Delay optimization method for ship fusion communication

Abstract

The present application provides a delay optimization method for ship integrated communication, which relates to the field of wireless communication technology, including: synchronously monitoring K communication links of the target ship to obtain K monitoring indicator set sequences; performing time series correlation analysis and monitoring node centralized indicator analysis on the K monitoring indicator set sequences to determine K first link reliability factors and K second link reliability factors; obtaining multiple communication tasks and multiple task basic data of the target ship in a preset communication window, combining K first link reliability factors and K second link reliability factors to perform task link allocation, and obtain the target link allocation result; optimizing the communication delay of multiple communication tasks based on the target link allocation result. The present application solves the technical problem that the existing ship communication method lacks a comprehensive analysis and dynamic evaluation mechanism for the communication link, and is difficult to effectively deal with communication delays, and significantly improves the real-time performance of ship communication and the efficiency of data transmission.

Description

Technical Field

[0001] The present application relates to the field of wireless communication technology, and in particular to a delay optimization method for ship-based integrated communications. Background Art

[0002] With the increasing intelligence of ship equipment and the complexity of missions, the communication needs between ships are becoming more diverse and efficient. However, the marine environment is complex and changeable, and communication links are easily affected by external interference, channel congestion and other factors, resulting in communication delays, which affect the mission coordination and information exchange of ships.

[0003] Existing ship communication methods usually focus on a fixed strategy of a single link, and select communication links by evaluating single indicators such as signal strength and link quality. However, this method has shortcomings in dealing with delays: on the one hand, link selection is usually evaluated based on static indicators (such as signal strength or link quality), lacking analysis of the temporal dynamics and multi-dimensional correlation of link performance, making it difficult to respond to changes in link status in a timely manner. On the other hand, the allocation of task links usually adopts fixed rules, which cannot flexibly adapt to real-time changes in task requirements and link conditions, and easily leads to low resource utilization or task delay problems. In addition, the existing scheme lacks a collaborative mechanism with communication delay optimization, and the link selection and task execution process do not form an effective closed loop, which limits the improvement of overall communication performance. Summary of the invention

[0004] The present application provides a delay optimization method for ship integrated communications, which solves the technical problem that existing ship communication methods lack a comprehensive analysis and dynamic evaluation mechanism for communication links, making it difficult to effectively deal with communication delays, and achieves the technical effect of significantly improving the real-time performance of ship communications and data transmission efficiency.

[0005] In view of the above problems, the present application provides a delay optimization method for ship integrated communication, the method comprising: obtaining K communication links of the target ship, and performing synchronous monitoring on the K communication links to obtain K monitoring indicator set sequences, wherein K is an integer greater than or equal to 1; performing timing correlation analysis based on the K monitoring indicator set sequences to determine K first link reliability factors; traversing the K monitoring indicator set sequences to perform centralized indicator analysis of monitoring nodes to determine K second link reliability factors; obtaining multiple communication tasks and multiple task basic data of the target ship in a preset communication window, and performing task link allocation in combination with the K first link reliability factors and the K second link reliability factors to obtain a target link allocation result; and optimizing communication delays for the multiple communication tasks based on the target link allocation result.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] K communication links of the target ship are obtained, and the K communication links are monitored synchronously to obtain K monitoring indicator set sequences. These monitoring indicator set sequences can reflect the link status and are the original basis for the subsequent analysis of link reliability. Based on the K monitoring indicator set sequences, a time series correlation analysis is performed to explore the relationship between the monitoring indicators and time changes, determine K first link reliability factors, evaluate the reliability of the link from the time dimension, so as to grasp the status of the communication link at different time points, and provide an important reference for the subsequent reasonable allocation of task links. The K monitoring indicator set sequences are traversed to perform centralized indicator analysis of the monitoring nodes and determine K second link reliability factors. By centrally analyzing the multi-dimensional characteristics of the link node performance, the results of the time series correlation analysis are supplemented to more comprehensively evaluate the stability and reliability of the link. Multiple communication tasks and multiple task basic data of the target ship in the preset communication window are obtained, and task link allocation is performed in combination with the K first link reliability factors and the K second link reliability factors to obtain the target link allocation result. Communication delay optimization is performed on the multiple communication tasks based on the target link allocation result. This process can reasonably allocate communication tasks to different links according to the actual conditions of the links, maximize the use of link resources, and avoid allocating tasks to unreliable or congested links, thereby effectively reducing data transmission delays in the links and improving the data transmission efficiency of the entire ship's integrated communications.

[0008] In summary, this application conducts comprehensive monitoring of the ship's communication links, analyzes the reliability of the links from different angles through timing correlation analysis and monitoring node centralized indicator analysis, and thus obtains an accurate link reliability factor. Then, task links are allocated in combination with the relevant data of the communication task, and finally communication delay optimization is achieved. This solution effectively solves the communication delay problem caused by unreasonable link allocation in ship integrated communications, improves data transmission efficiency, enhances the stability and reliability of ship communication systems in complex environments, and ensures efficient communication needs in ship-related activities.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A flowchart of a delay optimization method for ship fusion communications provided in an embodiment of the present application.

[0011] Figure 2A schematic diagram of a process for determining K first link reliability factors in a delay optimization method for ship fusion communication provided in an embodiment of the present application.

[0012] Figure 3 A schematic diagram of a flow chart for determining K second link reliability factors in a delay optimization method for ship fusion communication provided in an embodiment of the present application. DETAILED DESCRIPTION

[0013] The embodiment of the present application provides a delay optimization method for ship integrated communications. Through synchronous monitoring, timing correlation analysis and concentrated index analysis of multiple communication links, the reliability factor of the link is determined, task link allocation is performed in combination with the link reliability factor, and communication delay optimization is performed on the communication task based on the target link allocation result. This solves the technical problem that the existing ship communication methods lack a comprehensive analysis and dynamic evaluation mechanism for the communication links, which makes it difficult to effectively deal with communication delays, and achieves the technical effect of significantly improving the real-time performance of ship communications and data transmission efficiency.

[0014] like Figure 1 As shown, an embodiment of the present application provides a delay optimization method for ship fusion communication, the method comprising:

[0015] Step S1: Acquire K communication links of the target ship, and perform concurrent monitoring on the K communication links to obtain K monitoring indicator set sequences, where K is an integer greater than or equal to 1.

[0016] Specifically, the target ship can be any ship that needs to communicate in a specific mission and is the implementation object of the embodiment of the present application. The communication link refers to the data transmission path established between the ship and other communication nodes (such as other ships, satellites, base stations, etc.).

[0017] Deploy monitoring equipment on the target ship or activate the existing monitoring system of the ship to synchronously monitor the performance of K communication links on the target ship. In the embodiment of the present application, K is a positive integer, which represents the number of communication links available to the target ship. These communication links include different types of physical or wireless connections, such as satellite communications, radio waves, microwaves, etc. The monitoring equipment will record the key performance indicators of each link, such as signal strength, transmission rate, bit error rate, etc., to form a series of monitoring indicator sets. And the multiple monitoring indicator sets corresponding to each link are serialized according to time to obtain K monitoring indicator set sequences corresponding to K communication links. Each element in the sequence (i.e., monitoring indicator set) represents the monitoring data of a communication link at a point in time.

[0018] Through this step, the communication link performance of the target ship can be recorded in real time, providing accurate raw data for subsequent link analysis and task link allocation.

[0019] Step S2: Performing a time series correlation analysis based on the K monitoring indicator set sequences to determine K first link reliability factors.

[0020] Specifically, the first link reliability factor is a value obtained based on timing correlation analysis to measure the reliability of the communication link, and the factor reflects the reliability of the communication link at different time points.

[0021] For each monitoring indicator set sequence, the relationship between each monitoring indicator is analyzed in chronological order. For example, starting from the indicator value at the first time point, its changing trend and mutual influence with the indicator values ​​at subsequent time points are analyzed. During the analysis process, some mathematical methods or algorithms can be used to quantify this temporal correlation relationship. Through the temporal correlation analysis of the entire monitoring indicator set sequence, the performance of each indicator in the time dimension is comprehensively considered, and finally a value corresponding to each communication link is determined, that is, K first link reliability factors.

[0022] Through timing correlation analysis, the stability of each communication link can be evaluated from the time dimension, thus providing a reference for subsequent task link allocation and delay optimization.

[0023] Step S3: traverse the K monitoring indicator set sequences to perform centralized indicator analysis on monitoring nodes and determine K second link reliability factors.

[0024] Specifically, a monitoring node is a specific location or device in a communication link. There may be multiple communication devices on a ship as monitoring nodes, such as communication antennas, signal transceivers, etc. Various monitoring indicators such as signal strength and noise level can be obtained from these nodes. The second link reliability factor is a value used to measure the reliability of a communication link by analyzing the centralized indicators of the monitoring nodes. This factor reflects the reliability of the link from the perspective of the monitoring node.

[0025] Traverse the K monitoring indicator set sequences, and for each monitoring indicator set sequence, extract the indicators related to the monitoring nodes. The multiple monitoring nodes of the target ship may be located in different areas of the ship (such as the bridge, engine room, deck, etc.), and the communication link status of each node will be different. Perform centralized indicator analysis on the indicators of these monitoring nodes. For example, first summarize the same type of indicators of different monitoring nodes (such as the signal strength indicators of all monitoring nodes). Then, through statistical analysis (such as calculating the mean, standard deviation and other statistics of these indicators), or cluster analysis (classifying the indicators of monitoring nodes with similar characteristics into one category), the indicator information of each monitoring node is integrated, and finally the K second link reliability factors corresponding to each communication link are determined.

[0026] Through this step, the spatial reliability of each link can be comprehensively evaluated through the performance of multiple monitoring nodes, providing a more comprehensive basis for subsequent task link allocation.

[0027] Step S4: Acquire multiple communication tasks and multiple task basic data of the target ship in the preset communication window, perform task link allocation in combination with the K first link reliability factors and the K second link reliability factors, and obtain a target link allocation result.

[0028] Specifically, the preset communication window is a pre-set time range, during which the ship has communication needs and will arrange corresponding communication tasks. Communication tasks refer to various communication-related tasks that the target ship needs to complete within the preset communication window, such as sending navigation data, receiving instructions from the command center, and coordinating information exchange with other ships. Task basic data is the basic information related to each communication task, including the data volume of the communication task, the priority of the task, the transmission stability and speed requirements, etc., which will affect the task link allocation. The target link allocation result is the best match between each communication task and the communication link based on the task requirements and the reliability factor of the link.

[0029] Obtain multiple communication tasks that the target ship is to perform within the preset communication window and the task basic data corresponding to each task. For each communication task, the task link allocation algorithm is used to evaluate the adaptability of different links to the task based on its task basic data, comprehensively consider the first link reliability factor and the second link reliability factor of each communication link, select a suitable link for each task, and finally obtain the link allocation result for each task. For example, if a communication task has a large amount of data and a high priority, it needs to be allocated to a link that has both good timing stability (higher first link reliability factor) and excellent performance in monitoring node indicators (higher second link reliability factor).

[0030] This step ensures that all tasks can be assigned to the most suitable links within the preset communication window, thereby minimizing communication delays, improving task completion efficiency, and achieving optimal resource allocation.

[0031] Step S5: Optimizing communication delays for the multiple communication tasks based on the target link allocation result.

[0032] Specifically, according to the target link allocation result of step S4, the communication task is executed with the target communication link corresponding to each task within the preset window to reduce the communication delay. Before executing the task, it is also necessary to evaluate the load condition, bandwidth requirement and delay requirement of each link according to the target link allocation result to avoid the increase of communication delay due to overload of some links. Dynamically adjust the task allocation through hash load balancing, delay minimization scheduling algorithm, etc. to achieve load balancing. In practical applications, the state of the link may fluctuate over time. Therefore, it is also necessary to monitor the state of the communication link in real time within the preset window, and adjust the allocation of tasks according to the actual communication delay feedback. For example, if link 1 has a high delay at a certain moment, a part of the tasks will be migrated to link 2 to ensure the real-time nature of the communication task.

[0033] Through this step, further communication delay optimization is performed based on the task link allocation results, and the link load is dynamically adjusted to reduce communication delay and improve communication real-time performance.

[0034] Further, such as Figure 2 As shown, step S2 of the embodiment of the present application also includes:

[0035] Step S21: extracting the first K monitoring indicator sets and the second K monitoring indicator sets from the K monitoring indicator set sequences respectively.

[0036] Step S22: performing time series correlation analysis on the K first monitoring indicator sets and the K second monitoring indicator sets to obtain K first time series correlation monitoring indicator sets.

[0037] Step S23: extract the K third monitoring indicator sets located in the third position from the K monitoring indicator set sequences again, perform time series correlation analysis on the K third monitoring indicator sets and the K first time series correlation monitoring indicator sets, and determine K second time series correlation monitoring indicator sets.

[0038] Step S24: After multiple time series correlation analyses, until the end of the K monitoring indicator set sequence is reached, K target time series correlation monitoring indicator sets are obtained.

[0039] Step S25: performing link reliability analysis on the K target timing-related monitoring indicator sets to obtain the K first link reliability factors.

[0040] Specifically, the target time-series correlation monitoring indicator set is a monitoring data set obtained through time-series correlation analysis, and is used to describe the performance changes of the link in the time dimension.

[0041] From the collected K monitoring indicator set sequences, extract the first K monitoring indicator sets, recorded as the first monitoring indicator set; at the same time, extract the second K monitoring indicator sets, recorded as the second monitoring indicator set. The K first monitoring indicator sets represent the states of the K communication links at the first moment (t0), and the K second monitoring indicator sets represent the states of the K communication links at the second moment (t1).

[0042] Perform time series correlation analysis on the K first monitoring indicator sets and the K second monitoring indicator sets to monitor the changing rules of the indicators and obtain K first time series correlation monitoring indicator sets. Time series correlation analysis can reveal the changing patterns of link performance indicators (such as delay and bandwidth) in the time series and their mutual relationships, thereby helping to evaluate the stability of the link. For example, the delay of link 1 at time t0 is 50ms, and the delay at time t1 is 55ms. By analyzing the relationship between these two data points, the changing trend of link 1 during this period can be obtained.

[0043] The third monitoring indicator set is extracted from the monitoring indicator set sequence and recorded as the third monitoring indicator set. The K third monitoring indicator sets represent the states of the K communication links at the third moment (t2). The third monitoring indicator set corresponding to each link is subjected to a time series correlation analysis with the first time series correlation monitoring indicator set of the link obtained above. The relationship between the indicator at the third moment and the indicator after correlation analysis at the first two moments is considered. The development trend of the indicator is analyzed to see whether it continues or has new change characteristics, thereby determining the K second time series correlation monitoring indicator sets.

[0044] Repeat the above operation, extract a new monitoring indicator set at each location each time, and perform time series correlation analysis with the time series correlation monitoring indicator set obtained in the previous round, until the end of the K monitoring indicator set sequence. Through multiple time series correlation analyses, the time series characteristics of link performance are gradually constructed, and finally K target time series correlation monitoring indicator sets corresponding to K links are obtained.

[0045] Perform link reliability analysis on K target timing-related monitoring indicator sets, evaluate the reliability of each link, and obtain K first link reliability factors of the K links. Exemplarily, according to the set reliability evaluation criteria, weighted calculation, threshold judgment and other operations can be performed on the indicators in each target timing-related monitoring indicator set to generate the first link reliability factor of each link. These reliability factors will serve as the basis for subsequent task link allocation and delay optimization, helping to select the most stable and reliable links for task execution, thereby improving the overall performance and stability of the ship communication system.

[0046] Furthermore, step S22 further includes:

[0047] Step S221: Calculate the inner product of the K first monitoring indicator sets and the K second monitoring indicator sets to obtain K first time series association similarity sets.

[0048] Step S222: normalizing the K first time series association similarity sets using a softmax algorithm to obtain K first time series association matrices.

[0049] Step S223: performing convolution processing on the K second monitoring indicator sets and the K first time series association matrices to obtain K first time series association monitoring indicator sets.

[0050] Specifically, taking K first monitoring indicator sets and K second monitoring indicator sets as an example, the specific process of performing time series correlation analysis on the two monitoring indicator sets is explained.

[0051] First, each indicator in the K first monitoring indicator sets and the K second monitoring indicator sets is converted into a vector representation to obtain K first vector sets and K second vector sets. Then, the inner product of the vectors corresponding to the same indicator in the K first vector sets and the K second vector sets is calculated using the multidimensional space cosine formula to reflect the similarity of the indicator in the two sets. The larger the inner product result, the higher the similarity between the monitoring data at the two time points. Exemplarily, in a monitoring indicator set sequence of a communication link, for monitoring indicator I, its vector representation in the first monitoring indicator set and the second monitoring indicator set is A respectively. i and B i , calculate A i and B i The inner product of A i •B i , represents the change trend and degree of the indicator I of the communication link at the first moment and the second moment. By calculating the inner product of each monitoring indicator of each communication link one by one, K first time series correlation similarity sets corresponding to the K communication links are obtained.

[0052] Each first time series association similarity set is subjected to softmax normalization, that is, the original vector product result is replaced by the proportion of each vector product to the sum of all vector products in its set, and finally a normalized first time series association matrix is ​​obtained. Through normalization, the different similarity values ​​can be standardized to a unified range, eliminating the influence of dimension. At the same time, the normalized result also reflects the importance of the corresponding indicator in the change of the communication link state.

[0053] The K second monitoring indicator sets and the K first time series association matrices normalized by softmax are input into the graph convolutional network for convolution operation. The convolution operation can extract correlation features between time series data, so as to conduct a more detailed analysis of the performance of the link. The operation results are K first time series association monitoring indicator sets, which combine the information of the second monitoring indicator set and the first time series association matrix, and can better reflect the time series association characteristics of each communication link.

[0054] Furthermore, step S25 further includes:

[0055] Step S251: construct a link reliability analyzer.

[0056] Step S252: Utilize the link reliability analyzer to analyze the K target timing-related monitoring indicator sets to obtain the K first link reliability factors.

[0057] Specifically, a link reliability analyzer is a tool or model specifically used to analyze link reliability. It can be built based on mathematical formulas, algorithms, or machine learning models. The link reliability analyzer can output the reliability factor of the link based on the input target timing-related monitoring indicator set. The K target timing-related monitoring indicator sets are input into the constructed link reliability analyzer, and the constructed link reliability analyzer conducts in-depth analysis of these convolution-processed timing data, and finally obtains K first link reliability factors, thereby accurately evaluating the stability of the communication link in different time periods, and providing reference information in the time dimension for task link allocation and delay optimization.

[0058] Exemplarily, in one embodiment, a link reliability analyzer is constructed using a weighted summation model: specific monitoring indicators in the target timing-related monitoring indicator set are obtained, for example, the monitoring indicators include signal strength, signal fluctuation range, transmission rate, bit error rate, etc. The relative importance of each indicator to link reliability, i.e., the weight, is determined based on domain knowledge, expert experience, or data analysis methods. For example, if the signal strength has a greater impact on link reliability, a higher weight can be given; and if the bit error rate has a relatively small impact, a lower weight is given. The analytic hierarchy process can also be used to determine the weights of each indicator by building a hierarchical model, constructing a judgment matrix, and calculating a weight vector. Then, using the weighted summation formula, the value of each monitoring indicator in the target timing-related monitoring indicator set of each communication link is multiplied by the weight of the indicator, and then the sum is calculated to obtain the first link reliability factor of the communication link.

[0059] In another embodiment, a link reliability analyzer is constructed using a decision tree model: First, historical performance data of each communication link of the target ship communication system is collected. These historical data are labeled according to the actual performance of the link performance to reflect the status of the communication link. For example, a label of 1 indicates reliability, and a label of 0 indicates unreliability. The historical data is divided into a training set and a test set, and the decision tree model is trained using the training set. Then, the performance of the model is tested using the test set, and the accuracy of the model is verified by calculating indicators such as accuracy, recall rate, and F1 score. The analyzer is optimized according to the verification results, and the model such as the depth of the tree and the selection criteria for splitting nodes are adjusted until the model converges, and the trained link reliability analyzer is output. Finally, a set of K target timing-related monitoring indicators is input into the link reliability analyzer, and the analyzer outputs K first link reliability factors.

[0060] Further, such as Figure 3 As shown, step S3 of the embodiment of the present application also includes:

[0061] Step S31: traverse the K monitoring indicator set sequences to extract the monitoring indicator sets of the K communication links under the first monitoring node, and obtain K first monitoring node indicator sets.

[0062] Step S32: performing reliability analysis on the K first monitoring node indicator sets respectively to obtain K first monitoring node reliability factors.

[0063] Step S33: extract the K monitoring indicator set sequences again to extract the monitoring indicator sets of the K communication links under the second monitoring node, and obtain K second monitoring node indicator sets.

[0064] Step S34: performing reliability analysis on the K second monitoring node indicator sets respectively to obtain K second monitoring node reliability factors.

[0065] Step S35: After multiple extraction and analysis, a set of K monitoring node reliability factors is obtained.

[0066] Step S36: performing centralized analysis on the K monitoring node reliability factor sets to determine K second link reliability factors.

[0067] Specifically, the monitoring node reliability factor is used to represent the reliability of the link under a specific monitoring node and is used to quantify the performance of the link at the node.

[0068] The first monitoring node is a specific monitoring point in the communication link. Traverse the K monitoring indicator set sequences, and for each sequence, select the monitoring indicators related to the first monitoring node from all the monitoring indicator sets of the sequence, and combine these indicators to form K first monitoring node indicator sets. The link reliability analyzer constructed above will be used to perform reliability analysis on the K first monitoring node indicator sets to obtain K first monitoring node reliability factors.

[0069] Select another monitoring point in the communication link that is different from the first monitoring node and record it as the second monitoring node. Similar to step S31, traverse the K monitoring indicator set sequences again, extract the monitoring indicators related to the K communication links under the second monitoring node, and then combine these indicators into K second monitoring node indicator sets. Similar to step S32, perform reliability analysis on the K second monitoring node indicator sets respectively to obtain K second monitoring node reliability factors.

[0070] Repeat the extraction and analysis process from step S31 to step S32 for many times, and continuously obtain the monitoring node reliability factors of each monitoring node for different monitoring nodes. Integrate the monitoring node reliability factors of each monitoring node under each communication link, and finally obtain K monitoring node reliability factor sets. Centrally analyze the K monitoring node reliability factor sets, comprehensively consider the impact of each node on the link, and determine K second link reliability factors.

[0071] By traversing the monitoring indicator set sequence, analyzing the performance of the link node by node, and integrating the reliability factor set of all monitoring nodes, the second link reliability factor of each link is calculated, making the link reliability evaluation more comprehensive and accurate, and providing a comprehensive reference basis for subsequent task link allocation.

[0072] Furthermore, step S36 also includes:

[0073] Step S361: respectively calculating the mean values ​​of the K monitoring node reliability factor sets to obtain the mean values ​​of the K monitoring node reliability factors.

[0074] Step S362: taking the mean of the reliability factors of the K monitoring nodes as the cluster center and taking the preset centralized bandwidth as the cluster scale, performing centralized analysis on the set of reliability factors of the K monitoring nodes to determine K mean cluster neighborhoods.

[0075] Step S363: Count the K neighborhood densities of the K mean clustering neighborhoods. If the K neighborhood densities are greater than or equal to a preset neighborhood density threshold, use the mean of the K monitoring node reliability factors as the K second link reliability factors.

[0076] Furthermore, step S363 also includes:

[0077] Step S363-1: If not, randomly extract a monitoring node reliability factor from the neighborhood edge of the K mean clustering neighborhoods as K iterative clustering centers.

[0078] Step S363-2: Obtain the K iterative clustering neighborhoods of the K iterative clustering centers, and determine whether the neighborhood density of the K iterative clustering neighborhoods is greater than or equal to the preset neighborhood density threshold; if not, randomly extract a monitoring node reliability factor from the neighborhood edge of the K mean clustering neighborhoods to update the K iterative clustering centers.

[0079] Step S363-3: If yes, the K iterative clustering centers are used as the K second link reliability factors.

[0080] Specifically, the cluster center represents the core position of a cluster in the data clustering process and is used to divide data into different groups (clusters). The preset concentration bandwidth is a parameter that defines the neighborhood range, which indicates the distance or range that the cluster center can influence, that is, the cluster scale. The mean clustering neighborhood is a neighborhood area around the cluster center, consisting of factors within the clustering range. The neighborhood density is the number of factors in a neighborhood, which is used to judge the effectiveness of the neighborhood. The preset neighborhood density threshold is a pre-set value used to judge whether the reliable factors in the mean clustering neighborhood are dense enough. If the neighborhood density is greater than or equal to this threshold, it means that the reliable factors in the neighborhood are representative. The iterative cluster center is a new cluster center obtained by iterative updating when the initial clustering does not meet the requirements.

[0081] For each monitoring node reliability factor set, the average value of all monitoring node reliability factors in the set is calculated to obtain the mean value of the monitoring node reliability factor. This mean value represents the overall level of link reliability of the monitoring node to a certain extent.

[0082] The calculated mean of the reliability factors of K monitoring nodes is used as the cluster center, and the data in the corresponding monitoring node reliability factor set whose distance from the cluster center is less than the preset centralized bandwidth is divided into a neighborhood, thereby obtaining K mean cluster neighborhoods.

[0083] For K mean clustering neighborhoods, the number of reliability factors in each neighborhood is counted and used as the neighborhood density of each neighborhood. The neighborhood density of each neighborhood is compared with the preset neighborhood density threshold. If a neighborhood density is greater than or equal to the preset neighborhood density threshold, the corresponding monitoring node reliability factor mean is used as the second link reliability factor of the link.

[0084] If the density of a neighborhood is less than the preset neighborhood density threshold, then the monitoring node reliability factor is randomly extracted from the area near the neighborhood boundary in the mean clustering neighborhood, that is, the neighborhood edge, as the iterative clustering center. For the K iterative clustering centers, the K corresponding iterative clustering neighborhoods are determined according to a method similar to step S362, with the preset concentrated bandwidth as the clustering scale. The neighborhood densities of these iterative clustering neighborhoods are calculated, and it is determined whether these neighborhood densities are greater than or equal to the preset neighborhood density threshold. If the neighborhood density of the iterative clustering neighborhood is less than the preset neighborhood density threshold, a monitoring node reliability factor is randomly extracted from the neighborhood edge of the K mean clustering neighborhoods again to update the K iterative clustering centers. If the neighborhood density of the iterative clustering neighborhood is greater than or equal to the preset neighborhood density threshold, these iterative clustering centers are used as K second link reliability factors.

[0085] The above iterative steps are repeated continuously until K second link reliability factors corresponding to K links are determined. These second link reliability factors can fully reflect the overall performance of the link, thereby providing a more accurate reliability evaluation basis for subsequent task link allocation.

[0086] Furthermore, step S4 of the embodiment of the present application also includes:

[0087] Step S41: performing weighted calculation on the K first link reliability factors and the K second link reliability factors according to preset weight values ​​to obtain K link reliability coefficients.

[0088] Step S42: Utilize the task link allocator to analyze the plurality of task basic data and the K link reliability coefficients to obtain the target link allocation result.

[0089] Specifically, corresponding weight values ​​are set for the first link reliability factor and the second link reliability factor, respectively, and these weight values ​​reflect the relative importance of the first link reliability factor and the second link reliability factor in the final result. The first link reliability factor and the second link reliability factor corresponding to each communication link are weighted according to the preset weight values ​​to determine the link reliability coefficient of the communication link. This value combines the information of the two link reliability factors and can more comprehensively reflect the reliability of the link. The larger the link reliability coefficient, the higher the link reliability.

[0090] Multiple task basic data and K link reliability coefficients are input into the task link allocator. The task link allocator performs analysis according to the internal algorithm or model logic and outputs the target link allocation result. This allocation result clarifies the communication link corresponding to each task. Among them, the task link allocator is a tool or model for analyzing the matching degree between the communication task requirements and the link performance, and executing the allocation decision. The task link allocator can be constructed based on algorithms such as greedy algorithm and linear programming. Exemplarily, when the task link allocator is constructed based on the greedy algorithm, the task link allocator first sorts the tasks according to the priority, and then selects the link with the highest link reliability coefficient to allocate the highest priority task, and then allocates other tasks in turn. When the task link allocator is constructed based on the linear programming algorithm, first set the constraints, such as the bandwidth of the link must be greater than or equal to the task requirement, the link delay must be less than the task tolerance, and each task must be allocated to one or more links, and then construct the objective function with the goal of maximizing the weighted sum of the priority score of the task and the link reliability coefficient. For each task, the matching score of the task and each link is calculated using the objective function, and the result with the largest score is selected as the target link allocation result of the task.

[0091] The link reliability coefficient is determined by combining the first link reliability factor and the second link reliability factor to comprehensively evaluate the reliability of the communication link. Then, a suitable link is allocated to each task based on the specific requirements of the task and the reliability of the communication link. This maximizes the utilization efficiency of link resources while ensuring communication quality and reducing communication delays.

[0092] In summary, the delay optimization method for ship fusion communication provided by the embodiment of the present application has the following technical effects:

[0093] K communication links of the target ship are obtained, and the K communication links are monitored synchronously to obtain K monitoring indicator set sequences. These monitoring indicator set sequences can reflect the link status and are the original basis for the subsequent analysis of link reliability. Based on the K monitoring indicator set sequences, a time series correlation analysis is performed to explore the relationship between the monitoring indicators and time changes, determine K first link reliability factors, evaluate the reliability of the link from the time dimension, so as to grasp the status of the communication link at different time points, and provide an important reference for the subsequent reasonable allocation of task links. The K monitoring indicator set sequences are traversed to perform centralized indicator analysis of the monitoring nodes and determine K second link reliability factors. By centrally analyzing the multi-dimensional characteristics of the link node performance, the results of the time series correlation analysis are supplemented to more comprehensively evaluate the stability and reliability of the link. Multiple communication tasks and multiple task basic data of the target ship in the preset communication window are obtained, and task link allocation is performed in combination with the K first link reliability factors and the K second link reliability factors to obtain the target link allocation result. Communication delay optimization is performed on the multiple communication tasks based on the target link allocation result. This process can reasonably allocate communication tasks to different links according to the actual conditions of the links, maximize the use of link resources, and avoid allocating tasks to unreliable or congested links, thereby effectively reducing data transmission delays in the links and improving the data transmission efficiency of the entire ship's integrated communications.

[0094] In general, the embodiment of the present application comprehensively evaluates link reliability from two dimensions, namely, timing correlation analysis and monitoring node centralized index analysis, through the steps of link performance monitoring, timing and centralized index analysis, task link allocation, and delay optimization, and obtains the first and second link reliability factors. On this basis, the weighted allocation method is used to optimize the task link allocation in combination with task priority and basic data, and finally achieves communication delay optimization. This method realizes closed-loop optimization from link evaluation to task execution, minimizes communication link delay, significantly improves the real-time performance of ship communications and data transmission efficiency, enhances the stability and reliability of ship communication systems in complex marine environments, and ensures the efficient communication needs of ships in various navigation tasks.

[0095] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A delay optimization method for ship fusion communication, characterized in that: The method comprises: Acquire K communication links of the target ship, and monitor the K communication links simultaneously to obtain K monitoring indicator set sequences, where K is an integer greater than or equal to 1; Performing time series correlation analysis based on the K monitoring indicator set sequences to determine K first link reliability factors; Traversing the K monitoring indicator set sequences to perform centralized indicator analysis on monitoring nodes, and determining K second link reliability factors; Acquire multiple communication tasks and multiple task basic data of the target ship in a preset communication window, perform task link allocation in combination with the K first link reliability factors and the K second link reliability factors, and obtain a target link allocation result; Optimizing communication delays for the plurality of communication tasks based on the target link allocation result; Traversing the K monitoring indicator set sequences to perform monitoring node centralized indicator analysis and determine K second link reliability factors, including: Traversing the K monitoring indicator set sequences to extract the monitoring indicator sets of the K communication links under the first monitoring node, and obtaining K first monitoring node indicator sets; Performing reliability analysis on the K first monitoring node indicator sets respectively to obtain K first monitoring node reliability factors; Extract the K monitoring indicator set sequences again to extract the monitoring indicator sets of the K communication links under the second monitoring node, and obtain K second monitoring node indicator sets; Performing reliability analysis on the K second monitoring node indicator sets respectively to obtain K second monitoring node reliability factors; After multiple extraction and analysis, a set of K monitoring node reliability factors is obtained; The K monitoring node reliability factor sets are centrally analyzed to determine K second link reliability factors.

2. The delay optimization method for ship fusion communication according to claim 1, characterized in that: Performing a time series correlation analysis based on the K monitoring indicator set sequences to determine K first link reliability factors includes: Extracting the first K monitoring indicator sets and the second K second monitoring indicator sets from the K monitoring indicator set sequences respectively; Performing time series correlation analysis on the K first monitoring indicator sets and the K second monitoring indicator sets to obtain K first time series correlation monitoring indicator sets; Again, K third monitoring indicator sets located at the third position are respectively extracted from the K monitoring indicator set sequences, and time series correlation analysis is performed on the K third monitoring indicator sets and the K first time series correlation monitoring indicator sets to determine K second time series correlation monitoring indicator sets; After multiple time series correlation analyses, until the end of the K monitoring indicator set sequence is reached, K target time series correlation monitoring indicator sets are obtained; Perform link reliability analysis on the K target timing-related monitoring indicator sets to obtain the K first link reliability factors.

3. The delay optimization method for ship fusion communication according to claim 2, characterized in that: include: Calculating the inner product of the K first monitoring indicator sets and the K second monitoring indicator sets to obtain K first time series association similarity sets; Normalizing the K first time series association similarity sets using a softmax algorithm to obtain K first time series association matrices; The K second monitoring indicator sets and the K first time series association matrices are convolved to obtain K first time series association monitoring indicator sets.

4. The delay optimization method for ship fusion communication according to claim 2, characterized in that: Performing link reliability analysis on the K target timing-related monitoring indicator sets to obtain the K first link reliability factors includes: Building a link reliability analyzer; The link reliability analyzer is used to analyze the K target timing-related monitoring indicator sets to obtain the K first link reliability factors.

5. The delay optimization method for ship fusion communication according to claim 1, characterized in that: include: Calculating the mean values ​​of the K monitoring node reliability factor sets respectively to obtain the mean values ​​of the K monitoring node reliability factors; Taking the mean of the K monitoring node reliability factors as the cluster center and taking the preset centralized bandwidth as the cluster scale, the K monitoring node reliability factor sets are centrally analyzed to determine K mean cluster neighborhoods; The K neighborhood densities of the K mean clustering neighborhoods are counted, and if the K neighborhood densities are greater than or equal to a preset neighborhood density threshold, the mean of the K monitoring node reliability factors is used as the K second link reliability factors.

6. The delay optimization method for ship fusion communication according to claim 5, characterized in that: include: If not, randomly extract a monitoring node reliability factor from the neighborhood edge of the K mean clustering neighborhoods as K iterative clustering centers; Obtain K iterative clustering neighborhoods of the K iterative clustering centers, determine whether the neighborhood density of the K iterative clustering neighborhoods is greater than or equal to the preset neighborhood density threshold, and if not, randomly extract a monitoring node reliability factor from the neighborhood edge of the K mean clustering neighborhoods again to update the K iterative clustering centers; If so, the K iterative clustering centers are used as the K second link reliability factors.

7. The delay optimization method for ship fusion communication according to claim 1, characterized in that: Acquiring multiple communication tasks and multiple task basic data of the target ship in a preset communication window, performing task link allocation in combination with the K first link reliability factors and the K second link reliability factors, and obtaining a target link allocation result, including: Performing weighted calculation on the K first link reliability factors and the K second link reliability factors according to preset weight values ​​to obtain K link reliability coefficients; The task link allocator is used to analyze the multiple task basic data and the K link reliability coefficients to obtain the target link allocation result.

Images