A distributed network node heartbeat message transmission optimization method and device

Abstract

The present invention discloses a method and device for optimizing the transmission of heartbeat messages of distributed network nodes, including: obtaining network link data between distributed network nodes; extracting multiple influencing factors affecting the link communication of distributed network nodes from the network link data, inputting the multiple influencing factors into a trained gated recurrent neural network model, predicting the link communication state of the distributed network node and the adjustment strategy of the heartbeat message sending frequency through the gated recurrent neural network model; adjusting the sending frequency of the node heartbeat message according to the prediction result of the gated recurrent neural network model, and optimizing the message transmission link of the distributed network node by using the ant colony algorithm. The present invention can ensure the dynamic optimization of the heartbeat detection frequency in an environment with weak communication and limited bandwidth, significantly improve the heartbeat detection efficiency and reduce the network load.

Description

Technical Field

[0001] The present invention relates to the field of network communication technology, and in particular to a method and device for optimizing the transmission of distributed network node heartbeat messages. Background Art

[0002] In a distributed network system, nodes need to maintain synchronization and coordination to ensure the normal operation of the network system. However, in a low-bandwidth, high-latency or unstable communication network environment, it becomes difficult to maintain synchronization and coordination between distributed nodes. Heartbeat detection is a common method that monitors the status and connection of each node by sending heartbeat messages regularly. Usually, sending heartbeat messages at fixed time intervals is the most basic heartbeat detection method.

[0003] However, in harsh external scenarios, there are weak communications and low-bandwidth information transmission. Traditional methods usually adjust the heartbeat data packets within a set time based on the monitored network load. However, in complex weather scenarios or highly dynamic external environments, the use of heartbeat detection technology with fixed time intervals between distributed cluster nodes will bring great defects:

[0004] (1) Heartbeat messages sent at fixed time intervals may occupy valuable bandwidth resources under weak communication conditions. Especially when there are a large number of nodes, the cumulative occupation of heartbeat messages may cause network congestion.

[0005] (2) In a high-latency network environment, a fixed time interval may cause the actual arrival time of the heartbeat message to be inaccurate, affecting the judgment of the node status.

[0006] Therefore, there is an urgent need for a distributed network node heartbeat message transmission optimization method suitable for weak communication conditions to adapt to different communication needs. Summary of the invention

[0007] In view of the above defects or deficiencies in the prior art, the present invention provides a method and device for optimizing the transmission of heartbeat messages of distributed network nodes, which can fully or partially solve the above technical problems.

[0008] One aspect of the present invention provides a method for optimizing the transmission of heartbeat messages of distributed network nodes, comprising: obtaining network link data between distributed network nodes; extracting multiple influencing factors that affect the link communication of distributed network nodes from the network link data, inputting the multiple influencing factors into a trained gated recurrent neural network model, and predicting the link communication status of the distributed network nodes and the adjustment strategy of the heartbeat message sending frequency through the gated recurrent neural network model; adjusting the sending frequency of the node heartbeat messages according to the prediction results of the gated recurrent neural network model, and optimizing the message transmission links of the distributed network nodes using an ant colony algorithm.

[0009] On the other hand, the present invention also provides a transmission optimization device for distributed network node heartbeat messages, including: a monitoring module, configured to obtain network link data between distributed network nodes; a prediction module, configured to extract multiple influencing factors that affect the link communication of distributed network nodes from the network link data, and input the multiple influencing factors into a trained gated recurrent neural network model, and predict the link communication status of the distributed network node and the adjustment strategy of the heartbeat message sending frequency through the gated recurrent neural network model; an optimization module, configured to adjust the sending frequency of the node heartbeat message according to the prediction results of the gated recurrent neural network model, and use the ant colony algorithm to optimize the message transmission link of the distributed network node.

[0010] The present invention provides a distributed network node heartbeat message transmission optimization method and device, which has the following beneficial effects:

[0011] (1) Using a gated recurrent neural network model to perform distributed node adaptive heartbeat security detection under weak communication conditions can effectively capture the timing characteristics of the heartbeat signal, determine whether the communication interval of the communication link node needs to be adjusted, and identify potential anomalies. The present invention combines the advantages of deep learning to ensure that the heartbeat detection frequency can be dynamically optimized under weak communication and bandwidth-constrained environments.

[0012] (2) When the gated recurrent neural network model identifies the early warning stage, the message transmission link is optimized through the ant colony algorithm, and resources are reallocated according to the real-time link status, which can significantly improve the heartbeat detection efficiency and reduce the network load. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0014] Figure 1 This is a schematic diagram of a distributed network node heartbeat message transmission optimization method provided by an embodiment of the present application. Figure 1 ;

[0015] Figure 2 This is a schematic diagram of a distributed network node heartbeat message transmission optimization method provided by an embodiment of the present application. Figure 2 ;

[0016] Figure 3 It is a system logic diagram provided by an embodiment of the present application;

[0017] Figure 4 It is a structural diagram of a distributed network node heartbeat message transmission optimization device provided by an embodiment of the present application;

[0018] Figure 5 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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 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 are within the scope of protection of the present invention.

[0020] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0021] It should be understood that although the terms first, second, third, etc. may be used to describe the acquisition modules in the embodiments of the present invention, the acquisition modules should not be limited to these terms. These terms are only used to distinguish the acquisition modules from each other.

[0022] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0023] It should be noted that the directional words such as "upper", "lower", "left", and "right" described in the embodiments of the present invention are described at the angles shown in the drawings and should not be understood as limiting the embodiments of the present invention. In addition, in the context, it should also be understood that when it is mentioned that an element is formed "on" or "under" another element, it can not only be formed directly "on" or "under" another element, but also be formed "on" or "under" another element indirectly through an intermediate element.

[0024] This application proposes a distributed node adaptive time interval heartbeat transmission optimization mechanism in a weak communication and bandwidth-constrained environment. Figure 1-2 The detailed steps are as follows:

[0025] Step S101, obtaining network link data between distributed network nodes.

[0026] See also Figure 3 The present invention realizes the monitoring and maintenance of node dynamic links through functions such as node state maintenance, adaptive communication, state table merging, and topology information updating, and timely obtains node state changes and link network quality. Specifically, in a distributed network, each node maintains a state table, which contains state information of itself and other known nodes, such as node ID, last communication timestamp, node state, neighbor node list, etc. The node ID in the state information is used to uniquely identify each node, the last communication timestamp is used to record the time of the last successful communication, the node state indicates whether the node is online, and the neighbor node list indicates the connection between the node and other nodes. Exemplarily, the distributed topology structure has nodes A, B, C, and D. Node A selects node B for adaptive communication. The state table of node A includes "node ID: A, state: normal, last communication timestamp: T1, neighbor node list: {B, C}", and the state table of node B includes "node ID: B, state: normal, last communication timestamp: T2, neighbor node list: {A, D}". In addition, the monitored network link data includes but is not limited to: the actual used bandwidth of the link (the average transmission rate in a specific time period), the maximum available bandwidth of the link (theoretical maximum transmission rate), data packet sending time, data packet arrival time, the number of data packets, the number of data packets sent, the number of data packets successfully received, etc.

[0027] Step S102: extract multiple influencing factors that affect the link communication of distributed network nodes from the network link data, input the multiple influencing factors into a trained gated recurrent neural network model, and predict the link communication status of the distributed network nodes and the adjustment strategy of the heartbeat message sending frequency through the gated recurrent neural network model.

[0028] Dynamic bandwidth utilization, dynamic transmission delay rate, and dynamic packet loss rate are key factors affecting monitoring. Bandwidth utilization can be calculated by monitoring the ratio of the amount of data passing through the link within a specific time to the maximum carrying capacity of the link. Transmission delay can be measured by sending a test data packet and calculating its round-trip time (RTT). Packet loss rate can be calculated by counting the difference between the total number of data packets sent within a certain period of time and the number of data packets successfully received.

[0029] Exemplarily, the bandwidth utilization is calculated based on the monitored network link data , transmission delay and packet loss rate :

[0030] Calculate bandwidth utilization according to the following formula :

[0031]

[0032] in, To actually use the bandwidth, is the maximum available bandwidth;

[0033] The transmission delay is calculated according to the following formula :

[0034]

[0035] in, express , express , express ;

[0036] The packet loss rate is calculated according to the following formula :

[0037]

[0038] in, Indicates the number of packets sent. Indicates the number of packets successfully received.

[0039] The bandwidth utilization rate is calculated according to the following formula , transmission delay and packet loss rate Perform normalization:

[0040] ;

[0041] ;

[0042] ;

[0043]

[0044] ;

[0045] ;

[0046] The communication status information of the distributed network nodes is calculated according to the following formula: :

[0047] ;

[0048] 1;

[0049] in, for , , The weight coefficient of , , for , , Results after removing outliers;

[0050] The bandwidth utilization at the current moment , the transmission delay at the current moment , the packet loss rate at the current moment And the communication status information of the distributed network nodes at the previous moment The influence factors are input into the input layer of the gated recurrent neural network model. The gated recurrent neural network model is used to predict the link communication status of the distributed network nodes and the adjustment strategy of the heartbeat message sending frequency.

[0051] The gated recurrent neural network is a model that can solve problems such as long-term memory and gradient in back propagation. Its basic structure includes input layer, hidden layer and output layer. The input of reset gate and update gate in the gated recurrent unit is the node communication state at the current moment and the node communication state at the previous moment, that is, the current moment Value and the previous moment The output of the hidden layer is calculated by the fully connected layer with the activation function sigmoid.

[0052] The construction process of the gated recurrent neural network is as follows:

[0053] S1: Bandwidth utilization , transmission delay , Packet loss rate The communication status of the distributed network nodes at the previous moment As an impact factor;

[0054] S2: Quantify the impact factor data and expand the data features. The activation function of the hidden layer uses the Sigmoid function;

[0055] S3: Communication status of distributed network nodes at the last moment and the current time

[0056] Communication status of distributed network nodes To obtain two gated states, the gated recurrent neural network is divided into two gates, namely the reset gate and the update gate. The communication state of the distributed network nodes at the current moment Represents the bandwidth utilization at the current moment , the transmission delay at the current moment , the packet loss rate at the current moment The set consists of a reset gate and an update gate, respectively, obtained through the sigmoid function. The reset gate indicates how much information of the previous state is written into the current candidate set, and the update gate indicates the degree to which the information of the previous state is brought into the current state.

[0057] Time step The output values ​​of the reset gate and update gate are:

[0058] ( ;

[0059] ( ;

[0060] in, , , , is the weight matrix, is the hidden layer output value at the previous moment, is the current input data, , is the bias value, is the sigmoid activation function.

[0061] The candidate hidden layer state can be expressed as:

[0062] ( ;

[0063] When it approaches 0, the model discards the past hidden information. When it approaches 1, the model adds past hidden information to current information.

[0064] The final hidden state can be expressed as:

[0065]

[0066] in Contains information from the past. is the candidate hidden layer state, Update gate output value.

[0067] Step S103, according to the prediction result of the gated recurrent neural network model, the sending frequency of the node heartbeat message is adjusted, and the message transmission link of the distributed network node is optimized by using the ant colony algorithm.

[0068] Specifically, the influencing factors are input into the trained gated recurrent neural network model for prediction, where Y is the output layer, and Y(t) is the output of the output layer at time t. When the output result of Y(t) is less than 0.7, the output layer outputs the judgment result of the node adaptive communication and the state of the current node heartbeat, proving that the network delay and packet loss rate are low and stable; when the output result of Y(t) is greater than 0.7, the output layer outputs the judgment result of the node adaptive communication and the state of the current node heartbeat, proving that the network state is poor and the communication environment is harsh.

[0069] Phase 1: 0<Y(t)<0.7, this is a healthy state, and the dynamic link shortens the interval between sending adaptive heartbeat packets to receive more timely messages;

[0070] The second stage: 0.7≤Y(t)<1, which is the warning state. At this time, the heartbeat packet sending interval should be kept stable to prevent further deterioration of the network environment.

[0071] When the gated recurrent neural network model outputs Y(t) to 0.7≤Y(t)<1, the early warning stage is triggered. At this time, the heartbeat message needs to be adaptively adjusted and the entire dynamic link needs to be optimized to adapt to the complex communication environment requirements. The ant colony algorithm is an optimization algorithm that simulates the foraging behavior of ants in nature. It simulates the process of ants releasing pheromones in the search space and making decisions based on the amount of pheromones and heuristic information. The ant colony algorithm can optimize the message transmission links of distributed nodes and reallocate resources based on the real-time link status.

[0072] The following is a specific implementation of the ant colony algorithm in a distributed node dynamic link.

[0073] (1) Pheromone initialization

[0074] The initial pheromone concentration of each link (between two nodes) is set to a small positive value of 0.1, indicating that each link has a certain basic reliability. After each heartbeat detection or communication is completed, the pheromone is updated according to the communication result.

[0075] (2) Heuristic Information Computation

[0076] ① Heuristic information corresponding to delay

[0077] The communication delay between nodes can be obtained by calculating the response time of the most recent heartbeat packet. The lower the delay, the higher the heuristic information value, and an inverse relationship can be used, calculated as follows:

[0078]

[0079] in, is the delay from node i to node j.

[0080] ②Heuristic information corresponding to packet loss rate

[0081] The packet loss rate refers to the ratio of the number of heartbeat packets that fail to arrive successfully to the number of heartbeat packets sent within a certain period of time. The calculation method is as follows:

[0082]

[0083] in, is the packet loss rate from node i to node j.

[0084] ③Packet loss rate corresponding to node load

[0085] The lower the node's load (CPU and memory usage), the higher the heuristic information value, which is calculated as follows:

[0086]

[0087] in is the current load of node i.

[0088] The influence is incorporated into the pheromone amount by means of a weighted sum, calculated as follows:

[0089] ;

[0090] in, is the initial pheromone amount, is the pheromone evaporation coefficient, , , They represent the influence values ​​of different relationships respectively.

[0091] (3) Comprehensive behavioral assessment

[0092] Each ant selects the next communication node based on the amount of pheromone and heuristic information, and makes decisions based on the amount of pheromone and heuristic information. Select the next node Probability It can be calculated by the following formula:

[0093] ;

[0094] in, Indicates the node in the current link state To Node pheromone concentration; Represents heuristic information; is the weight parameter that adjusts the pheromone and heuristic information.

[0095] (4) Pheromone update

[0096] After each heartbeat detection or communication is completed, the pheromone is updated according to the communication result, which can be calculated by the following formula:

[0097] ;

[0098] in, represents the pheromone volatilization rate, Indicates the amount of pheromone added to the link state when ant s passes through the currently selected link. When the communication is successful and the link quality is good Take a constant value when communication fails or link quality is poor is 0; Indicates the node in the previous link state To Node pheromone concentration.

[0099] (5) Repeated iteration

[0100] The ant behavior simulation and pheromone update process are repeated until the maximum number of iterations K is met.

[0101] (6) Calculate the optimal solution

[0102] After multiple iterations, the optimal distributed node communication strategy in a weak communication environment (i.e., the communication path with the most ants) found by the ant colony is taken as the final result of the complex communication link environment.

[0103] This embodiment combines the gated recurrent neural network model and the ant colony algorithm to achieve more efficient distributed node adaptive heartbeat security detection under weak communication conditions. The gated recurrent neural network is responsible for determining whether the adaptive communication interval of the communication link node needs to be adjusted and identifying anomalies. When it is identified as the early warning stage, the ant colony algorithm optimizes the message transmission link and reallocates resources according to the real-time link status. The algorithm combination of this embodiment not only improves the accuracy of anomaly detection, but also helps the system maintain stability and robustness in complex external communication environments and during network fluctuations and node failures.

[0104] See also Figure 4 Another embodiment of the present invention further provides a distributed network node heartbeat message transmission optimization device 200, including a monitoring module 201, a prediction module 202 and an optimization module 203. The device 200 can execute the distributed network node heartbeat message transmission optimization method in the method embodiment.

[0105] Specifically, the device 200 includes:

[0106] The monitoring module 201 is configured to obtain network link data between distributed network nodes;

[0107] The prediction module 202 is configured to extract multiple influencing factors affecting the link communication of the distributed network nodes from the network link data, input the multiple influencing factors into the trained gated recurrent neural network model, and predict the link communication state of the distributed network nodes and the adjustment strategy of the heartbeat message sending frequency through the gated recurrent neural network model;

[0108] The optimization module 203 is configured to adjust the sending frequency of the node heartbeat message according to the prediction result of the gated recurrent neural network model, and optimize the message transmission link of the distributed network node by using the ant colony algorithm.

[0109] Furthermore, the prediction module is configured to calculate the bandwidth utilization rate based on the monitored network link data. , transmission delay and packet loss rate ; The bandwidth utilization calculated according to the following formula , transmission delay and packet loss rate Perform normalization:

[0110] ;

[0111] ;

[0112] ;

[0113]

[0114] ;

[0115] ;

[0116] The state parameter value of the distributed network node is calculated according to the following formula :

[0117] ;

[0118] 1;

[0119] in, for , , The weight coefficient of , , for , Results after removing outliers;

[0120] The bandwidth utilization at the current moment , the transmission delay at the current moment , the packet loss rate at the current moment The communication status of the distributed network nodes at the previous moment It is input as an influencing factor into the input layer of the gated recurrent neural network model.

[0121] Furthermore, the prediction module is configured as follows: if the output of the gated recurrent neural network model is less than a preset threshold, the link communication of the distributed network node is in a healthy state, and the sending interval of the heartbeat message is shortened; if the output of the gated recurrent neural network model is greater than the preset threshold, the link communication of the distributed network node is in a warning state, and the sending interval of the heartbeat message is maintained unchanged.

[0122] Furthermore, the optimization module is configured to calculate the current node according to the following formula: Select next node Probability :

[0123] ;

[0124] ;

[0125] ;

[0126] ;

[0127] ;

[0128] ;

[0129] in, Indicates the node in the current link state To Node pheromone concentration; It represents the set of path nodes that the ants have not visited in the process of continuously searching for paths; Represents heuristic information; is the weight parameter for adjusting pheromone and heuristic information; represents the pheromone volatilization rate; Indicates the amount of pheromone added to the link state when ant s passes through the currently selected link. When the communication is successful and the link quality is good Take a constant value when communication fails or link quality is poor is 0; Indicates the node in the previous link state To Node pheromone concentration; is the heuristic information value corresponding to the transmission delay; is the heuristic information value corresponding to the packet loss rate; Indicates the heuristic information value corresponding to the node load; is the delay from node i to node j; is the packet loss rate from node i to node j; is the current load of node i; is the initial pheromone concentration; , , yes , , The corresponding weight coefficient;

[0130] The probability The highest node is used as the next node for message transmission, and the ant behavior simulation and pheromone update are repeated until the maximum number of iterations is met, and the message transmission path with the most ants is taken as the optimal path.

[0131] Furthermore, the prediction module is configured as follows:

[0132] Calculate bandwidth utilization according to the following formula :

[0133]

[0134] in, To actually use the bandwidth, is the maximum available bandwidth;

[0135] The transmission delay is calculated according to the following formula :

[0136]

[0137] in, express , express , express ;

[0138] The packet loss rate is calculated according to the following formula :

[0139]

[0140] in, Indicates the number of packets sent. Indicates the number of packets successfully received.

[0141] It should be noted that the distributed network node heartbeat message transmission optimization device 200 provided in this embodiment corresponds to a technical solution that can be used to execute each method embodiment. Its implementation principle and technical effect are similar to the method and will not be repeated here.

[0142] See also Figure 5 Another embodiment of the present invention provides a schematic diagram of the structure of an electronic device 300, which is used to implement the transmission optimization method of distributed network node heartbeat messages in the method embodiment. The electronic device 300 in the embodiment of the present invention may include but is not limited to a PC and a server. Figure 5 The electronic device 300 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.

[0143] like Figure 5 As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 to a random access memory (RAM) 303 to implement the method of the embodiment of the present invention. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0144] Typically, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Figure 5 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0145] The above description is only a preferred embodiment of the present invention. Those skilled in the art should understand that the scope of disclosure involved in the present invention is not limited to the technical solution formed by a specific combination of the above technical features, but also should cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present invention (but not limited to) to form a technical solution.

Claims

1. A method for optimizing the transmission of heartbeat messages of distributed network nodes, characterized in that: include: Obtain network link data between distributed network nodes; Extracting multiple influencing factors affecting the link communication of distributed network nodes from the network link data, inputting the multiple influencing factors into a trained gated recurrent neural network model, and predicting the link communication state of the distributed network nodes and the adjustment strategy of the heartbeat message sending frequency through the gated recurrent neural network model; According to the prediction results of the gated recurrent neural network model, the sending frequency of node heartbeat messages is adjusted, and the message transmission link of distributed network nodes is optimized using the ant colony algorithm; The step of extracting a plurality of influencing factors affecting distributed network node link communication from the network link data comprises: Calculate bandwidth utilization based on monitored network link data , transmission delay and packet loss rate ; The bandwidth utilization rate is calculated according to the following formula , transmission delay and packet loss rate Perform normalization: ； ； ； ； ； The state parameter value of the distributed network node is calculated according to the following formula : ； 1； in, for , , The weight coefficient of , , for , Results after removing outliers; The bandwidth utilization at the current moment , the transmission delay at the current moment , the packet loss rate at the current moment The communication status of the distributed network nodes at the previous moment As an influencing factor, it is input into the input layer of the gated recurrent neural network model; The step of predicting the link communication status of the distributed network node and the adjustment strategy of the heartbeat message sending frequency by using the gated recurrent neural network model includes: If the output of the gated recurrent neural network model is less than the preset threshold, the link communication of the distributed network node is in a healthy state, and the sending interval of the heartbeat message is shortened; If the output of the gated recurrent neural network model is greater than the preset threshold, the link communication of the distributed network node is in a warning state, and the sending interval of the heartbeat message is maintained unchanged; The step of optimizing the message transmission link of the distributed network nodes by using the ant colony algorithm includes: Calculate the current node according to the following formula Select next node Probability : ； ； ； ； ； ； in, Indicates the node in the current link state To Node pheromone concentration; It represents the set of path nodes that the ants have not visited in the process of continuously searching for paths; Represents heuristic information; is the weight parameter for adjusting pheromone and heuristic information; represents the pheromone volatilization rate; Indicates the amount of pheromone added to the link state when ant s passes through the currently selected link. When the communication is successful and the link quality is good Take a constant value when communication fails or link quality is poor is 0; Indicates the node in the previous link state To Node pheromone concentration; is the heuristic information value corresponding to the transmission delay; is the heuristic information value corresponding to the packet loss rate; Indicates the heuristic information value corresponding to the node load; Is a node i To Node j Delays; Is a node i To Node j Packet loss rate; Is a node i The current load; is the initial pheromone concentration; , , yes 、 、 The corresponding weight coefficient; The probability The highest node is used as the next node for message transmission, and the ant behavior simulation and pheromone update are repeated until the maximum number of iterations is met, and the message transmission path with the most ants is taken as the optimal path.

2. The method for optimizing the transmission of heartbeat messages of distributed network nodes according to claim 1, characterized in that: Calculate bandwidth utilization according to the following formula : in, To actually use the bandwidth, is the maximum available bandwidth; The transmission delay is calculated according to the following formula: : in, express , express , express ; The packet loss rate is calculated according to the following formula : in, Indicates the number of packets sent. Indicates the number of packets successfully received.

3. A distributed network node heartbeat message transmission optimization device, characterized in that: include: A monitoring module configured to obtain network link data between distributed network nodes; A prediction module is configured to extract multiple influencing factors affecting the link communication of distributed network nodes from the network link data, input the multiple influencing factors into a trained gated recurrent neural network model, and predict the link communication state of the distributed network nodes and the adjustment strategy of the heartbeat message sending frequency through the gated recurrent neural network model; an optimization module, configured to adjust the sending frequency of the node heartbeat message according to the prediction result of the gated recurrent neural network model, and optimize the message transmission link of the distributed network node by using the ant colony algorithm; Wherein, the prediction module is further configured as follows: Calculate bandwidth utilization based on monitored network link data , transmission delay and packet loss rate ; The bandwidth utilization rate is calculated according to the following formula , transmission delay and packet loss rate Perform normalization: ； ； ； ； ； The state parameter value of the distributed network node is calculated according to the following formula : ； 1； in, for , , The weight coefficient of , , for , Results after removing outliers; The bandwidth utilization at the current moment , the transmission delay at the current moment , the packet loss rate at the current moment The communication status of the distributed network nodes at the previous moment As an influencing factor, it is input into the input layer of the gated recurrent neural network model; Wherein, the prediction module is further configured as follows: If the output of the gated recurrent neural network model is less than the preset threshold, the link communication of the distributed network node is in a healthy state, and the sending interval of the heartbeat message is shortened; If the output of the gated recurrent neural network model is greater than the preset threshold, the link communication of the distributed network node is in a warning state, and the sending interval of the heartbeat message is maintained unchanged; Wherein, the optimization module is further configured as follows: Calculate the current node according to the following formula Select next node Probability : ； ； ； ； ； ； in, Indicates the node in the current link state To Node pheromone concentration; It represents the set of path nodes that the ants have not visited in the process of continuously searching for paths; Represents heuristic information; is the weight parameter for adjusting pheromone and heuristic information; represents the pheromone volatilization rate; Indicates the amount of pheromone added to the link state when ant s passes through the currently selected link. When the communication is successful and the link quality is good Take a constant value when communication fails or link quality is poor is 0; Indicates the node in the previous link state To Node pheromone concentration; is the heuristic information value corresponding to the transmission delay; is the heuristic information value corresponding to the packet loss rate; Indicates the heuristic information value corresponding to the node load; Is a node i To Node j Delays; Is a node i To Node j Packet loss rate; Is a node i The current load; is the initial pheromone concentration; , , yes 、 、 The corresponding weight coefficient; The probability The highest node is used as the next node for message transmission, and the ant behavior simulation and pheromone update are repeated until the maximum number of iterations is met, and the message transmission path with the most ants is taken as the optimal path.

4. A distributed network node heartbeat message transmission optimization device according to claim 3, characterized in that: The prediction module is configured to: Calculate bandwidth utilization according to the following formula : in, To actually use the bandwidth, is the maximum available bandwidth; The transmission delay is calculated according to the following formula: : in, express , express , express ; The packet loss rate is calculated according to the following formula : in, Indicates the number of packets sent. Indicates the number of packets successfully received.

Images