Intelligent decision-making method and system applied to cross-domain communication networking and electronic equipment

Abstract

The invention discloses an intelligent decision-making method and system applied to cross-domain communication networking and electronic equipment, the intelligent decision-making method and system are applied to the cross-domain communication networking comprising a plurality of cross-domain communication nodes and a plurality of heterogeneous communication links, and the method comprises the following steps: for each communication node in the cross-domain communication networking, in the process of executing a data transmission task by the communication node, sending the data to the communication node; inputting the communication task data, the link state data of the cross-domain communication networking at the current moment and the communication node state data into a transmission strategy decision model corresponding to the current moment to obtain target communication decision information of the communication node at the next moment, and adjusting a transmission strategy of the communication node at the next moment based on the target communication decision information, so that the communication node executes a data transmission task at the next moment based on the transmission strategy, taking the next moment as the current moment, and determining a routing strategy with high transmission efficiency and strong real-time performance for cross-domain communication networking. And the overall data transmission efficiency of cross-domain communication networking is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer processing technology, and in particular to an intelligent decision-making method, system and electronic equipment applied to cross-domain communication networking. Background Art

[0002] With the further exploration and development of the ocean by humans, cross-domain communication networking, as a key link in the coordinated control of communication nodes and the interconnection of data information, is of great significance for ensuring and promoting the development of marine resources and the development of the marine economy. However, cross-domain communication networking faces challenges such as multi-domain communication nodes, heterogeneous communication links, dynamic communication scenarios, and complex communication tasks. For example, in cross-surface and underwater communication scenarios, due to the presence of water media, radio communication alone is difficult to meet the communication transmission needs.

[0003] At present, the communication decision-making method of communication networking can determine the routing strategy in a single domain or homogeneous communication network, but it is difficult to meet the cross-domain communication transmission needs of the air domain, surface domain and underwater domain, and it is impossible to determine a routing strategy with high transmission efficiency and strong real-time performance for the cross-domain communication network. Summary of the invention

[0004] The present invention provides an intelligent decision-making method, system and electronic device applied to cross-domain communication networking, so as to determine a routing strategy with high transmission efficiency and strong real-time performance for cross-domain communication networking, and improve the overall data transmission efficiency of the cross-domain communication networking.

[0005] In a first aspect, the present invention provides an intelligent decision-making method applied to a cross-domain communication network, wherein the cross-domain communication network includes multiple communication nodes in an air domain, a surface domain, and an underwater domain, and the communication mode between the communication nodes is a radio communication mode, a laser communication mode, or an underwater acoustic communication mode, including:

[0006] For each communication node in the cross-domain communication network, in the process of the communication node performing the data transmission task, the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node are input into the transmission strategy decision model corresponding to the current moment, so as to obtain the target communication decision information of the communication node at the next moment; wherein the target communication decision information includes: the target link, the target link communication mode and the target link communication rate, the transmission strategy decision model is determined after optimizing the transmission strategy decision model corresponding to the previous moment based on the communication information rights and interests attribute at the previous moment, and the communication information rights and interests attribute is determined by the link communication feedback parameters corresponding to the communication node and the networking feedback parameters corresponding to the cross-domain communication network after the communication node completes the data transmission task based on the historical communication decision information at the historical moment;

[0007] The transmission strategy of the communication node at the next moment is adjusted based on the target communication decision information, so that the communication node performs the data transmission task based on the transmission strategy at the next moment, and takes the next moment as the current moment.

[0008] In a second aspect, an embodiment of the present invention further provides an intelligent decision-making system applied to cross-domain communication networking, wherein the cross-domain communication networking includes multiple communication nodes in an air domain, a surface domain, and an underwater domain, and the communication mode between the communication nodes is a radio communication mode, a laser communication mode, or an underwater acoustic communication mode, and the system includes:

[0009] A target information determination module is used to input the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node into the transmission strategy decision model corresponding to the current moment for each communication node in the cross-domain communication network during the process of the communication node performing the data transmission task, so as to obtain the target communication decision information of the communication node at the next moment; wherein the target communication decision information includes: the target link, the target link communication mode and the target link communication rate, the transmission strategy decision model is determined after optimizing the transmission strategy decision model corresponding to the previous moment based on the communication information rights and interests attribute at the previous moment, and the communication information rights and interests attribute is determined by the link communication feedback parameters corresponding to the communication node and the networking feedback parameters corresponding to the cross-domain communication network after the communication node completes the data transmission task based on the historical communication decision information at the historical moment;

[0010] The target information adjustment module is used to adjust the transmission strategy of the communication node at the next moment based on the target communication decision information, so that the communication node performs the data transmission task based on the transmission strategy at the next moment and takes the next moment as the current moment.

[0011] In a third aspect, an embodiment of the present invention further provides an electronic device, the electronic device comprising:

[0012] one or more processors;

[0013] a storage system for storing one or more programs,

[0014] When one or more programs are executed by one or more processors, the one or more processors implement an intelligent decision-making method applied to cross-domain communication networking as described in any of the embodiments of the present invention.

[0015] The technical solution of the embodiment of the present invention is applied to a cross-domain communication network of multiple communication nodes in at least an air domain, a surface domain, and an underwater domain. The method includes: for each communication node in the cross-domain communication network, in the process of the communication node performing a data transmission task, the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node are input into the transmission strategy decision model corresponding to the current moment, and the target link, the target link communication mode, and the target link communication rate of the communication node at the next moment are obtained; based on the target communication decision information, the transmission strategy of the communication node at the next moment is adjusted, so that the communication node performs the data transmission task based on the transmission strategy at the next moment, and the next moment is used as the current moment. The technical solution provided by the embodiment of the present invention, for such a complex communication scenario of the cross-domain communication network, can determine a routing strategy with high transmission efficiency and strong real-time performance according to the task attributes of the transmission task, the real-time link information in the cross-domain communication network, and the real-time node information, thereby improving the overall data transmission efficiency of the cross-domain communication network.

[0016] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 This is a flow chart of an intelligent decision-making method applied to cross-domain communication networking provided according to Embodiment 1 of the present invention;

[0019] Figure 2 is a schematic diagram of a link communication mode between communication nodes corresponding to different spatial domains according to an embodiment of the present invention;

[0020] Figure 3 This is a flow chart of an intelligent decision-making method applied to cross-domain communication networking provided according to Embodiment 2 of the present invention;

[0021] Figure 4 is a schematic diagram of various links associated with the communication nodes involved in the second embodiment of the present invention;

[0022] Figure 5 Schematic diagram of the architecture of an intelligent decision-making method applied to cross-domain communication networking provided in accordance with Embodiment 3 of the present invention;

[0023] Figure 6 A schematic diagram of the structure of an intelligent decision-making system applied to cross-domain communication networking provided by an embodiment of the present invention;

[0024] Figure 7 The present invention is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0027] Before introducing this solution, the application scenario of this solution can be introduced first. This embodiment can be applied to any situation where it is necessary to determine the communication strategy of each communication node in the cross-domain communication network. With the further exploration and development of the ocean by mankind, cross-domain communication networking, as a key link in the collaborative control of communication nodes and the networking and intercommunication of data information, is of great significance for ensuring and promoting the development of marine resources and the development of the marine economy. The so-called cross-domain communication networking refers to a communication network composed of communication nodes in different spatial domains. The cross-domain communication network includes at least multiple communication nodes in the air domain, the surface domain, and the underwater domain. Among them, the communication node refers to the connection point in the cross-domain communication network, which can be a redistribution point or a communication endpoint, such as a server, a computer, a drone, a robot, etc.

[0028] More specifically, the multiple communication nodes include at least one air domain communication node, at least one surface domain communication node, and at least one underwater domain communication node. That is, it can be understood that the cross-domain communication network includes multiple air domain communication nodes corresponding to the air domain (for example, drones flying in mid-air), multiple surface domain communication nodes working in the surface domain (for example, servers configured on ships), and multiple underwater domain communication nodes working in the underwater domain (for example, servers configured underwater, underwater robots, anchor points in the sea, etc.). The communication method between the communication nodes is a radio communication method, a laser communication method, or an underwater acoustic communication method.

[0029] However, cross-domain communication networking faces challenges such as multi-domain communication nodes, heterogeneous communication links, dynamic communication scenarios, and complex communication tasks. For communication networking between multi-domain communication nodes in the air, on the surface, and underwater, the traditional homogeneous single communication method is difficult to meet the needs of cross-domain communication networking, which is mainly reflected in: 1) Radio communication uses electromagnetic wave transmission strategy to achieve high coverage and high transmission rate communication in the air, but electromagnetic waves attenuate greatly in water and are difficult to meet the requirements of long-distance communication transmission across the surface or underwater; 2) Underwater acoustic communication uses sound wave transmission, and the transmission loss of water waves in water is relatively small. It is the main means of underwater wireless long-distance communication transmission, but the communication rate is very low; 3) Laser communication uses laser as a carrier for data transmission, which is suitable for cross-surface communication.

[0030] Although traditional communication networking methods and systems can better solve the problems of single-domain or homogeneous communication networks, it is difficult to meet the cross-domain communication transmission needs in the air, on the surface, and underwater. For example, long-distance communication transmission across the surface and underwater cannot be achieved using only radio communication, and communication transmission in the air domain cannot be achieved using only hydroacoustic communication. Therefore, it is necessary to build a cross-domain communication networking solution for air-surface-underwater cross-domain communication, which is a hybrid heterogeneous cross-domain communication of radio communication, hydroacoustic communication, and laser communication. In response to the above problems, this application integrates cross-domain communication nodes in the air, on the surface, and underwater, effectively integrates and gives play to the complementary advantages of the communication capabilities of three heterogeneous communication links of radio, hydroacoustic, and laser communication, and builds a networking intelligent decision-making model for cross-domain communication scenarios and data transmission tasks, realizing cross-domain communication networking intelligent decision-making between underwater, surface, and air nodes, ensuring the resilience and interconnection of air domain communication networking and the robust transmission of information, thereby supporting the application needs of ocean monitoring, ocean rescue, etc.

[0031] Embodiment 1

[0032] Figure 1 An intelligent decision-making method applied to cross-domain communication networking is provided in Example 1 of the present invention.

[0033] In this embodiment, the cross-domain communication website includes multiple communication nodes, and the processing method for each communication node is consistent. Here, any one of the communication nodes is taken as an example for description.

[0034] like Figure 1 As shown, the intelligent decision-making method applied to cross-domain communication networking includes:

[0035] S110. For each communication node in the cross-domain communication network, when the communication node executes a data transmission task, the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node are input into the transmission strategy decision model corresponding to the current moment to obtain the target communication decision information of the communication node at the next moment.

[0036] Among them, the data transmission task refers to the process of transmitting data between the data source and the data sink through one or more links in accordance with appropriate procedures. This process is also called data communication, and its purpose is to realize the transmission and exchange of information between points. The communication task data is the characteristic description information corresponding to the data transmission task. For example, the communication task data may include information such as the task name, task type, and task priority. Whenever a data transmission task is generated, the communication task data corresponding to the data transmission task is determined. The link status data includes the link start node, the link end node, whether the link is invalid, and the link rate. The communication node status data mainly includes node type (air domain communication node, surface domain communication node or underwater domain communication node), node function (send, receive, forward), node working mode (send frequency, operating power), node physical properties (location information, longitude, latitude, altitude) and other information.

[0037] Among them, the target communication decision information refers to the information content (i.e., link, link communication mode, and link communication rate) on which the communication node performs the data transmission task at the next moment for a certain communication node. More specifically, the target communication decision information includes: target link, target link communication mode, and target link communication rate. A link refers to a communication line from one communication node to an adjacent communication node in data communication, and there are no other switching nodes between the two communication nodes. For a certain communication node, the adjacent communication node that is directly connected to it for communication is determined, that is, the link corresponding to it is determined. Link communication modes include radio communication mode, laser communication mode, or underwater acoustic communication mode. Link communication rate refers to the speed at which data is transmitted through a link in a computer network, also known as data rate or bit rate. This rate is usually measured in bits per second (bps), which represents the unit time (usually one second).

[0038] In this embodiment, the transmission strategy decision model is a reinforcement learning model that processes and analyzes the communication task data, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node, and outputs the target communication decision information. In the specific application process, the transmission strategy decision model can be updated in real time, that is, the model parameters in the transmission strategy decision model can be optimized. For example, the model optimization method can include adjusting the number of hidden layer neurons, adjusting the learning rate, and adjusting the discount factor and other parameters to improve the adaptability and learning efficiency of the model, so as to achieve the purpose of model updating, so that the target communication decision information determined by the real-time updated transmission strategy decision model has stronger adaptability and higher accuracy.

[0039] The transmission strategy decision model is determined after optimizing the transmission strategy decision model corresponding to the previous moment based on the communication information rights and interests attributes of the previous moment. The communication information rights and interests attributes are determined by the link communication feedback parameters corresponding to the communication node after the communication node completes the data transmission task based on the historical communication decision information at the historical moment and the networking feedback parameters corresponding to the cross-domain communication network.

[0040] Among them, the communication information rights and interests attribute can be understood as the reward information determined by the link communication feedback parameters and the networking feedback parameters after the communication node performs the data transmission task based on the target communication decision information at the previous moment. The historical moment is the moment before the current moment, for example, the historical moment is the moment before the current moment. The link communication feedback parameters may include link transmission delay information, link packet loss rate, and target link rate corresponding to each link associated with the communication node. The networking feedback parameters include load balancing.

[0041] In this embodiment, the communication information rights and interests attributes can be determined based on the link communication feedback parameters corresponding to the communication node after the data transmission task is completed according to the historical communication decision information and the networking feedback parameters corresponding to the cross-domain communication network. Then, the model parameters of the transmission strategy decision model are optimized according to the communication information rights and interests attributes, thereby realizing the optimization process of the transmission strategy decision model.

[0042] In this embodiment, when a communication node starts to perform a data transmission task, an initial transmission strategy decision model can be obtained, and the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node can be obtained. Then, these data can be input into the initial transmission strategy decision model to obtain the target link, target link communication mode and target link communication rate of the communication node at the next moment. Subsequently, after the communication node completes a one-hop data transmission task based on the target link, target link communication mode and target link communication rate, the link communication feedback parameter corresponding to the communication node and the networking feedback parameter corresponding to the cross-domain communication network are obtained, and the communication information rights and interests attributes are determined according to the link communication feedback parameter and the networking feedback parameter, and then, the model parameters of the transmission strategy decision model are optimized according to the communication information rights and interests attributes to obtain the first transmission strategy decision model. The first transmission strategy decision model can be saved so that the first transmission strategy decision model can be called at the next control moment.

[0043] On this basis, the technical solution provided by this implementation is further described with the second control moment as the current moment. At the second control moment, the transmission strategy decision model corresponding to the current moment is the first transmission strategy decision model that completes the model optimization at the first control moment, and obtains the link status data and communication node status data of the cross-domain communication network at the current moment (ie, the second control moment). Thus, these link status data, communication node status data, and communication task data are input into the first transmission strategy decision model, and the first transmission strategy decision model analyzes and processes these data and outputs the target communication decision information at the next moment (ie, the third control moment). Based on the same communication information determination method, the target communication decision information of the communication nodes corresponding to the third control moment, the fourth control moment, ..., and the Nth control moment at the next moment can be determined.

[0044] More specifically, the transmission strategy decision model may include an action value mapping layer and an action decision layer, wherein the action value mapping layer is used to determine the value parameter corresponding to at least one preset action through a value function based on the communication task data, the link state data of the cross-domain communication network at the current moment (i.e., the second control moment), and the communication node state data as state information. The action decision layer is used to determine the target communication decision information to be finally executed based on each value parameter.

[0045] Optionally, the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node are input into the transmission strategy decision model corresponding to the current moment, and the specific implementation method of obtaining the target communication decision information of the communication node at the next moment may include:

[0046] S1101. Input the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node into the action value mapping layer to obtain the value parameters corresponding to each preset execution action in the preset execution action set.

[0047] The preset execution action includes link information, a communication mode corresponding to the link, and a preset communication rate range corresponding to the communication mode.

[0048] In this embodiment, a preset execution action set can be pre-constructed, and the construction method of the preset execution action set includes: constructing a preset execution action set based on at least one link, at least one communication mode, and at least one preset communication rate range corresponding to each communication mode. That is, it can be understood that for a certain communication node, the communication node and each communication node except itself can determine a link, and the communication mode corresponding to the link can be determined according to the node type corresponding to the communication node and the node type corresponding to the communication node that forms a link with it. More specifically, for a schematic diagram of the link communication mode between communication nodes corresponding to different spatial domains, see Figure 2 ,like Figure 2 As shown, the link between the air domain communication node and the air domain communication node corresponds to the radio communication mode; the link between the air domain communication node and the surface domain communication node corresponds to the radio communication mode; the link between the air domain communication node and the underwater domain communication node corresponds to the laser communication mode; the link between the surface domain communication node and the surface domain communication node corresponds to the radio communication mode; the link between the surface domain communication node and the underwater domain communication node corresponds to the hydroacoustic communication mode; the link between the underwater domain communication node and the underwater domain communication node corresponds to the hydroacoustic communication mode. Since different communication modes correspond to different communication rate ranges, for each communication mode, multiple preset communication rate ranges corresponding thereto can be predetermined. Based on this, according to at least one link, at least one communication mode and at least one preset communication rate range corresponding to each communication mode, multiple combination strategies can be determined, and each combination strategy can be regarded as a preset execution action. Based on this, a preset execution action set including a large number of preset execution actions can be obtained.

[0049] Exemplarily, a communication node 1 is an air domain communication node, and another communication node 2 is an underwater domain communication node. Communication node 1 and communication node 2 correspond to link A, and the communication mode corresponding to link A is laser communication mode, which corresponds to 5 preset communication rate ranges, namely: first level communication rate, second level communication rate, third level communication rate, fourth level communication rate, and fifth level communication rate. In the above situation, the preset execution action set may at least include: first preset action {link A, laser communication mode, first level communication rate}, second preset action {link A, laser communication mode, second level communication rate}, third preset action {link A, laser communication mode, third level communication rate}, fourth preset action {link A, laser communication mode, fourth level communication rate}, and fifth preset action {link A, laser communication mode, fifth level communication rate}. A large number of communication nodes are included in the cross-domain communication network, and a large number of preset execution actions can be constructed based on the same method.

[0050] In this embodiment, the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node can be input into the action value mapping layer, and the action value mapping layer can output the value parameters corresponding to each preset execution action.

[0051] Based on the above examples, the output of the action value mapping layer can be represented in the form of a vector, that is, {Q1, Q2, Q3, Q4, Q5, ..., QN}, where Q1 represents the value parameter corresponding to the first preset action, Q2 represents the value parameter corresponding to the second preset action, Q3 represents the value parameter corresponding to the third preset action, Q4 represents the value parameter corresponding to the fourth preset action, Q5 represents the value parameter corresponding to the fifth preset action, and QN represents the value parameter corresponding to the nth preset action.

[0052] S1102. Input each value parameter into the action decision layer to obtain a target value parameter that meets preset conditions.

[0053] Among them, the preset condition is to determine the maximum value parameter value as the target value parameter.

[0054] In this embodiment, the value parameters corresponding to the first preset action, the second preset action, the third preset action, the fourth preset action and the fifth preset action can be input into the action decision layer, and the action decision layer outputs the target value parameter, which is the maximum value parameter value.

[0055] S1103. Determine the target preset execution action corresponding to the target value parameter as the target link, target link communication mode, and target link communication rate of the communication node at the next moment.

[0056] Based on the above example, the value parameter Q5 has the largest value, and Q5 is used as the target value parameter. The fifth preset action is the target preset execution action, that is, the fifth preset action (i.e. link A, laser communication mode, and fifth level communication rate) is determined as the target link, target link communication mode, and target link communication rate of the communication node at the next moment.

[0057] S120. Adjust the transmission strategy of the communication node at the next moment based on the target communication decision information, so that the communication node performs the data transmission task based on the transmission strategy at the next moment, and takes the next moment as the current moment.

[0058] In this embodiment, based on the target communication decision information, the transmission strategy (i.e., link information, link communication mode, and link communication rate) of the communication node at the next moment can be adjusted according to the target communication decision information, so that the communication node can perform the data transmission task based on these target links, target link communication modes, and target link communication rates at the next moment. Subsequently, the next moment is taken as the current moment, and the above steps S110-S120 are repeatedly executed.

[0059] The technical solution of the embodiment of the present invention is applied to a cross-domain communication network of multiple communication nodes in at least an air domain, a surface domain, and an underwater domain. The method includes: for each communication node in the cross-domain communication network, in the process of the communication node performing a data transmission task, the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node are input into the transmission strategy decision model corresponding to the current moment, and the target link, the target link communication mode, and the target link communication rate of the communication node at the next moment are obtained; based on the target communication decision information, the transmission strategy of the communication node at the next moment is adjusted, so that the communication node performs the data transmission task based on the transmission strategy at the next moment, and the next moment is used as the current moment. The technical solution provided by the embodiment of the present invention, for such a complex communication scenario of the cross-domain communication network, can determine a routing strategy with high transmission efficiency and strong real-time performance according to the task attributes of the transmission task, the real-time link information in the cross-domain communication network, and the real-time node information, thereby improving the overall data transmission efficiency of the cross-domain communication network.

[0060] Embodiment 2

[0061] Figure 3 It is a flowchart of an intelligent decision-making method applied to cross-domain communication networking provided in Embodiment 2 of the present invention. Based on the above-mentioned embodiment, S110 is described in detail, wherein the technical terms that are the same as or corresponding to the above-mentioned embodiment are not repeated here.

[0062] like Figure 3 As shown, the method includes:

[0063] S210. After the communication node completes the data transmission task based on the historical communication decision information at the historical moment, obtain the link communication feedback parameter corresponding to the communication node and the networking feedback parameter corresponding to the cross-domain communication networking.

[0064] The historical communication decision information is the target communication decision information corresponding to the communication node at the last moment. The historical communication decision information may specifically include historical links, historical link communication modes, and historical link communication rates. Link communication feedback parameters include target transmission delay information, target packet loss rate, and target link rates corresponding to each link associated with the communication node. For example, the schematic diagram of each link associated with the communication node is shown in Figure 4 ,like Figure 4 As shown in the figure, there are four nodes, namely, node 0, node 1, node 2 and node 3. Among them, the black-filled nodes 0 and 2 represent air domain communication nodes, the gray-filled node 3 represents the surface domain communication node, and the white-filled node 1 represents the underwater domain communication node. Figure 4 In the example, node 2 represents a communication node for which communication decision information needs to be determined, and other communication nodes associated with the communication node include node 0, node 1, and node 3, so that the links associated with the communication node (i.e., node 2) include: link A corresponding to node 2 and node 0, link B corresponding to node 2 and node 1, and link C corresponding to node 2 and node 3. The target link rate corresponding to link A can be represented by V1, the target link rate corresponding to link B can be represented by V2, and the target link rate corresponding to link C can be represented by V3.

[0065] Among them, the networking feedback parameters include the target load balancing degree, and the target load balancing degree of the network can be evaluated by calculating the sum of the standard deviations of the used bandwidths of all links in the cross-domain communication networking.

[0066] In this embodiment, after the communication node completes the data transmission task based on the historical communication decision information at the historical moment, the network status of the cross-domain communication network has changed. At this time, the networking feedback parameters corresponding to the cross-domain communication network and the link communication feedback parameters corresponding to the communication node can be extracted.

[0067] S220: Determine a target rate symmetry parameter based on each target link rate.

[0068] Among them, considering the asymmetric links in cross-domain communication, for example, the transmission rate of wireless communication links is usually higher than that of underwater acoustic communication links, the rate symmetry parameter can be determined according to the asymmetric differences of different links of nodes. The so-called target rate symmetry parameter refers to the rate symmetry parameter corresponding to a specific communication node.

[0069] In this embodiment, a specific implementation manner of determining the target rate symmetry parameter based on each target link rate may include: determining the target rate symmetry parameter based on a ratio between each target link rate.

[0070] Based on the above examples, the ratio of the target link rate V1 corresponding to link A to the target link rate V2 corresponding to link B can be expressed as S1=V1 / V2; the ratio of the target link rate V2 corresponding to link B to the target link rate V3 corresponding to link C can be expressed as S2=V2 / V3; the ratio of the target link rate V1 corresponding to link A to the target link rate V3 corresponding to link C can be expressed as S3=V1 / V3; and furthermore, the average of S1, S2 and S3 can be determined as the target rate symmetry parameter.

[0071] S230: Determine the communication information rights and interests attribute based on the target load balancing degree, the target transmission delay information, the target packet loss rate, and the target rate symmetry parameter.

[0072] In this embodiment, the reward values ​​corresponding to the target load balancing degree, the target transmission delay information, the target packet loss rate and the target rate symmetry parameter can be determined respectively, and then the communication information rights and interests attributes can be determined according to each reward value.

[0073] Optionally, based on the target load balancing degree, the target transmission delay information, the target packet loss rate, and the target rate symmetry parameter, the specific implementation steps of determining the communication information equity attribute may include:

[0074] (1) Determine a first equity attribute based on a mapping relationship between a target load balancing degree and a first reward value.

[0075] The first reward value mapping relationship includes a mapping relationship between a load balancing degree and a first preset reward value.

[0076] In this embodiment, the network load balancing is evaluated by calculating the sum of the standard deviations of the used bandwidth of all links in the cross-domain communication network, and a positively correlated reward mechanism is designed accordingly. Specifically, the first reward value mapping relationship can be understood as follows: the higher the load balancing degree, that is, the larger the sum of the standard deviations, the more uniform the load distribution of the network, and the larger the corresponding reward value. On the contrary, the smaller the reward value, the network load is balanced. Based on this, the first equity attribute can be obtained according to the mapping relationship between the target load balancing degree and the first reward value.

[0077] (2) Determine the second equity attribute based on the mapping relationship between the target transmission delay information and the second reward value.

[0078] The second reward value mapping relationship includes a mapping relationship between a transmission delay and a second preset reward value.

[0079] In this embodiment, a reward mechanism is designed that is inversely proportional to the transmission delay. The second reward value mapping relationship can be understood as follows: the higher the transmission delay, the lower the corresponding reward value; conversely, the lower the transmission delay, the higher the reward value obtained, thereby seeking a low-latency communication transmission path. Based on this, the second equity attribute can be obtained according to the mapping relationship between the target load balancing degree and the second reward value.

[0080] (3) Determine the third equity attribute based on the mapping relationship between the target packet loss rate and the third reward value.

[0081] The third reward value mapping relationship includes a mapping relationship between a packet loss rate and a third preset reward value.

[0082] In this embodiment, packet loss events have a serious negative impact on the quality of communication transmission. When a packet loss event is detected, the corresponding packet loss negative reward mechanism will be triggered. The severity of the packet loss event will be weighted and punished according to the priority and importance of the data packet or task. The third reward value mapping can be understood as follows: if a critical node loses important data, a larger negative reward will be given; and for ordinary data packet loss at non-critical nodes, although a negative reward will also be given, its value will be relatively small, thereby incentivizing network nodes to avoid packet loss, especially the loss of high priority or important data. Based on this, the third equity attribute can be obtained according to the mapping relationship between the target packet loss rate and the third reward value.

[0083] (4) Determine the fourth equity attribute based on the mapping relationship between the target rate symmetry parameter and the fourth reward value.

[0084] The fourth reward value mapping relationship includes a mapping relationship between a rate symmetry parameter and a fourth preset reward value.

[0085] In this embodiment, taking into account the asymmetric links existing in cross-domain communications, such as wireless links, whose transmission rate is usually higher than that of underwater acoustic links, a reward mechanism is designed according to the asymmetric differences of different links of nodes. That is, the fourth reward value mapping relationship can be understood as follows: when the asymmetric difference is close to 1, it indicates that the rates of the uplink and downlink links are relatively balanced, and a higher reward will be obtained at this time; and when the difference deviates significantly from 1, the reward value will be reduced, so as to avoid extreme differences in the rates of the uplink and downlink links. Based on this, the fourth equity attribute can be obtained according to the target rate symmetry parameter and the fourth reward value mapping relationship:

[0086] (5) Determine the rights and interests attribute of the communication information based on the first rights and interests attribute, the second rights and interests attribute, the third rights and interests attribute, and the fourth rights and interests attribute.

[0087] In this embodiment, the communication information equity attribute may be obtained by performing a weighted sum operation on the first equity attribute, the second equity attribute, the third equity attribute, and the fourth equity attribute.

[0088] S240: Based on the communication information rights and interests attribute, optimize the model parameters of the transmission strategy decision model corresponding to the previous moment to obtain the transmission strategy decision model corresponding to the current moment.

[0089] In this embodiment, the model parameters of the transmission strategy decision model may be optimized once every preset control moment, or the model parameters of the transmission strategy decision model may be optimized once at each control moment. The specific implementation method of optimizing the model parameters of the transmission strategy decision model may be: using the communication information equity attribute as the reward value in the reinforcement learning technology, calculating the target Q value through the Bellman equation and the reward value, and updating the model parameters in the transmission strategy decision model by minimizing the difference between the predicted Q value (i.e., the target communication decision information output by the transmission strategy decision model at the previous moment) and the target Q value.

[0090] In this embodiment, the communication information rights and interests attribute can be determined by comprehensively adopting rewards for each time step (for example, calculating and assigning rewards such as transmission delay in real time for each time step), time segment rewards (for example, performing segment cumulative rewards at certain key time nodes or after multiple time steps), event-triggered rewards (for example, negative rewards triggered by events such as packet loss), and sparse rewards (scoring the overall network quality assessment). At the same time, the reward is composed of multiple sub-rewards after regularization (mainly transmission delay sub-rewards, network load balancing sub-rewards, packet loss event sub-rewards, and link asymmetry difference sub-rewards, etc.), where each sub-reward function selects and designs corresponding linear functions, nonlinear functions, or multi-parameter functions according to the specific situation.

[0091] S250. For each communication node in the cross-domain communication network, when the communication node performs a data transmission task, the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node are input into the transmission strategy decision model corresponding to the current moment to obtain the target link, target link communication mode and target link communication rate of the communication node at the next moment.

[0092] The communication task data also includes task category information, and the task category information is a data transmission communication task, a control information communication task, a status information communication task category, or a fault alarm information communication task.

[0093] In this embodiment, according to the data flow carried in the data transmission task, the task category information corresponding to the data transmission task can be determined. For example, the data flow is classified into raw data, control information, status information and fault alarm information according to the data type. Each type of data can be further subdivided, for example, raw data can be divided into text, voice, image and video, and control information, status information and fault alarm information can be divided into system level, device level, board level and unit level. Based on supervised machine learning, a network such as a convolutional neural network (CNN) is used to extract features, and a classification model is constructed by a classifier such as an extreme gradient boosting tree (XGBoost) to achieve the task category information corresponding to the data transmission task according to the data flow of the data transmission task. Different task categories correspond to different communication service priority levels. For example, queue-based priority management can be used to manage multiple data transmission tasks of different task categories.

[0094] S260: Determine target communication rate information based on the task category information in the communication task data and a preset mapping relationship between the task category and the rate, so as to update the target link communication rate based on the target communication rate information.

[0095] In this embodiment, a preset mapping relationship between task categories and transmission rates can be predefined. In particular, for different link transmission modes, three preset mapping relationships between task categories and rates can be predefined respectively. Specifically, for the radio communication mode, the first wireless transmission rate corresponding to the data transmission communication task, the second wireless transmission rate corresponding to the control information communication task, the third wireless transmission rate corresponding to the status information communication task category, and the fourth wireless transmission rate corresponding to the fault alarm information communication task can be predetermined; for the laser communication mode, the first laser transmission rate corresponding to the data transmission communication task, the second laser transmission rate corresponding to the control information communication task, the third laser transmission rate corresponding to the status information communication task category, and the fourth laser transmission rate corresponding to the fault alarm information communication task can be predetermined; for the laser electrical communication mode, the first laser transmission rate corresponding to the data transmission communication task, the second laser transmission rate corresponding to the control information communication task, the third laser transmission rate corresponding to the status information communication task category, and the fourth laser transmission rate corresponding to the fault alarm information communication task can be predetermined. The invention relates to a method for determining a first underwater acoustic transmission rate corresponding to a data transmission communication task, a second underwater acoustic transmission rate corresponding to a control information communication task, a third underwater acoustic transmission rate corresponding to a status information communication task category, and a fourth underwater acoustic transmission rate corresponding to a fault alarm information communication task. For the underwater acoustic communication method, the first underwater acoustic transmission rate corresponding to the data transmission communication task, the second underwater acoustic transmission rate corresponding to the control information communication task, the third underwater acoustic transmission rate corresponding to the status information communication task category, and the fourth underwater acoustic transmission rate corresponding to the fault alarm information communication task can be predetermined. For the underwater acoustic-electrical communication method, the first underwater acoustic transmission rate corresponding to the data transmission communication task, the second underwater acoustic transmission rate corresponding to the control information communication task, the third underwater acoustic transmission rate corresponding to the status information communication task category, and the fourth underwater acoustic transmission rate corresponding to the fault alarm information communication task can be predetermined.

[0096] On the basis of the above-mentioned embodiments, the target communication rate information can be determined according to the task category information, the target communication mode and each preset mapping relationship in the communication task data. Furthermore, the target link communication rate can be updated according to the target communication rate information to determine the communication rate that the final communication node is about to adopt. In this embodiment, by mapping the graded communication link rate with the classified communication task type, service-rate perception and matching are achieved, so that different communication rates are allocated to different services as much as possible under limited resources, and priority is given to ensuring the reliable transmission of important cross-domain communication services, such as high real-time or high reliability services, and adaptively switching the transmission rate, including increasing or decreasing the transmission rate.

[0097] S270. Adjust the transmission strategy of the communication node at the next moment based on the updated target communication decision information, so that the communication node performs the data transmission task based on the transmission strategy at the next moment, and takes the next moment as the current moment.

[0098] The technical solution of the embodiment of the present invention is to obtain the link communication feedback parameters corresponding to the communication node and the networking feedback parameters corresponding to the cross-domain communication network after the communication node completes the data transmission task based on the historical communication decision information at the historical moment, and then determine the target rate symmetry parameters based on the target link rate and the historical transmission rate corresponding to each link associated with the communication node, and determine the communication information equity attributes based on the target load balancing degree, the target transmission delay information, the target packet loss rate and the target rate symmetry parameters, and then, based on the communication information equity attributes, optimize the model parameters of the transmission strategy decision model corresponding to the previous moment to obtain the transmission strategy decision model corresponding to the current moment, and by continuously optimizing the transmission strategy decision model parameters, the generalization ability and adaptability of the model are improved, and the overall data transmission efficiency of the cross-domain communication network is further improved. In addition, before adjusting the transmission strategy of the communication node at the next moment based on the target communication decision information, the target communication rate information can also be determined based on the task category information in the communication task data and the preset mapping relationship between the task category and the rate, so as to update the target link communication rate based on the target communication rate information, and achieve service-rate perception and matching by mapping the graded communication link rates with the classified communication task types, so as to allocate different communication rates to different services as much as possible under limited resources, and give priority to ensuring the reliable transmission of important cross-domain communication services, such as high real-time or high reliability services, and adaptively switch the transmission rate.

[0099] Embodiment 3

[0100] Next, a specific example is used to illustrate the specific implementation process of the intelligent decision-making method applied to cross-domain communication networking. Figure 5 This is a schematic diagram of the architecture of an intelligent decision-making method applied to cross-domain communication networking. Figure 5 As shown, the intelligent decision-making method applied to cross-domain communication networking mainly involves a cross-domain communication environment, an intelligent control layer and an intelligent decision-making layer.

[0101] Cross-domain communication network environment: mainly includes cross-domain communication nodes, cross-domain communication links and intelligent decision-making computing terminals. Among them, cross-domain communication nodes include air domain communication nodes, surface domain communication nodes and underwater domain communication nodes; cross-domain communication links include radio communication links, underwater acoustic communication links and laser communication links; intelligent decision-making computing terminals include central processing units (CPU), graphics processing units (GPU) and neural network processors (NPU).

[0102] Intelligent control layer: From bottom to top, access and extract cross-domain communication network data through the data interaction southbound interface, and then pre-process the data, and then input the data information into the intelligent decision model of the intelligent decision layer through the data interaction northbound interface; from top to bottom, the strategy (i.e., target communication decision information) output by the intelligent decision model of the intelligent decision layer is deployed and applied to the intelligent decision computing end in the cross-domain communication network environment through the intelligent strategy configuration module.

[0103] Intelligent decision-making layer: The cross-domain communication networking intelligent decision-making layer is the core layer of the cross-domain communication networking intelligent decision-making system, which mainly includes: integrated resource management intelligent decision-making model, dynamic routing selection intelligent decision-making model, and adaptive rate switching intelligent decision-making model.

[0104] More specifically, the integrated resource management intelligent decision-making model: for different communication nodes in the air, surface and underwater domains, a cross-domain communication network resource management access controller is built to monitor, acquire and analyze information such as resource status in real time. At the same time, the prediction algorithm based on supervised learning is improved, and the traffic conditions of cross-domain communication nodes, links, networks and tasks are intelligently predicted according to the actual needs of various communication platforms and tasks, etc., to optimize the integrated resource management performance, thereby improving the resource utilization and transmission efficiency of the cross-domain communication network.

[0105] The main functions of the integrated resource management intelligent decision-making model include:

[0106] Node and link status management: Node status mainly includes node type (air domain, surface, underwater), node function (send, receive, forward), node working mode (send frequency, operating power), node physical properties (location information, longitude, latitude, altitude); link status mainly includes link start node, link end node, link failure and link rate (can be obtained, the speed of data transmission) and other states. Cross-domain communication network topology management: The cross-domain communication network system is a dynamic network that evolves over time, and its network topology also changes over time. In order to reflect the evolution of the cross-domain communication network topology over time, the time domain is sliced ​​and a series of continuous static network topologies are used to represent the dynamic topology. In a very short period of time, the network can be regarded as a static topology network at that moment. Based on the cross-domain communication network topology model, the cross-domain communication network topology management performs real-time topology structure perception and adjustment according to the dynamic changes of the topology structure in different time domain slices, thereby realizing the detection and optimization of the cross-domain communication network topology structure. Node and task priority management: Since there are a large number of different communication nodes in the cross-domain communication network, and the information transmitted by each communication node is also different, it is necessary to define the priority of the communication node and the priority of the communication task, and perform queue priority management on the priorities of different communication nodes and different communication tasks. Among them, the priority of cross-domain communication nodes, when multiple nodes use the same communication link to perform reception and forwarding, the communication transmission of important communication nodes is given priority according to the importance of the node (such as the importance of the backbone node is greater than that of the edge node). The priority of cross-domain communication tasks, when a node performs communication reception and forwarding, the smooth communication of general services is guaranteed through the first-in-first-out order of the same priority, and the queue priority management is performed according to the importance of the cross-domain communication task (high real-time task or high reliability task). Through high-priority services, such as: control instructions, etc., an immediate queue-jumping strategy is implemented to ensure the reliable and efficient transmission of important cross-domain communication tasks. Communication data transmission service quality management (Quality of Service, QoS): Communication data transmission QoS management mainly focuses on service quality, that is, the bandwidth occupied by end-to-end transmission of communication data, transmission delay and packet loss rate, etc., to build a dynamic detection and optimization mechanism for QoS service quality, to detect the delay, bandwidth, packet loss rate, etc. of cross-domain communication data transmission, to achieve QoS management and improvement in reducing communication transmission delay, ensuring communication bandwidth and reducing data packet loss rate.

[0107] The main function of the dynamic routing selection intelligent decision model is: for each communication node in the cross-domain communication network, when the communication node performs the data transmission task, the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node are input into the transmission strategy decision model corresponding to the current moment, and the target communication decision information of the communication node at the next moment is obtained, that is, the output node routing strategy information.

[0108] Adaptive rate switching intelligent decision model: Determine the target communication rate information based on the task category information in the communication task data and the preset mapping relationship between the task category and the rate. The model maps the hierarchical communication link rate with the classified communication service type to achieve service-rate perception and matching, so as to allocate different communication rates to different services as much as possible under limited resources, give priority to ensuring the reliable transmission of important cross-domain communication services, such as high real-time or high reliability services, and adaptively switch the transmission rate, including increasing or decreasing the transmission rate.

[0109] In the cross-domain communication networking intelligent decision algorithm model inference stage, it can be deployed on CPU, GPU or NPU devices. The device deployment of the cross-domain communication networking intelligent decision algorithm model mainly includes the following key steps: Step 1) Complete the development of the cross-domain communication networking intelligent decision algorithm model: Complete the algorithm design of the cross-domain communication resource management intelligent decision model, the routing selection intelligent decision model and the adaptive rate switching intelligent decision model respectively, and complete the development of the cross-domain communication networking intelligent decision algorithm model. Step 2) Configure the artificial intelligence (AI) computing device development environment: According to the deployment requirements of the cross-domain communication networking intelligent decision algorithm model and the requirements of the computing device (mainly including the CPU central processing unit, GPU graphics processing unit, and NPU neural network processor), install and run the development environment, and fully configure the dependency list and environment variables. Step 3) Cross-domain communication networking intelligent decision algorithm model migration: The cross-domain communication networking intelligent decision algorithm model is usually first developed on the CPU or GPU processor. When the cross-domain communication networking intelligent decision algorithm model needs to be migrated from the GPU or CPU to the NPU, it is necessary to develop operators for the unsupported AI model operators, etc. Step 4) Environment configuration and cross-domain communication networking intelligent decision-making model training and debugging: Configure environment variables, and use computers or industrial computers or servers equipped with CPUs or GPUs or NPU processors to perform training and debugging of cross-domain communication networking intelligent decision-making algorithm models. Step 5) Save and export the cross-domain communication networking intelligent decision-making algorithm model: After the training and debugging of the cross-domain communication networking intelligent decision-making algorithm model is completed, save and export the algorithm model as a model file suitable for computing processors such as CPUs or GPUs or NPUs. Step 6) AI computing device inference of cross-domain communication networking algorithm model: After the cross-domain communication networking intelligent decision-making algorithm model is trained, debugged, saved and exported, load and infer the cross-domain communication networking algorithm model on computing processors such as CPUs or GPUs or NPUs to complete the resource management, routing selection, and adaptive rate switching intelligent decision-making functions of the cross-domain communication networking intelligent decision-making algorithm.

[0110] Embodiment 4

[0111] Figure 6 A structural schematic diagram of an intelligent decision-making system applied to cross-domain communication networking provided in an embodiment of the present invention, wherein the cross-domain communication networking includes multiple communication nodes in the air domain, the surface domain, and the underwater domain, and the communication mode between the communication nodes is a radio communication mode, a laser communication mode, or an underwater acoustic communication mode. The system includes: a target information determination module 310 and a target information adjustment module 320.

[0112] Among them, the target information determination module 310 is used for each communication node in the cross-domain communication network. In the process of the communication node performing the data transmission task, the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node are input into the transmission strategy decision model corresponding to the current moment to obtain the target communication decision information of the communication node at the next moment; wherein the target communication decision information includes: the target link, the target link communication mode and the target link communication rate, the transmission strategy decision model is determined after optimizing the transmission strategy decision model corresponding to the previous moment based on the communication information rights and interests attribute at the previous moment, and the communication information rights and interests attribute is determined by the link communication feedback parameters corresponding to the communication node and the networking feedback parameters corresponding to the cross-domain communication network after the communication node completes the data transmission task based on the historical communication decision information at the historical moment;

[0113] The target information adjustment module 320 is used to adjust the transmission strategy of the communication node at the next moment based on the target communication decision information, so that the communication node performs the data transmission task based on the transmission strategy at the next moment and takes the next moment as the current moment.

[0114] Based on the above system, optionally, the transmission strategy decision model includes an action value mapping layer and an action decision layer, and a target information determination module, including:

[0115] The value parameter determination unit is used to input the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node into the action value mapping layer to obtain the value parameter corresponding to each preset execution action in the preset execution action set; wherein the preset execution action includes link information, the communication mode corresponding to the link, and the preset communication rate range corresponding to the communication mode;

[0116] The target value determination unit is used to input each value parameter into the action decision layer to obtain the target value parameter that meets the preset conditions;

[0117] The target link information determination unit is used to determine the target preset execution action corresponding to the target value parameter as the target link, target link communication mode and target link communication rate of the communication node at the next moment.

[0118] Based on the above system, optionally, the intelligent decision-making system applied to cross-domain communication networking further includes: an action set determination module;

[0119] The action set determination module is used to construct a preset execution action set based on at least one link, at least one communication mode and at least one preset communication rate range corresponding to each communication mode.

[0120] Based on the above system, optionally, the intelligent decision-making system applied to cross-domain communication networking further includes: a transmission strategy decision model determination module; the transmission strategy decision model determination module includes:

[0121] A feedback parameter acquisition unit, used to acquire the link communication feedback parameters corresponding to the communication node and the networking feedback parameters corresponding to the cross-domain communication network after the communication node completes the data transmission task based on the historical communication decision information at the historical moment; wherein the link communication feedback parameters include the target transmission delay information, the target packet loss rate and the target link rate corresponding to each link associated with the communication node, and the networking feedback parameters include the target load balancing degree;

[0122] A rate symmetry determination unit, configured to determine a target rate symmetry parameter based on each target link rate;

[0123] A rights attribute determination unit, used to determine the rights attribute of communication information based on the target load balancing degree, the target transmission delay information, the target packet loss rate and the target rate symmetry parameter;

[0124] The decision model optimization unit is used to optimize the model parameters of the transmission strategy decision model corresponding to the previous moment based on the communication information rights and interests attributes to obtain the transmission strategy decision model corresponding to the current moment.

[0125] Based on the above system, optionally, a rate symmetry determination unit is specifically used to determine a target rate symmetry parameter based on a ratio between target link rates.

[0126] Based on the above system, optionally, the equity attribute determination unit is specifically used to determine the first equity attribute based on the target load balancing degree and the first reward value mapping relationship; wherein the first reward value mapping relationship includes the mapping relationship between the load balancing degree and the first preset reward value; determine the second equity attribute based on the target transmission delay information and the second reward value mapping relationship; wherein the second reward value mapping relationship includes the mapping relationship between the transmission delay and the second preset reward value; determine the third equity attribute based on the target packet loss rate and the third reward value mapping relationship; wherein the third reward value mapping relationship includes the mapping relationship between the packet loss rate and the third preset reward value; determine the fourth equity attribute based on the target rate symmetry parameter and the fourth reward value mapping relationship; wherein the fourth reward value mapping relationship includes the mapping relationship between the rate symmetry parameter and the fourth preset reward value; determine the communication information equity attribute based on the first equity attribute, the second equity attribute, the third equity attribute and the fourth equity attribute.

[0127] Based on the above system, optionally, the communication task data includes task category information, and the task category information is a data transmission communication task, a control information communication task, a status information communication task category or a fault alarm information communication task.

[0128] Based on the above system, optionally, the intelligent decision-making system applied to cross-domain communication networking further includes: a rate update module;

[0129] The rate update module is used to determine the target communication rate information based on the task category information in the communication task data and the preset mapping relationship between the task category and the rate, so as to update the target link communication rate based on the target communication rate information.

[0130] The technical solution provided by the embodiment of the present invention can dynamically adjust the link weight in real time according to the link receiving and sending information of different links in the cross-domain communication network in such a complex communication scenario as the cross-domain communication network, so as to determine the route corresponding to the current moment according to the adjusted link weight, and can determine the route with the highest transmission efficiency, thereby improving the overall data transmission efficiency of the cross-domain communication network.

[0131] The intelligent decision-making system for cross-domain communication networking provided by the embodiment of the present invention can execute the intelligent decision-making method for cross-domain communication networking provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0132] It is worth noting that the various units and modules included in the above system are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the protection scope of the embodiments of the present invention.

[0133] Embodiment 5

[0134] Figure 7 The present invention is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 7 A block diagram of an exemplary electronic device 40 suitable for implementing exemplary embodiments of the present invention is shown. Figure 7 The electronic device 40 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0135] like Figure 7As shown, the electronic device 40 is in the form of a general computing device. The components of the electronic device 40 may include, but are not limited to: one or more processors or processing units 401, a system memory 402, and a bus 403 connecting different system components (including the system memory 402 and the processing unit 401). The bus 403 represents one or more of several types of bus structures, including a memory bus or a memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any bus structure in a variety of bus structures. For example, these architectures include, but are not limited to, an industrial standard architecture (ISA) bus, a microchannel architecture (MAC) bus, an enhanced ISA bus, a video electronics standard association (VESA) local bus, and a peripheral component interconnect (PCI) bus. The electronic device 40 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device 40, including volatile and non-volatile media, removable and non-removable media.

[0136] System memory 402 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 404 and / or cache memory 405. Electronic device 40 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 406 may be used to read and write non-removable, non-volatile magnetic media ( Figure 7 not shown, usually called a "hard drive"). Although Figure 7 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 403 via one or more data medium interfaces. Memory 402 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of various embodiments of the present invention.

[0137] A program / utility 408 having a set (at least one) of program modules 407 may be stored, for example, in memory 402, such program modules 407 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 407 generally perform the functions and / or methods of the embodiments described herein.

[0138] The electronic device 40 may also communicate with one or more external devices 409 (e.g., keyboards, pointing devices, displays 810, etc.), may communicate with one or more devices that enable a user to interact with the electronic device 40, and / or may communicate with any device that enables the electronic device 40 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication may be performed via an input / output (I / O) interface 411. Furthermore, the electronic device 40 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 412. As shown, the network adapter 412 communicates with other modules of the electronic device 40 via the bus 403. It should be understood that although Figure 7 Not shown, other hardware and / or software modules may be used in combination with the electronic device 40, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc. The processing unit 401 executes various functional applications and page processing by running programs stored in the system memory 402, such as implementing the intelligent decision-making method for cross-domain communication networking provided in the embodiment of the present invention.

Claims

1. An intelligent decision-making method applied to cross-domain communication networking, characterized in that: The cross-domain communication network includes multiple communication nodes in the air domain, the surface domain, and the underwater domain. The communication mode between the communication nodes is radio communication mode, laser communication mode, or underwater acoustic communication mode, including: For each of the communication nodes in the cross-domain communication networking, in the process of the communication node performing a data transmission task, the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication networking at the current moment, and the status data of each communication node are input into the transmission strategy decision model corresponding to the current moment, so as to obtain the target communication decision information of the communication node at the next moment; wherein the target communication decision information includes: the target link, the target link communication mode and the target link communication rate, the transmission strategy decision model is determined after optimizing the transmission strategy decision model corresponding to the previous moment based on the communication information rights and interests attribute at the previous moment, and the communication information rights and interests attribute is determined by the link communication feedback parameters corresponding to the communication node and the networking feedback parameters corresponding to the cross-domain communication networking after the communication node completes the data transmission task based on the historical communication decision information at the historical moment; The transmission strategy of the communication node at the next moment is adjusted based on the target communication decision information, so that the communication node performs the data transmission task based on the transmission strategy at the next moment, and takes the next moment as the current moment.

2. The method according to claim 1, characterized in that The transmission strategy decision model includes an action value mapping layer and an action decision layer, wherein the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node are input into the transmission strategy decision model corresponding to the current moment, and the target communication decision information of the communication node at the next moment is obtained, including: The communication task data corresponding to the data transmission task, the link status data of the cross-domain communication network at the current moment, and the status data of each communication node are input into the action value mapping layer to obtain the value parameters corresponding to each preset execution action in the preset execution action set; wherein the preset execution action includes link information, the communication mode corresponding to the link, and the preset communication rate range corresponding to the communication mode; Input each of the value parameters into the action decision layer to obtain a target value parameter that meets the preset conditions; The target preset execution action corresponding to the target value parameter is determined as the target link, target link communication mode and target link communication rate of the communication node at the next moment.

3. The method according to claim 2, characterized in that Also includes: A preset execution action set is constructed based on at least one link, at least one communication mode, and at least one preset communication rate range corresponding to each of the communication modes.

4. The method according to claim 1, characterized in that: The transmission strategy decision model corresponding to the current moment is determined in the following manner: After the communication node completes the data transmission task based on the historical communication decision information at the historical moment, the link communication feedback parameter corresponding to the communication node and the networking feedback parameter corresponding to the cross-domain communication networking are obtained; wherein the link communication feedback parameter includes the target transmission delay information, the target packet loss rate and the target link rate corresponding to each link associated with the communication node, and the networking feedback parameter includes the target load balancing degree; Determining a target rate symmetry parameter based on each of the target link rates; Determining communication information equity attributes based on the target load balancing degree, the target transmission delay information, the target packet loss rate, and the target rate symmetry parameter; Based on the communication information rights and interests attribute, the model parameters of the transmission strategy decision model corresponding to the previous moment are optimized to obtain the transmission strategy decision model corresponding to the current moment.

5. The method according to claim 4, characterized in that The determining a target rate symmetry parameter based on each of the target link rates includes: A target rate symmetry parameter is determined based on a ratio between the target link rates.

6. The method according to claim 1, characterized in that The determining of the communication information equity attribute based on the target load balancing degree, the target transmission delay information, the target packet loss rate, and the target rate symmetry parameter includes: Determine a first equity attribute based on the target load balancing degree and the first reward value mapping relationship; wherein the first reward value mapping relationship includes a mapping relationship between the load balancing degree and the first preset reward value; Determine a second equity attribute based on the target transmission delay information and the second reward value mapping relationship; wherein the second reward value mapping relationship includes a mapping relationship between the transmission delay and the second preset reward value; Determine a third equity attribute based on the target packet loss rate and the third reward value mapping relationship; wherein the third reward value mapping relationship includes a mapping relationship between the packet loss rate and the third preset reward value; Determining a fourth equity attribute based on the target rate symmetry parameter and the fourth reward value mapping relationship; wherein the fourth reward value mapping relationship includes a mapping relationship between the rate symmetry parameter and the fourth preset reward value; Based on the first equity attribute, the second equity attribute, the third equity attribute, and the fourth equity attribute, a communication information equity attribute is determined.

7. The method according to claim 1, characterized in that The communication task data includes task category information, and the task category information is a data transmission communication task, a control information communication task, a status information communication task category, or a fault alarm information communication task.

8. The method according to claim 7, characterized in that Before adjusting the transmission strategy of the communication node at the next moment based on the target communication decision information, the method further includes: Based on the task category information in the communication task data and a preset mapping relationship between the task category and the rate, target communication rate information is determined to update the target link communication rate based on the target communication rate information.

9. An intelligent decision-making system applied to cross-domain communication networking, characterized in that: The cross-domain communication network includes multiple communication nodes in the air domain, the surface domain, and the underwater domain. The communication mode between the communication nodes is radio communication mode, laser communication mode, or underwater acoustic communication mode, including: A target information determination module, for each communication node in the cross-domain communication networking, during the process of the communication node performing a data transmission task, inputting the communication task data corresponding to the data transmission task, the link status data of the cross-domain communication networking at the current moment, and the status data of each communication node into the transmission strategy decision model corresponding to the current moment, so as to obtain the target communication decision information of the communication node at the next moment; wherein the target communication decision information includes: a target link, a target link communication mode, and a target link communication rate, the transmission strategy decision model is determined after optimizing the transmission strategy decision model corresponding to the previous moment based on the communication information rights and interests attribute of the previous moment, and the communication information rights and interests attribute is determined by the link communication feedback parameters corresponding to the communication node and the networking feedback parameters corresponding to the cross-domain communication networking after the communication node completes the data transmission task based on the historical communication decision information at the historical moment; The target information adjustment module is used to adjust the transmission strategy of the communication node at the next moment based on the target communication decision information, so that the communication node performs the data transmission task based on the transmission strategy at the next moment and takes the next moment as the current moment.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the intelligent decision-making method applied to cross-domain communication networking as described in any one of claims 1-8.

Images