: S61. Establish a multi-satellite link aggregation pool to simultaneously connect more than three satellite communication systems in different orbits; S62. Adaptive fragmentation is performed based on the priority characteristics of the encrypted data packet set, with high-priority data being fragmented using multiple replicas; S63. Through real-time monitoring of link quality, dynamically select the optimal transmission path and perform seamless switching; S64. Implement a fountain code-based error correction control mechanism to adapt to the characteristics of high-latency, high-error-rate satellite channels. Specific operations include: S641. The encoding process generates an infinite number of encoded symbols. The probability distribution of each input symbol with a generation degree of d is as follows: ; in Let d be the probability and d be the degree. For ideal soliton distribution; S642, The receiver only needs to receive any The original data can be recovered with a high probability using only a few coded symbols. The number of original data packets; S643. Dynamically configure redundancy coefficients based on prediction results output by the satellite channel state assessment model; S65. Continuously optimize the fragmentation strategy and link allocation scheme during transmission.

[0029] Integrity verification and reconstruction in S7 includes the following operations: S71. Perform digital signature verification and decryption processing on the received data packets; S72. Check the integrity of the data packet and request retransmission of lost and corrupted data fragments; S73. Reconstruct the original data stream based on the fragment number and timestamp information; S74. Verify the temporal consistency and logical integrity of the recombined data; S75. Output the reconstructed multimodal dataset with quality labels.

[0030] Data validity analysis and anomaly detection in S8 include the following operations: S81. Perform outlier detection on the reconstructed multimodal dataset to identify data deviations caused by sensor malfunctions and transmission anomalies; S82. Confirm the authenticity of data through multi-source data cross-validation technology. Specific implementation includes: S821. Establish a physical mechanism correlation model among multi-source sensor data; S822. When data from a single sensor is abnormal, use data from related sensors to verify reliability. S823. A voting mechanism is adopted, and when more than 2 / 3 of the data from similar sensors are consistent, the data is determined to be reliable. S824. Mark data that cannot pass cross-validation and initiate a manual review process; S83. Calculate data integrity rate, timeliness and accuracy indicators; S84. Generate a data quality assessment report that includes data quality scores and anomaly alerts; S85. Feedback the assessment report to the cage site control system and shore-based monitoring platform.

[0031] The feedback adjustment of data acquisition and transmission strategies in S9 includes the following operations: S91. Adjust the sensor sampling frequency and transmission trigger threshold according to the data quality assessment report; S92. Optimize satellite communication link selection strategy and transmission timing arrangement.

[0032] The construction and continuous optimization of the transmission performance evaluation model in S10 includes the following operations: S101. Construct a transmission performance quantification model with throughput, energy efficiency ratio, and transmission success rate as the core evaluation dimensions. The transmission performance evaluation model is implemented through the following multivariate evaluation functions: ; in The overall score for transmission performance at time t. Let be the throughput at time t. Let be the energy consumption ratio at time t. Let be the transmission success rate at time t. Let be the weighting coefficient, satisfying ; S102. Continuously optimize the transmission parameter configuration based on the reinforcement learning algorithm. The reinforcement learning algorithm adopts the Q-learning algorithm, and its Q-value update follows the following Bellman equation: ; in The state at time t includes channel quality, data queue length, and remaining energy. The actions at time t include channel selection, transmission power, and modulation and coding scheme. The reward at time t is output by the transmission performance evaluation model. Calculations show that For learning rate, Discount factor; S103. Through continuous evaluation of the model and dynamic optimization of the algorithm, the transmission strategy can be adaptively adjusted.

[0033] The operational steps of this satellite communication data transmission method for deep-sea cages are as follows: Step 1: Multimodal Data Acquisition and Aggregation A multimodal sensor cluster deployed on deep-sea cages simultaneously collects multi-dimensional monitoring data on hydrological environment, biological behavior, structural safety, and energy status. The sensor array is distributed according to the optimal observation geometry to ensure the comprehensiveness and accuracy of data acquisition, generating a raw multimodal monitoring dataset with spatiotemporal alignment characteristics.

[0034] Step 2: Intelligent Preprocessing and Priority Evaluation An adaptive data cleaning engine is employed to standardize the original dataset, and a sliding window mechanism is used to detect and remove outliers, while also unifying the conversion of multi-source heterogeneous data formats. Based on the analytic hierarchy process (AHP) and fuzzy comprehensive evaluation theory, a data priority evaluation matrix is ​​constructed to finely classify data streams from multiple dimensions, including real-time requirements, security level, and business criticality, resulting in a three-level scheduling strategy for emergency transmission, regular transmission, and delayed transmission.

[0035] Step 3: Channel Awareness and Transmission Decision A satellite channel state assessment model is established to predict the quality change trends of each channel in real time through time series analysis. Based on the prediction results and data priority matrix, the optimal satellite communication protocol and channel coding scheme are dynamically selected, and an enhanced forward error correction mechanism is automatically activated for high-priority data to generate an intelligent transmission configuration scheme that integrates channel state and service requirements.

[0036] Step 4: Robust Transmission and Link Optimization A virtual transmission pool is constructed using multi-satellite link aggregation technology, and channel utilization is improved through multi-path parallel transmission and dynamic load balancing mechanisms. A fountain code encoding scheme is introduced, and elastically encoded data packets are generated based on a degree distribution algorithm to achieve reliable decoding under a specified packet loss rate threshold. Link quality is continuously monitored during transmission, and a switching mechanism ensures the continuous transmission of critical data.

[0037] Step 5: Closed-loop optimization and adaptive adjustment A transmission performance evaluation model is constructed to quantitatively analyze key indicators such as throughput, energy efficiency ratio, and transmission success rate. A reinforcement learning algorithm is employed, using the performance evaluation output as a reward signal to continuously optimize transmission parameter configuration. Through closed-loop control of evaluation-optimization-execution, the system achieves adaptive adjustment of the transmission strategy and self-optimization, ensuring optimal transmission performance in complex marine environments.

[0038] Step Six: Integrity Verification and Quality Assurance A multi-level integrity verification mechanism is implemented at the receiving end, using digital signature verification and data reassembly algorithms to ensure data integrity and authenticity. A multi-source data cross-validation model is established, using logical consistency checks and anomaly detection algorithms to identify transmission anomalies, generate a data quality assessment report, and feed it back to the sending end, forming an end-to-end quality assurance closed loop.

[0039] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0040] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A satellite communication data transmission method for deep-sea cages, characterized in that: The method includes the following steps: S1. Collect environmental parameters, biological behavior data, and equipment operation status data of the cages by deploying a multimodal sensor cluster on the deep-sea cages, and generate the original multimodal monitoring dataset; S2. Perform adaptive data cleaning and format standardization on the original multimodal monitoring dataset to generate a standardized multimodal data stream; S3. Based on the standardized multimodal data stream, dynamically evaluate the transmission priority and generate a data transmission priority matrix; S4. Select the optimal satellite communication protocol and channel coding scheme based on the data transmission priority matrix and real-time satellite channel status information; S5. Based on the selected satellite communication protocol, the standardized multimodal data stream is subjected to distributed compression and encryption processing to generate a set of encrypted data packets; S6. The encrypted data packet set is fragmented and transmitted using multi-satellite link aggregation technology, and dynamic link switching and error correction control are implemented during the transmission process; S7. Perform integrity verification and reassembly on the received data packets at the shore-based data center to generate a reconstructed multimodal dataset; S8. Based on the reconstructed multimodal dataset, perform data validity analysis and anomaly detection, and generate a data quality assessment report; S9. Adjust the data acquisition and transmission strategy of deep-sea cages based on the feedback from the data quality assessment report; S10. Construct a satellite communication transmission performance evaluation model to achieve continuous optimization of the data transmission process.

2. The satellite communication data transmission method for deep-sea cages according to claim 1, characterized in that: The collection of cage environmental parameters, biological behavior data, and equipment operating status data in S1 includes the following operations: S11. Collect water temperature, salinity, dissolved oxygen concentration and water flow velocity data through the underwater sensor array of the net cage, and generate a hydrological environment dataset; S12. Collect data on fish activity frequency, feeding behavior and group distribution using bioacoustic monitoring equipment to generate a biological behavior dataset; S13. Collect data on netting tension, anchor chain stress, and floating frame displacement using netting structure monitoring sensors to generate a structural safety dataset. S14. Collect data on solar panel output power, battery energy storage status and equipment energy consumption through the energy management system, and generate an energy status dataset. S15. Align the hydrological environment dataset, biological behavior dataset, structural safety dataset, and energy status dataset according to time series to generate the original multimodal monitoring dataset.

3. The satellite communication data transmission method for deep-sea cages according to claim 1, characterized in that: The adaptive data cleaning and format standardization process in S2 includes the following operations: S21. Perform outlier detection and removal on the original multimodal monitoring dataset, and use the sliding window-based Z-score algorithm to identify and replace data points that exceed the threshold range; The Z-score standardization formula is as follows: ； Z ij Let X be the standard score of the i-th data point of the j-th sensor. ij μ is the original data value. j Let σ be the mean of the data from the j-th sensor within the sliding window. j Let be the standard deviation of the j-th sensor data within the sliding window; when When an outlier occurs, it is identified as an anomaly and replaced. S22. Perform timestamp alignment and resampling on sensor data with different sampling frequencies to make all data streams reach a unified time reference. S23. Convert all types of data into standard JSON format and add metadata descriptors to identify the data source, acquisition time, and sensor type; S24. Perform semantic consistency verification on the standardized data stream to ensure the correctness of data logical associations; S25. Output a standardized multimodal data stream with integrity markers.

4. The satellite communication data transmission method for deep-sea cages according to claim 1, characterized in that: The dynamic evaluation of transmission priority in S3 includes the following operations: S31. Establish a data priority evaluation index system, including data real-time requirements, security level weight, business criticality index, and data scarcity coefficient; S32. Calculate the weight vectors of each indicator based on the analytic hierarchy process (AHP) to generate a priority evaluation matrix: ； in For priority judgment matrix, Let p represent the importance of the p-th priority indicator relative to the q-th indicator, and n be the number of evaluation indicators. S33. Dynamically adjust the evaluation matrix parameters according to the cage operation scenario, and increase the weight of safety-related data in emergency situations; S34. Use the fuzzy comprehensive evaluation method to score the priority of each data stream and generate a data transmission priority matrix; S35. Based on the priority matrix, the data stream is divided into three levels: urgent transmission, regular transmission, and delayed transmission.

5. The satellite communication data transmission method for deep-sea cages according to claim 1, characterized in that: The selection of the optimal satellite communication protocol and channel coding scheme in S4 includes the following operations: S41. Real-time monitoring of signal strength, bit error rate, bandwidth utilization, and transmission delay parameters of available satellite channels; S42. Establish a satellite channel status assessment model to predict the communication quality of each channel in future time windows; S43. Select the lowest power communication protocol that meets the quality of service requirements based on the data transmission priority matrix; S44. Automatically select an enhanced forward error correction coding scheme for high-priority data, and generate an optimized configuration scheme for satellite communication protocols and coding schemes accordingly.

6. The satellite communication data transmission method for deep-sea cages according to claim 1, characterized in that: The distributed compression and encryption processing in S5 includes the following operations: S51. Select compression algorithm based on data characteristics: use lossy compression for environmental data and lossless compression for equipment status data; S52. Distributed compression technology is adopted, and compression operations are performed in parallel on the computing nodes at the edge of the cage. S53. Generate a dynamic encryption key based on the national cryptographic algorithm, and perform layered encryption on the compressed data packet; S54. Add a digital signature and integrity check code to each data packet; S55. Group and package the encrypted data packets according to transmission priority to generate a set of encrypted data packets.

7. The satellite communication data transmission method for deep-sea cages according to claim 1, characterized in that: The multi-satellite link aggregation and fragmented transmission in S6 includes the following operations: S61. Establish a multi-satellite link aggregation pool to simultaneously connect more than three satellite communication systems in different orbits; S62. Adaptive fragmentation is performed based on the priority characteristics of the encrypted data packet set, with high-priority data being fragmented using multiple replicas; S63. Through real-time monitoring of link quality, dynamically select the optimal transmission path and perform seamless switching; S64. Implement a fountain code-based error correction control mechanism to adapt to the characteristics of high-latency and high-error-rate satellite channels; S65. Continuously optimize the fragmentation strategy and link allocation scheme during transmission.

8. The satellite communication data transmission method for deep-sea cages according to claim 1, characterized in that: The integrity verification and reconstruction in S7 includes the following operations: S71. Perform digital signature verification and decryption processing on the received data packets; S72. Check the integrity of the data packet and request retransmission of lost and corrupted data fragments; S73. Reconstruct the original data stream based on the fragment number and timestamp information; S74. Verify the temporal consistency and logical integrity of the recombined data; S75. Output the reconstructed multimodal dataset with quality labels.

9. The satellite communication data transmission method for deep-sea cages according to claim 1, characterized in that: The data validity analysis and anomaly detection in S8 include the following operations: S81. Perform outlier detection on the reconstructed multimodal dataset to identify data deviations caused by sensor malfunctions and transmission anomalies; S82. Confirm the authenticity of the data through multi-source data cross-validation technology; S83. Calculate data integrity rate, timeliness and accuracy indicators; S84. Generate a data quality assessment report that includes data quality scores and anomaly alerts; S85. Feedback the assessment report to the cage site control system and shore-based monitoring platform.

10. The satellite communication data transmission method for deep-sea cages according to claim 1, characterized in that: The feedback adjustment of the data acquisition and transmission strategy in S9 includes the following operations: S91. Adjust the sensor sampling frequency and transmission trigger threshold according to the data quality assessment report; S92. Optimize satellite communication link selection strategy and transmission timing arrangement.

Images