Method for ship communication link establishment based on perception control

Abstract

The application relates to the technical field of ship wireless communication network, and discloses a ship communication link establishment method based on perception control. The method extracts a ship historical communication behavior data set, analyzes the dense and violent change degree of the time distribution, calculates a communication environment dynamic index, and thus screens candidate areas with stable communication environment. For each candidate area, a perception feature vector and a link establishment feature vector of the ship are respectively constructed, and the perception contribution degree and the communication node weight are calculated. The node weight is sequenced and hierarchically divided, a hierarchical communication network architecture is formed, and the communication mode is adaptively allocated to different levels. The application quantitatively evaluates the communication environment dynamicity and the node behavior characteristics, realizes dynamic optimization of the network structure and resource allocation, and enhances the adaptability and overall reliability of the ship communication system in a complex sea environment.

Description

Technical Field

[0001] This invention relates to the field of ship wireless communication network technology, specifically a ship communication link establishment method based on perception control. Background Technology

[0002] Traditional ship communication networking methods are typically based on static network planning or pre-defined fixed link protocols. These technologies treat the communication area as a stable and unchanging environment, making decisions primarily based on geographical location or instantaneous signal strength. Existing solutions lack quantitative analysis of the temporal characteristics of communication behavior and cannot effectively capture and respond to the inherent time-varying characteristics of maritime radio channels. This results in insufficient adaptability and predictability in network construction in dynamic sea areas with dense or rapidly changing communication behaviors.

[0003] Existing technologies for assigning network node roles and selecting paths largely rely on nodes' physical location, remaining energy, or simple communication history. Conventional methods fail to extract and correlate node perception and link-building behavior characteristics separately, and the calculation model for node weights is relatively simplistic. This approach cannot accurately assess the comprehensive performance of nodes in complex environments and is insufficient to support the construction of hierarchical network architectures that combine high connectivity and strong environmental adaptability.

[0004] This invention aims to address how to dynamically assess the time-varying characteristics of the marine communication environment to guide the initial selection of networking areas, and how to perform accurate weight calculation and hierarchical division based on the multidimensional behavioral characteristics of nodes, thereby achieving intelligent matching of communication resources with the dynamic environment and node status. Summary of the Invention

[0005] The purpose of this invention is to provide a ship communication link establishment method based on perception control to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides a ship communication link establishment method based on perception control, the method comprising:

[0007] Based on the ship's historical wireless carrier communication records, a communication behavior dataset corresponding to different communication areas is extracted. The communication behavior dataset contains multiple communication behavior data, and each communication behavior data corresponds to one communication event.

[0008] By analyzing the density and drastic changes in the distribution of communication behavior data over time within each communication area, a dynamic index of the communication environment for each communication area is obtained.

[0009] Candidate communication regions for constructing communication links are selected based on the dynamic index of the communication environment.

[0010] For each candidate communication region, using the communication behavior data of all ships within the candidate communication region, a perception feature vector reflecting the ship's perception behavior characteristics and a link-building feature vector reflecting the ship's link-building behavior characteristics are constructed respectively.

[0011] Based on the perception feature vector of each ship, the perception contribution of the ship in the candidate communication area is calculated;

[0012] Based on the link establishment feature vector of each ship and the perception contribution, the communication node weight of the ship in the candidate communication area is calculated.

[0013] Based on the weights of all ships' communication nodes, a hierarchical communication network architecture is generated within the candidate communication area.

[0014] Based on the characteristics of different levels in the hierarchical communication network architecture, appropriate wireless carrier communication modes are assigned to ships at different levels to form the final ship communication link establishment scheme.

[0015] Preferably, the communication behavior data includes at least: the start and end time of each communication event, the duration of communication, the identifier of the initiating vessel, the identifier of the target vessel, the channel identifier used during communication, the carrier sensing duration before communication is initiated, the signal strength value when communication is established, and an identifier indicating whether this communication is a re-establishment of connection.

[0016] Preferably, the step of obtaining the communication environment dynamic index for each communication area includes:

[0017] For a given communication area, the communication behavior data within that area are arranged in chronological order.

[0018] At a preset time granularity, the number of communication events occurring in each time unit is counted to form a sequence of communication event counts.

[0019] Calculate the variance of the sequence of communication events as a measure of fluctuation reflecting the degree of change;

[0020] Identify consecutive time periods in the sequence of communication event counts. If the number of communication events in a consecutive time period is consistently higher than the regional historical average, calculate the cumulative duration percentage of all consecutive time periods as a measure of sustained activity intensity.

[0021] The fluctuation amount and the sustained activity are weighted and fused, and the result is normalized and output as the communication environment dynamic index of the communication area.

[0022] Preferably, the step of screening candidate communication regions for constructing a communication link includes:

[0023] The communication environment dynamic index is compared with a preset dynamic threshold.

[0024] Communication regions with a communication environment dynamics index higher than the dynamics threshold are marked as high-dynamic regions;

[0025] Calculate the contribution ratio of the fluctuation in each high-dynamic region to its communication environment dynamics index to obtain the fluctuation contribution rate.

[0026] Several high-dynamic regions with the highest fluctuation contribution were selected as candidate communication regions for the communication link to be constructed.

[0027] Preferably, the steps for constructing the perceptual feature vector include:

[0028] For a single vessel, extract all records of the vessel as the communication initiator from its communication behavior data within the candidate communication area;

[0029] For each record where the communication is initiated, calculate the total time from the start of carrier sensing to the successful establishment of a communication connection, and record it as the single sensing connection establishment time.

[0030] The proportion of successful chain establishment times where the single sensing chain establishment time is less than a preset threshold within a preset statistical period is used as the sensing agility.

[0031] Calculate the average signal strength value at the time of communication establishment in all records of the ship as the initiator of communication, and use it as the average perceived quality;

[0032] The perception feature vector of the ship is composed of the perception agility and the average perception quality.

[0033] Preferably, the step of calculating the ship's sensing contribution within the candidate communication area includes:

[0034] Obtain the perception feature vectors of all ships within the candidate communication area;

[0035] Calculate the mean cosine similarity between the perceived feature vector of the ship and the perceived feature vectors of all other ships in the region;

[0036] The ship's perception agility is multiplied by the mean cosine similarity to obtain the ship's perception behavior consistency factor.

[0037] The average perception quality of the ship is fused with the perception behavior consistency factor to calculate the perception contribution of the ship in the candidate communication area.

[0038] Preferably, the steps for constructing the chain feature vector include:

[0039] For a single vessel, extract all communication records in which the vessel participated, regardless of whether it was the initiator or the target, from its communication behavior data within the candidate communication area;

[0040] The total duration of the communication connection successfully maintained by the ship within a preset statistical period and the total number of times the ship attempted to initiate a communication connection are statistically analyzed. The ratio of the total duration of the successfully maintained communication connection to the total number of times the communication connection was attempted to be initiated is calculated as the connection establishment and maintenance efficiency.

[0041] The proportion of records marked as reconnection in the ship's communication behavior data is used as a measure of the ship's link reconnection tendency.

[0042] The link establishment and maintenance efficiency and the link reconnection tendency constitute the link establishment feature vector of the ship.

[0043] Preferably, the step of calculating the communication node weights of the ship within the candidate communication area includes:

[0044] Obtain the chain-building feature vector and perception contribution of the ship;

[0045] Calculate the product of the chain establishment and maintenance efficiency and the perception contribution to obtain the preliminary weighting factor;

[0046] The link reconnection tendency is processed using a preset compensation function to obtain a stability compensation coefficient;

[0047] The initial weighting factor is multiplied by the stability compensation coefficient, and the output is the communication node weight of the ship in the candidate communication area.

[0048] Preferably, the step of generating a hierarchical communication network architecture within a candidate communication area includes:

[0049] All ships within the candidate communication area are sorted from highest to lowest according to their communication node weights;

[0050] Select ships that rank highly and whose communication node weights are greater than a preset weight threshold to form the core node set;

[0051] The remaining ships that have a history of successful communication with any ship in the core node set constitute the stable node set.

[0052] All ships that do not belong to the core node set or the stable node set are formed into the edge node set;

[0053] The core node set, stable node set, and edge node set together constitute the hierarchical communication network architecture of the candidate communication region.

[0054] Preferably, the step of assigning appropriate wireless carrier communication modes to ships at different levels includes:

[0055] A coordinated communication mode is assigned to ships belonging to the core node set. In this mode, the ships are responsible for listening to the common channel and coordinating communication requests between ships in the stable node set and the edge node set.

[0056] Auxiliary communication mode is assigned to ships belonging to the stable node set. In this mode, ships need to listen to the channel status information broadcast by the core node set ships before initiating communication.

[0057] A free-competition communication mode is assigned to ships belonging to the edge node set, in which ships communicate using a carrier-aware random access mechanism.

[0058] Compared with the prior art, the beneficial effects of the present invention are:

[0059] By analyzing the density and drastic changes in communication behavior data over time, a communication environment dynamics index is calculated, enabling a quantitative assessment of maritime communication conditions. This method transforms the characteristics of communication areas from static physical divisions to dynamic behavioral analysis, allowing for precise responses to the time-varying nature of the real environment from the initial stages of network construction. Using this index to select candidate communication areas proactively avoids regions with overly unstable or sparse communication environments, prioritizing network resources in maritime areas with good communication potential. This lays a crucial foundation for building a stable network architecture and enhances the intelligence and foresight of the networking process.

[0060] A perception feature vector reflecting the ship's perception behavior and a link-building feature vector reflecting its link-building behavior are constructed respectively. Based on these vectors, the perception contribution and communication node weights of nodes are calculated, realizing a multi-dimensional capability assessment and comprehensive weighting of network nodes. This mechanism innovatively decouples the "perception" capability and "relay" capability of nodes before weighted fusion, so that the calculation of node weights not only considers its connectivity performance as a relay node, but also emphasizes its contribution to the network environment perception capability as a perception node. Based on these weights, a hierarchical division is performed to generate a layered communication network architecture. This results in a network with not only good connectivity, but also stronger environmental perception capability and overall robustness, optimizing network resource allocation and improving the network's survivability and service continuity under harsh sea conditions. Attached Figure Description

[0061] Figure 1 This is a schematic diagram illustrating the working principle of the ship communication link establishment method based on perception control described in this invention.

[0062] Figure 2 A flowchart for calculating the dynamic index of the communication environment;

[0063] Figure 3 A flowchart for filtering candidate communication regions;

[0064] Figure 4 A dual-axis composite diagram for dynamic analysis of ship communication areas;

[0065] Figure 5 A regional hierarchical comparison chart of the weights of ship communication nodes. Detailed Implementation

[0066] 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.

[0067] Please see Figure 1 This invention provides a method for establishing ship communication links based on perception control. The method includes: extracting communication behavior datasets corresponding to different communication areas based on historical wireless carrier communication records of ships. These datasets contain multiple communication behavior data points, each corresponding to a single communication event. The density and drastic changes of the communication behavior data over time within each communication area are analyzed to obtain a communication environment dynamic index for each area. Candidate communication areas for establishing communication links are selected based on the communication environment dynamic index. For each candidate communication area, a perception feature vector reflecting the ship's perception behavior characteristics and a link-establishment feature vector reflecting the ship's link-establishment behavior characteristics are constructed using the communication behavior data of all ships within the candidate communication area. Based on the perception feature vector of each ship, the perception contribution of that ship within the candidate communication area is calculated. Based on the link-establishment feature vector and perception contribution of each ship, the communication node weight of that ship within the candidate communication area is calculated. The communication node weights of all ships are sorted and hierarchically divided to generate a hierarchical communication network architecture within the candidate communication area. Based on the characteristics of different levels in the hierarchical communication network architecture, appropriate wireless carrier communication modes are assigned to ships at different levels to form the final ship communication link establishment scheme.

[0068] In one embodiment of the present invention, see [reference] Figure 2Each communication event is recorded, including its start and end times, duration, initiating vessel identifier, target vessel identifier, channel identifier used, carrier sensing duration before communication initiation, signal strength at communication establishment, and whether the communication is a re-establishment of a connection. For a given communication area, the communication behavior data within that area is arranged chronologically. At a preset time granularity, the number of communication events occurring within each time unit is counted, forming a communication event quantity sequence. The variance of this communication event quantity sequence is calculated as a fluctuation measure reflecting the degree of drastic change. Continuous periods within the communication event quantity sequence are identified; if the number of communication events within a continuous period is consistently higher than the historical average for the area, the cumulative duration of all continuous periods is calculated as the sustained activity level reflecting the intensity of change. The fluctuation measure and sustained activity level are weighted and fused, and the result, after normalization, is output as the communication environment dynamic index for that communication area.

[0069] In specific implementations, communication behavior data includes at least the start and end times of each communication event, the duration of communication, the identifier of the initiating vessel, the identifier of the target vessel, the channel identifier used during communication, the carrier sensing duration before communication initiation, the signal strength value at the time of communication establishment, and an identifier indicating whether this communication is a re-establishment of connection. These data constitute the basic elements of the communication behavior dataset. In some embodiments, for a communication area, the communication behavior data within the communication area is arranged chronologically to form a timestamp-sorted record sequence. The number of communication events occurring within each time unit is counted at a preset time granularity. The time unit can be minutes, hours, or days. The statistical results form a communication event quantity sequence. It can be understood that the communication event quantity sequence reflects the temporal distribution of communication activities. The variance of the communication event quantity sequence is calculated as a fluctuation measure reflecting the degree of change. Continuous time periods in the communication event quantity sequence are identified. Within a continuous time period, the number of communication events is consistently higher than the regional historical average, which is calculated based on the historical number of communication events. The cumulative duration percentage of all continuous time periods is counted as the sustained activity level reflecting the intensity.

[0070] In practice, the fluctuation amount and the sustained activity level are weighted and fused. The result is then normalized and output as the communication environment dynamic index of the communication area. The weighting and fusion uses preset weight coefficients, such as α for fluctuation amount and β for sustained activity level. The normalization process maps the weighted sum to a preset range. The formula for calculating the communication environment dynamic index can be expressed as:

[0071] in: This represents an index indicating the dynamic nature of the communication environment. Indicates the fluctuation amount. Indicates sustained activity level. and These are preset weighting coefficients that satisfy... , This represents a normalization function, such as linear normalization to the [0,1] interval. Volatility. It is the variance of the sequence of communication event counts, representing the sustained activity. It is the ratio of the cumulative duration of a continuous period to the total duration.

[0072] In practical implementation, the calculation of the communication environment dynamic index relies on the completeness of communication behavior data. This data is extracted from historical wireless carrier communication records, with each data point corresponding to a single communication event. The construction of the communication event sequence is based on the division of time units. The volatility measures the temporal dispersion of the number of communication events; a higher value indicates more drastic changes. The sustained activity measures the concentration of high-activity periods; a higher value indicates higher communication density. Optionally, continuous time periods are identified using a sliding window algorithm. The window size is related to the time granularity. The average number of communication events within the window is compared to the regional historical average. Windows exceeding the regional historical average are marked as active windows, and consecutive active windows are merged into continuous time periods. The communication environment dynamic index combines volatility and sustained activity; a high value indicates a highly dynamic communication environment, while a low value indicates a relatively stable environment.

[0073] In one embodiment of the present invention, see [reference] Figure 3 The communication environment dynamics index is compared with a preset dynamics threshold. Communication areas with a dynamics index higher than the threshold are marked as high-dynamic areas. The contribution ratio of fluctuation in each high-dynamic area to its communication environment dynamics index is calculated to obtain the fluctuation contribution. Several high-dynamic areas with the highest fluctuation contribution are selected as candidate communication areas for the communication link to be constructed.

[0074] In practical implementation, the communication environment dynamics index is compared with a preset dynamics threshold, which is a preset numerical benchmark used to distinguish the strength of communication environment dynamics. In some embodiments, the dynamics threshold is set based on the statistical distribution of historical communication environment dynamics indices, for example, taking the median or a specific percentile of historical communication environment dynamics indices as the dynamics threshold. Communication areas with communication environment dynamics indices higher than the dynamics threshold are marked as high-dynamic areas. High-dynamic areas represent communication events with a high degree of drastic change or density in their temporal distribution. It can be understood that the screening of high-dynamic areas is a process of identifying areas with complex communication environments that require priority construction of optimized links.

[0075] In practical implementation, the contribution ratio of fluctuation in each high-dynamic region to the dynamic index of the communication environment is calculated to obtain the fluctuation contribution degree. The fluctuation contribution degree reflects the relative influence of fluctuation on the value of the dynamic index of the communication environment. Optionally, the formula for calculating the fluctuation contribution degree is:

[0076] in: Indicates the contribution of volatility. Indicates the fluctuation amount. This represents an index indicating the dynamic nature of the communication environment. Volatility. The Communication Environment Dynamics Index is a numerical value calculated when analyzing the dynamics of the communication environment, reflecting the degree of drastic change. It is the result of combining volatility and sustained activity, after normalization. In some embodiments, volatility contribution... The value range is between 0 and 1, and the fluctuation contribution is... A higher value indicates that the dynamism of this high-dynamic region primarily stems from dramatic fluctuations in the number of communication events rather than a sustained high level of activity. It can be understood that the fluctuation contribution is used to further differentiate the variations within a high-dynamic region.

[0077] In practice, several high-dynamic regions with the highest fluctuation contribution are selected as candidate communication regions for building communication links. The number of regions selected is a preset parameter, such as selecting the top three high-dynamic regions in terms of fluctuation contribution. Optionally, the selection process sorts all high-dynamic regions from highest to lowest according to their fluctuation contribution, and then truncates the top region of the sorted list according to a preset number. The output of this step is a clear list of candidate communication regions. The subsequent construction of sensing feature vectors and link-building feature vectors will be based on the ship communication behavior data within these candidate communication regions.

[0078] In one embodiment of the present invention, for a ship, all records of the ship acting as a communication initiator are extracted from its communication behavior data within a candidate communication area. For each record of acting as a communication initiator, the total duration from carrier sensing to successful establishment of a communication connection is calculated and recorded as the single sensing link establishment time. The proportion of successful link establishment times with a single sensing link establishment time lower than a preset threshold within a preset statistical period is counted as the sensing agility. The average signal strength value at the time of communication establishment is calculated for all records of the ship acting as a communication initiator and is recorded as the average sensing quality. The sensing agility and the average sensing quality constitute the sensing feature vector of the ship. The sensing feature vectors of all ships within the candidate communication area are obtained. The mean cosine similarity between the sensing feature vector of the ship and the sensing feature vectors of all other ships within the area is calculated. The sensing agility of the ship is multiplied by the mean cosine similarity to obtain the sensing behavior consistency factor of the ship. The average sensing quality and the sensing behavior consistency factor of the ship are fused and calculated to output the sensing contribution of the ship within the candidate communication area.

[0079] For a given vessel, all communication records in which the vessel participated, regardless of whether it was the initiator or the target, are extracted from its communication behavior data within a candidate communication area. The total duration of successfully maintained communication connections and the total number of attempts to initiate communication connections are calculated, and the ratio of the total duration of successfully maintained connections to the total number of attempts is taken as the link establishment and maintenance efficiency. The proportion of records marked as re-established connections out of the total number of communication records is taken as the link reconnection tendency. The link establishment and maintenance efficiency and the link reconnection tendency constitute the vessel's link establishment feature vector.

[0080] In practical implementation, for a single vessel, all records of the vessel acting as a communication initiator are extracted from the vessel's communication behavior data within the candidate communication area. Records of the vessel acting as a communication initiator refer to communication behavior data where the initiating vessel's identifier matches the current vessel's identifier. For each record of the vessel acting as a communication initiator, the total duration from carrier sensing to successful establishment of a communication connection is calculated and recorded as the single sensing link establishment time. The single sensing link establishment time is the sum of the carrier sensing time and the time used for the communication connection establishment process. The proportion of successful link establishments with a single sensing link establishment time lower than a preset threshold within a preset statistical period is used as the sensing agility. The preset threshold is a time threshold value used to distinguish between agile and sluggish link establishment processes. The average signal strength value at the time of communication establishment in all records of the vessel acting as a communication initiator is calculated as the average sensing quality. The average sensing quality reflects the channel conditions typically encountered by the vessel when initiating communication. The sensing agility and average sensing quality constitute the vessel's sensing feature vector, which is a two-dimensional vector in the form of (sensing agility, average sensing quality).

[0081] In specific implementation, the perception feature vectors of all ships within the candidate communication area are obtained. The mean cosine similarity between the perception feature vector of a ship and the perception feature vectors of all other ships within the area is calculated. The mean cosine similarity measures the degree of similarity between the ship's perception behavior pattern and the average pattern of other ships within the area. The perception agility of the ship is multiplied by the mean cosine similarity to obtain the ship's perception behavior consistency factor. The average perception quality of the ship and the perception behavior consistency factor are then fused to calculate the ship's perception contribution within the candidate communication area. In some embodiments, the fusion calculation uses a weighted summation or multiplication method; for example, the formula for calculating the perception contribution S is:

[0082] in: Indicates perceived contribution. Represents the mean cosine similarity. This represents perception agility. It can be understood that perception contribution combines the agility of the ship's own perception behavior with the coordination of that behavior with group patterns. Optionally, the calculation of mean cosine similarity requires normalization of the perception feature vectors of all ships to eliminate the influence of dimensions.

[0083] In specific implementation, for a single vessel, all communication records in which the vessel participated, regardless of whether it was the initiator or the target, are extracted from the vessel's communication behavior data within the candidate communication area. All participating communication records refer to records where the initiating vessel's identifier or the target vessel's identifier matches the current vessel's identifier. The total duration of successfully maintained communication connections and the total number of attempts to initiate communication connections are statistically analyzed within a preset statistical period. The total number of attempts is obtained by counting the number of records where the vessel was the initiator. The ratio of the total duration of successfully maintained communication connections to the total number of attempts is calculated as the connection establishment and maintenance efficiency, reflecting the average duration of communication that the vessel can maintain per attempt. The proportion of records marked as re-established connections in the vessel's communication behavior data to the total number of communication records is calculated as the link reconnection tendency, reflecting the stability of the vessel's communication links. The connection establishment and maintenance efficiency and the link reconnection tendency constitute the vessel's connection establishment feature vector, a two-dimensional vector in the form of (connection establishment and maintenance efficiency, link reconnection tendency). In some embodiments, the preset statistical period is coordinated with the time granularity used when calculating the communication environment dynamics index. It is understandable that the link establishment feature vector characterizes the link establishment behavior of ships from two dimensions: communication maintenance capability and link stability.

[0084] In one embodiment of the present invention, the link establishment feature vector and perception contribution of the ship are obtained. The product of the link establishment and maintenance efficiency and the perception contribution is calculated to obtain a preliminary weighting factor. A preset compensation function is used to process the link reconnection tendency to obtain a stability compensation coefficient. The preliminary weighting factor is multiplied by the stability compensation coefficient, and the output is the communication node weight of the ship in the candidate communication area.

[0085] In specific implementation, the ship's link establishment feature vector and perception contribution are obtained. The link establishment feature vector includes two components: link establishment and maintenance efficiency and link reconnection tendency. The perception contribution is a scalar value characterizing the ship's perception behavior's contribution to the area. The product of link establishment and maintenance efficiency and perception contribution is calculated to obtain a preliminary weighting factor, which reflects the ship's comprehensive performance in maintaining communication efficiency and perception contribution. A preset compensation function is used to process the link reconnection tendency to obtain a stability compensation coefficient. The compensation function maps the link reconnection tendency to an adjustment coefficient; a higher link reconnection tendency generally means worse stability, and the mapped stability compensation coefficient value is lower. In some embodiments, the compensation function can be a piecewise linear function or a lookup table mapping. Its design principle is to output a higher stability compensation coefficient value when the link reconnection tendency is low, and a lower stability compensation coefficient value when the link reconnection tendency is high. It can be understood that the role of the stability compensation coefficient is to correct the preliminary weighting factor to reflect the negative impact of frequent link reconnection on node reliability.

[0086] In practice, the stability compensation coefficient can be determined through a preset mapping relationship. For example, a preset mapping table can be used to map the numerical range of link reconnection tendency to specific stability compensation coefficient values, see Table 1.

[0087] Table 1: Stability Compensation Coefficient Mapping Table

[0088] Link reconnection tendency range Stability compensation coefficient value [0.0,0.1) 1.0 [0.1,0.3) 0.8 [0.3,0.5) 0.6 [0.5,0.7) 0.4 [0.7,1.0] 0.2

[0089] Optionally, the compensation function can also be a continuous mathematical function, such as a negatively correlated exponential decay function. Multiplying the initial weighting factor by the stability compensation coefficient outputs the communication node weights of the ship within the candidate communication area. The communication node weights are the single numerical values ​​ultimately used to evaluate the importance of communication nodes within the area. In some embodiments, the formula for calculating the communication node weights is:

[0090] in: Indicates the weight of the communication node. Indicates the initial weighting factor. Represents the stability compensation coefficient. Preliminary weighting factor. It is the product of chain establishment and maintenance efficiency and perceived contribution, and the stability compensation coefficient. This is the output value of the link reconnection tendency after processing by the compensation function. It can be understood that the communication node weight integrates three factors: link establishment efficiency, perceived contribution, and link stability. After calculating the communication node weight for each ship, these weight values ​​will be used for subsequent sorting and hierarchical division.

[0091] See Figure 4 This is a dual-axis composite chart analyzing the dynamics of ship communication areas, a core data chart in the "candidate communication area screening stage." The dynamics index is positively correlated with the combined performance of "fluctuation volume + sustained activity," reflecting the logic that "areas with drastic and dense changes are more dynamic." It can quickly locate highly dynamic areas; these areas have complex communication behaviors and are candidate areas for priority communication link construction. Distinguishing the communication characteristics of different areas provides a differentiated basis for subsequent link construction strategies for different areas. If the dynamics index of a certain area is abnormal, it can be determined that its communication demand is weak, and excessive link construction resources do not need to be invested, improving resource allocation efficiency.

[0092] In one embodiment of the present invention, all ships within the candidate communication area are sorted from highest to lowest according to their communication node weights. Ships ranked higher with communication node weights greater than a preset weight threshold are selected to form a core node set. Ships among the remaining ships that have a historical successful communication record with any ship in the core node set are selected to form a stable node set. All ships that do not belong to either the core node set or the stable node set are selected to form an edge node set. The core node set, stable node set, and edge node set together constitute the hierarchical communication network architecture of the candidate communication area. A coordinated communication mode is assigned to ships belonging to the core node set. In this mode, the ships are responsible for listening to the common channel and coordinating communication requests between ships in the stable node set and the edge node set. An auxiliary communication mode is assigned to ships belonging to the stable node set. In this mode, ships need to listen to channel state information broadcast by ships in the core node set before initiating communication. A free-competition communication mode is assigned to ships belonging to the edge node set. In this mode, ships use a carrier-aware random access mechanism for communication.

[0093] In practice, all ships within the candidate communication area are sorted from highest to lowest according to their communication node weights. The weights are calculated using the aforementioned steps, and the sorting process generates an ordered list of ships. Ships ranked high and with communication node weights greater than a preset weight threshold are selected to form the core node set. The preset weight threshold is a numerical threshold used to define core nodes. "Ranked high" can be defined as the top N ships in the list or ships with weights in the top X percentiles. Ships meeting these two conditions are included in the core node set. Ships among the remaining ships that have a history of successful communication with any ship in the core node set are selected to form the stable node set. The remaining ships refer to all ships in the sorted list that are not part of the core node set. A history of successful communication means that the ship's communication behavior data contains at least one record, with the initiating ship's identifier and the target ship's identifier corresponding to the ship to be judged and a ship in the core node set, respectively. All ships that do not belong to either the core node set or the stable node set are selected to form the edge node set. The core node set, stable node set, and edge node set together constitute the hierarchical communication network architecture of the candidate communication area.

[0094] In some embodiments, the preset weight threshold can be set based on the statistical distribution of the weights of all ship communication nodes, such as taking the average or upper quartile of the weights of all communication nodes as the preset weight threshold. It is understood that ships in the core node set have higher communication node weights, indicating that these ships perform optimally in terms of perception contribution, link establishment efficiency, and link stability. Optionally, when determining whether there are historical successful communication records, the success or failure of the communication record can be determined by whether the communication duration is greater than zero or whether there is a corresponding communication establishment confirmation flag.

[0095] In practical implementation, a coordinated communication mode is assigned to ships belonging to the core node set. In this mode, ships are responsible for monitoring the common channel and coordinating communication requests between ships within the stable node set and the edge node set. Monitoring the common channel refers to continuously or periodically performing carrier sensing on a designated common control channel. An auxiliary communication mode is assigned to ships belonging to the stable node set. In this mode, before initiating communication, ships need to monitor channel status information broadcast by ships in the core node set. This channel status information may include the currently available channel identifier, channel busy / idle status, or suggested communication parameters. A free-competition communication mode is assigned to ships belonging to the edge node set. In this mode, ships use a carrier-aware random access mechanism for communication, such as carrier sense multiple access / collision avoidance mechanisms.

[0096] In some embodiments, ships within the core node set periodically broadcast beacon frames on a common channel to publish coordination information. Ships within the stable node set receive and parse beacon frames to obtain the current network status before initiating communication. Ships within the edge node set independently compete for channel resources following standard carrier-aware random access procedures. It can be understood that the allocation of different communication modes is based on the ship's position in the hierarchical communication network architecture and its expected communication role. The coordinated communication mode assigns network management functions to core node ships, the auxiliary communication mode requires stable node ships to follow the guidance of core nodes, and the free-competition communication mode allows edge node ships to access the network in a distributed manner. Optionally, ships at different levels may use different physical layer parameters or access protocols to implement their assigned modes, ultimately forming a ship communication link establishment scheme.

[0097] See Figure 5This is a regional hierarchical comparison chart of ship communication node weights, corresponding to the "network architecture generation stage." Core nodes have the highest and most stable weights, while edge nodes also have high weights, representing the region with the most concentrated node resources. Core nodes have relatively high weights, but edge nodes have relatively low weights. A pattern of "core node weight > stable node > edge node" is observed across all regions, consistent with the design logic of a hierarchical network. Clearly defining the node hierarchy distribution in each region allows for priority allocation of coordinated communication resources to high-weight areas, improving network efficiency. If the weight of an edge node in a certain region is too low, its communication mode can be adjusted to enhance the communication stability of the edge node.

[0098] 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 process, method, article, or apparatus.

[0099] 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 method for establishing ship communication links based on perception and control, characterized in that, include: Based on the ship's historical wireless carrier communication records, a communication behavior dataset corresponding to different communication areas is extracted. The communication behavior dataset contains multiple communication behavior data, and each communication behavior data corresponds to one communication event. By analyzing the density and drastic changes in the distribution of communication behavior data over time within each communication area, a dynamic index of the communication environment for each communication area is obtained. Candidate communication regions for constructing communication links are selected based on the dynamic index of the communication environment. For each candidate communication region, using the communication behavior data of all ships within the candidate communication region, a perception feature vector reflecting the ship's perception behavior characteristics and a link-building feature vector reflecting the ship's link-building behavior characteristics are constructed respectively. Based on the perception feature vector of each ship, the perception contribution of the ship in the candidate communication area is calculated; Based on the link establishment feature vector of each ship and the perception contribution, the communication node weight of the ship in the candidate communication area is calculated. Based on the weights of all ships' communication nodes, a hierarchical communication network architecture is generated within the candidate communication area. Based on the characteristics of different levels in the hierarchical communication network architecture, appropriate wireless carrier communication modes are assigned to ships at different levels to form the final ship communication link establishment scheme.

2. The ship communication link establishment method based on perception control according to claim 1, characterized in that, The communication behavior data includes at least: the start and end time of each communication event, the duration of communication, the identifier of the initiating vessel, the identifier of the target vessel, the channel identifier used during communication, the carrier sensing duration before communication is initiated, the signal strength value when communication is established, and an identifier indicating whether this communication is a re-establishment of connection.

3. The ship communication link establishment method based on perception control according to claim 1, characterized in that, The step of obtaining the communication environment dynamic index for each communication region includes: For a given communication area, the communication behavior data within that area are arranged in chronological order. At a preset time granularity, the number of communication events occurring in each time unit is counted to form a sequence of communication event counts. Calculate the variance of the sequence of communication events as a measure of fluctuation reflecting the degree of change; Identify consecutive time periods in the sequence of communication event counts. If the number of communication events in a consecutive time period is consistently higher than the regional historical average, calculate the cumulative duration percentage of all consecutive time periods as a measure of sustained activity intensity. The fluctuation amount and the sustained activity are weighted and fused, and the result is normalized and output as the communication environment dynamic index of the communication area.

4. The ship communication link establishment method based on perception control according to claim 1, characterized in that, The step of selecting candidate communication regions for the communication link to be constructed includes: The communication environment dynamic index is compared with a preset dynamic threshold. Communication regions with a communication environment dynamics index higher than the dynamics threshold are marked as high-dynamic regions; Calculate the contribution ratio of the fluctuation in each high-dynamic region to its communication environment dynamics index to obtain the fluctuation contribution rate. Several high-dynamic regions with the highest fluctuation contribution were selected as candidate communication regions for the communication link to be constructed.

5. The ship communication link establishment method based on perception control according to claim 1, characterized in that, The steps for constructing the perceptual feature vector include: For a single vessel, extract all records of the vessel as the communication initiator from its communication behavior data within the candidate communication area; For each record where the communication is initiated, calculate the total time from the start of carrier sensing to the successful establishment of a communication connection, and record it as the single sensing connection establishment time. The proportion of successful chain establishment times where the single sensing chain establishment time is less than a preset threshold within a preset statistical period is used as the sensing agility. Calculate the average signal strength value at the time of communication establishment in all records of the ship as the initiator of communication, and use it as the average perceived quality; The perception feature vector of the ship is composed of the perception agility and the average perception quality.

6. The ship communication link establishment method based on perception control according to claim 1, characterized in that, The step of calculating the ship's sensing contribution within the candidate communication area includes: Obtain the perception feature vectors of all ships within the candidate communication area; Calculate the mean cosine similarity between the perceived feature vector of the ship and the perceived feature vectors of all other ships in the region; The ship's perception agility is multiplied by the mean cosine similarity to obtain the ship's perception behavior consistency factor. The average perception quality of the ship is fused with the perception behavior consistency factor to calculate the perception contribution of the ship in the candidate communication area.

7. The ship communication link establishment method based on perception control according to claim 1, characterized in that, The steps for constructing the chain-building feature vector include: For a single vessel, extract all communication records in which the vessel participated, regardless of whether it was the initiator or the target, from its communication behavior data within the candidate communication area; The total duration of the communication connection successfully maintained by the ship within a preset statistical period and the total number of times the ship attempted to initiate a communication connection are statistically analyzed. The ratio of the total duration of the successfully maintained communication connection to the total number of times the communication connection was attempted to be initiated is calculated as the connection establishment and maintenance efficiency. The proportion of records marked as reconnection in the ship's communication behavior data is used as a measure of the ship's link reconnection tendency. The link establishment and maintenance efficiency and the link reconnection tendency constitute the link establishment feature vector of the ship.

8. The ship communication link establishment method based on perception control according to claim 1, characterized in that, The step of calculating the communication node weights of the ship within the candidate communication area includes: Obtain the chain-building feature vector and perception contribution of the ship; Calculate the product of the chain establishment and maintenance efficiency and the perception contribution to obtain the preliminary weighting factor; The link reconnection tendency is processed using a preset compensation function to obtain a stability compensation coefficient; The initial weighting factor is multiplied by the stability compensation coefficient, and the output is the communication node weight of the ship in the candidate communication area.

9. The ship communication link establishment method based on perception control according to claim 1, characterized in that, The step of generating a hierarchical communication network architecture within a candidate communication area includes: All ships within the candidate communication area are sorted from highest to lowest according to their communication node weights; Select ships that rank highly and whose communication node weights are greater than a preset weight threshold to form the core node set; The remaining ships that have a history of successful communication with any ship in the core node set constitute the stable node set. All ships that do not belong to the core node set or the stable node set are formed into the edge node set; The core node set, stable node set, and edge node set together constitute the hierarchical communication network architecture of the candidate communication region.

10. The ship communication link establishment method based on perception control according to claim 1, characterized in that, The steps of assigning appropriate wireless carrier communication modes to ships at different levels include: A coordinated communication mode is assigned to ships belonging to the core node set. In this mode, the ships are responsible for listening to the common channel and coordinating communication requests between ships in the stable node set and the edge node set. Auxiliary communication mode is assigned to ships belonging to the stable node set. In this mode, ships need to listen to the channel status information broadcast by the core node set ships before initiating communication. A free-competition communication mode is assigned to ships belonging to the edge node set, in which ships communicate using a carrier-aware random access mechanism.

Images