Ad hoc network information processing method for mass internet of things devices

Abstract

The application relates to the technical field of information processing, in particular to a self-organizing network information processing method for a mass Internet of Things device, which comprises the following steps: acquiring a network topology structure of a target self-organizing network, performing graph structure modeling on the network topology structure to obtain a network undirected graph corresponding to the network topology structure; calculating a node weight of each node in the network undirected graph based on a node position of each node in the network undirected graph, and calculating a node label probability of each node in the network undirected graph based on the node weight. The application evaluates the importance of network nodes by calculating the weight of the nodes and based on the node label probability, which can help to divide the whole network into multiple control domains according to the characteristics of different nodes, and each control domain is independently managed by dividing the network into multiple sub-target self-organizing networks, so that the flexibility of network management is improved, and the management of different areas in the network can be more precise.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of information processing, and in particular to a self-organizing network information processing method for massive Internet of Things devices. BACKGROUND

[0002] Massive Internet of Things devices refer to a large number of physical devices connected through the Internet, which usually have low power consumption, low data transmission demand, low bandwidth requirement, and intelligent processing and transmission capabilities. With the development of Internet of Things technology, especially in the fields of smart city, industrial automation, smart home, and smart agriculture, the scale of massive Internet of Things devices is continuously expanding.

[0003] Traditional methods usually adopt static network topology structure, which cannot dynamically adapt to changes in network environment. In actual Internet of Things (IoT) applications, the number of devices is large and constantly changing, and traditional methods often cannot effectively cope with such dynamic changes, resulting in inefficient and non-scalable network management. Moreover, traditional methods often rely on simple global optimization for network management, and cannot finely divide different areas for independent management. The role of each node and the importance of the network are often not fully evaluated, resulting in chaotic management hierarchy and low efficiency. Furthermore, traditional methods often rely on pre-set rules or fixed control strategies for network management and congestion control, lacking the ability of real-time prediction and dynamic adjustment. Once the network is congested, it is often difficult to identify and take effective measures in advance, resulting in decreased network performance. In addition, traditional methods are usually based on simple resource allocation rules, without fully considering the complex relationship between nodes and the optimization of resource utilization. In the face of massive device networks, traditional methods often lead to unbalanced, wasted, or insufficient resource scheduling, affecting network throughput and device energy efficiency. SUMMARY

[0004] The technical problem to be solved by the present application is to overcome the shortcomings of the above-mentioned prior art and provide a self-organizing network information processing method for massive Internet of Things devices.

[0005] The technical solution adopted to solve the above technical problems is: a self-organizing network information processing method for massive Internet of Things devices, comprising:

[0006] obtaining the network topology structure of a target self-organizing network, and modeling the network topology structure as a graph structure to obtain a network undirected graph corresponding to the network topology structure;

[0007] calculating the node weight of each node in the network undirected graph based on the node position of each node in the network undirected graph, and calculating the node label probability of each node in the network undirected graph based on the node weight;

[0008] The penalty coefficient of each node in the network undirected graph is calculated based on the node label probability, and the penalty coefficient of each node is compared with a preset penalty threshold table to divide the target ad hoc network into multiple control domains, i.e., the target ad hoc network is divided into multiple sub-target ad hoc networks, and the sub-target ad hoc network corresponds to the control domain;

[0009] The state of the sub-target ad hoc network in the control domain is monitored according to a preset monitoring period to obtain state data corresponding to the sub-target ad hoc network;

[0010] The state data is subjected to congestion prediction based on a pre-trained congestion prediction model to obtain a congestion level label corresponding to the sub-target ad hoc network;

[0011] A congestion control model is constructed based on the state data and the congestion level label corresponding to the sub-target ad hoc network, and the congestion control model is solved based on a reinforcement learning algorithm to obtain a congestion control scheme corresponding to the sub-target ad hoc network.

[0012] Preferably, the network undirected graph includes a node set and an edge set, wherein the nodes in the node set correspond to Internet of Things devices, and the edges in the edge set correspond to communication links between two Internet of Things devices.

[0013] Preferably, the calculation formula of the node weight of each node in the network undirected graph is as follows:

[0014]

[0015] wherein W i denotes the node weight of the i th node in the network undirected graph, TL i denotes the maximum value of the delay between the i th node and all neighbor nodes of the i th node in the network undirected graph, and v j denotes the j th node in the network undirected graph, N S denotes the neighbor node set of the i th node in the network undirected graph, d ij denotes the distance between the i th node and the j th node in the network undirected graph, and p denotes the electromagnetic wave propagation speed.

[0016] Preferably, the calculation formula of the node label probability of each node in the network undirected graph is as follows:

[0017]

[0018] wherein l i denotes the node label, and l i = 0, 1, P(l i = 1) denotes the node label li node label probability, and P(l i = 0) represents the node label probability of the node label l i = 0, and P(l i = 0) = 1 - P(l i = 1), W min and W max represent the minimum and maximum values of the node weight.

[0019] Preferably, the penalty coefficient of each node in the network undirected graph is calculated based on the node label probability, comprising:

[0020] A node association function is constructed based on the node weight, wherein the node association function is as follows:

[0021]

[0022] wherein AC(i, j) represents the node association degree between two nodes in the network undirected graph, η represents the distance coefficient, and Rt represents the node communication range, and σ represents the standard deviation of the node weight;

[0023] A node association penalty function is constructed based on the node association function, wherein the association penalty function is as follows:

[0024]

[0025] wherein ACP i represents the node association penalty coefficient of the i th node in the network undirected graph, V represents the node set of the network undirected graph, δ(l i , l j ) represents the association indicator function, and

[0026] A node self-penalty function is constructed based on the node label probability, wherein the node self-penalty function is as follows:

[0027] SCP i = -ln(P(l i = l j ));

[0028] wherein SCP i represents the node self-penalty coefficient of the i th node in the network undirected graph, P(l i = l j ) represents the probability that the node label of the i th node in the network undirected graph belongs to the real node label;

[0029] constructing a penalty function of the network undirected graph based on the node self-penalty function and the node association penalty function, wherein the penalty function of the network undirected graph is as follows:

[0030] LF i = β · SCP i + ACP i ;

[0031] wherein, LF i represents a penalty coefficient of the i-th in the network undirected graph, and β represents a preset important factor.

[0032] Preferably, the expression of the state data corresponding to the sub-target ad hoc network is as follows:

[0033] SD = {RTT avg , Ack num , Thr avg , cwnd cl , v, RTT, MinRTT, Jitter e , loss} ;

[0034] wherein, SD represents the state data corresponding to the sub-target ad hoc network, RTT avg represents the average packet delay in the monitoring period, Ack num represents the number of valid acknowledgement messages in the monitoring period, Thr avg represents the average throughput in the monitoring period, cend cl represents the current congestion window size in the monitoring period, v represents the data sending rate at the end of the monitoring period, RTT represents the last RTT in the monitoring period, MinRTT represents the minimum RTT in the recent preset time, Jitter e represents the round-trip delay jitter value, and loss represents the packet loss rate of the communication link.

[0035] Preferably, the state data is subjected to congestion prediction based on a pre-trained congestion prediction model to obtain the congestion level label corresponding to the sub-target ad hoc network, comprising:

[0036] performing feature extraction on the state data to obtain a congestion feature set, wherein the expression of the congestion feature set is as follows:

[0037]

[0038] wherein, FS represents the congestion feature set.

[0039] performing congestion prediction on the congestion feature set based on a pre-trained congestion prediction model to obtain the congestion level label corresponding to the sub-target ad hoc network, wherein the congestion prediction model adopts a support vector machine model.

[0040] Preferably, the congestion control model is constructed based on the state data corresponding to the sub-target MANET and the congestion level label, and comprises:

[0041] A state space of the congestion control model is constructed based on the state data corresponding to the sub-target MANET;

[0042] A reward function of the congestion control model is constructed based on the state data corresponding to the sub-target MANET and the congestion level label;

[0043] An action space of the congestion control model is constructed based on all communication links of the sub-target MANET. Preferably, the expression of the state space of the congestion control model is as follows:

[0044] s t =(SD t ,SD t-1 ,...,SD t-k );

[0045] wherein s t represents the state space of the congestion control model, SD t represents the state data corresponding to the sub-target MANET at time t;

[0046] The reward function of the congestion control model is as follows:

[0047]

[0048] wherein reward represents the reward function of the congestion control model, p represents the congestion level corresponding to the sub-target MANET, and p th represents a preset congestion level threshold.

[0049] Preferably, the congestion control model is solved based on a reinforcement learning algorithm to obtain the congestion control scheme corresponding to the sub-target MANET, and the method comprises:

[0050] An action is selected from the action space according to a policy π and executed in the state space environment through the monitored state space s t , so as to receive a reward reward and enter a next state space s t+1 , and the policy π is a mapping from the state space s t to the action space;

[0051] The reinforcement learning algorithm is learned based on the maximum cumulative reward to obtain a policy π corresponding to the maximum reward, wherein the policy π is the congestion control scheme corresponding to the sub-target MANET, and the learning formula of the cumulative reward is as follows:

[0052]

[0053] Among them, E t represents the cumulative reward, and T represents the number of iterations of the reinforcement learning algorithm.

[0054] The beneficial effects of the present invention are as follows: (1) The present invention obtains an undirected graph representation by graph structure modeling of the network topology of the self-organizing network. This modeling method can clearly describe the network nodes and connection relationships, making subsequent management and optimization more controllable, and by calculating the weight of the node and evaluating the importance of the network node based on the node label probability, it can help divide the entire network into multiple control domains according to the characteristics of different nodes, and by dividing the network into multiple sub-target self-organizing networks, each control domain is managed independently and matched with its corresponding sub-target self-organizing network. This division improves the flexibility of network management and makes the management of different areas in the network more refined; (2) The present invention uses a pre-trained congestion prediction model to predict the congestion level of the network status data, so that The congestion situation of each sub-target self-organizing network is understood in real time, and measures are taken in advance to avoid the occurrence of congestion. Based on the congestion prediction, the congestion control model is solved by the reinforcement learning algorithm. This means that the congestion control strategy can be dynamically optimized according to the real-time network status in the process of continuous learning, thereby reducing network congestion, improving throughput, and improving the performance of the overall network; (3) The present invention can make adaptive adjustments according to the constantly changing network status through the reinforcement learning algorithm. When the network conditions change, the model can automatically adjust the control scheme, avoiding the limitations of traditional static methods. Moreover, since the characteristics of self-organizing networks are a large number of devices and their wide distribution, the use of graph structure modeling and control domain division methods can effectively manage each sub-target self-organizing network independently, thereby ensuring stable operation under the condition of massive devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 The figure is a flowchart of the steps of the overall method in one embodiment of the present invention. DETAILED DESCRIPTION

[0056] Example 1, as Figure 1 As shown, the present invention proposes a method for processing information of a self-organizing network for massive IoT devices, comprising:

[0057] S1. Obtain the network topology of the target ad hoc network, perform graph structure modeling on the network topology to obtain an undirected network graph corresponding to the network topology;

[0058] S2. Calculate the node weight of each node in the network undirected graph based on the node position of each node in the network undirected graph, and calculate the node label probability of each node in the network undirected graph based on the node weight;

[0059] S3, calculate a penalty coefficient of each node in the network undirected graph based on the node label probability, compare the penalty coefficient of each node with a preset penalty threshold table, and divide the target ad hoc network into multiple control domains, i.e., divide the target ad hoc network into multiple sub-target ad hoc networks, and the sub-target ad hoc network corresponds to a control domain;

[0060] S4, perform state monitoring on the sub-target ad hoc network in the control domain in a preset monitoring period to obtain state data corresponding to the sub-target ad hoc network;

[0061] S5, perform congestion prediction on the state data based on a pre-trained congestion prediction model to obtain a congestion level label corresponding to the sub-target ad hoc network;

[0062] S6, construct a congestion control model based on the state data and the congestion level label corresponding to the sub-target ad hoc network, and solve the congestion control model based on a reinforcement learning algorithm to obtain a congestion control scheme corresponding to the sub-target ad hoc network.

[0063] In the present application, the ad hoc network is a kind of wireless network structure, the equipment (node) in the network can establish, maintain and manage the connection by itself, without relying on fixed infrastructure or central control, each node not only acts as a data source, but also can forward the data of other nodes, therefore, the state of node and network topology changes dynamically;Network topology refers to the connection mode between each node in the network, in the Internet of Things, the topology structure describes how the devices (such as sensors, terminals, etc.) connect and interact, the topology may be wired or wireless, or fixed or dynamically changing;Network topology can be represented by graph theory, nodes represent devices or sensors in the network, edges represent communication connections between nodes, graph structure modeling means using the concept of graph theory to represent network topology, network topology is converted into an undirected graph, that is, the connection between nodes has no direction, which represents the relationship between devices for mutual communication;Node label is a kind of identification of node, which may represent the state or category of node, node label probability refers to the probability that each node belongs to a certain category (for example, important node, low-power node, etc.), these labels may affect the processing and communication strategy of node;Penalty threshold table is a preset standard, which defines the threshold corresponding to different penalty coefficients, to help decide how to divide the network, if the penalty coefficient of a node exceeds the threshold, it may be divided into different sub-networks (or control domains);Control domain refers to a certain area or subnetwork in the network, which is usually composed of nodes with certain common characteristics (such as similar functions, tasks or distances), each control domain has a certain autonomy inside, which can centrally process control tasks, monitor data, etc.;Sub-target ad hoc network is the sub-network after the division of target ad hoc network, each sub-target ad hoc network represents the device set of a certain area or sub-domain in the network, these sub-networks or sub-target ad hoc networks are usually divided according to control domain, and the nodes inside each sub-target ad hoc network can independently transmit and manage data;Congestion prediction model is a prediction model based on historical data and current state of network, which is used to predict the network congestion that may occur in a certain period of time in the future;Monitoring period refers to the time interval of the system for periodically monitoring the state of a certain network domain (such as sub-target ad hoc network in control domain);Congestion control model is a control algorithm or strategy, which is used to adjust resource allocation, load balancing, routing strategy, etc. in the network, so as to alleviate or solve the problem of network congestion, by adjusting the state of network (such as bandwidth, delay, load, etc.) in real time, the transmission efficiency of network can be improved;Reinforcement learning is a kind of machine learning method, in which the agent learns how to maximize the cumulative reward by selecting actions through interaction with the environment.

[0064] In the second embodiment, the ad hoc network information processing method for massive Internet of Things devices further includes: the network undirected graph includes a node set and an edge set, wherein the nodes in the node set correspond to Internet of Things devices, and the edges in the edge set correspond to communication links between two Internet of Things devices.

[0065] In this embodiment, the undirected graph is a graph structure in which each edge has no direction, meaning that the two endpoints of the edge are equal, the nodes represent devices, and the edges represent communication links between devices. The characteristic of the undirected graph is that if node A has an edge with node B, then node B can also have the same communication with node A without directional restrictions.

[0066] In an optional embodiment, the formula for calculating the node weight of each node in the network undirected graph is as follows:

[0067]

[0068] wherein W i represents the node weight of the i-th node in the network undirected graph, TL i represents the maximum value of the delay between the i-th node and all its neighbor nodes in the network undirected graph, and v j represents the j-th node in the network undirected graph, N S represents the neighbor node set of the i-th node in the network undirected graph, d ij represents the distance between the i-th node and the j-th node in the network undirected graph, and p represents the electromagnetic wave propagation speed.

[0069] In an optional embodiment, the formula for calculating the node label probability of each node in the network undirected graph is as follows:

[0070]

[0071] wherein l i represents the node label, and l i = 0, 1, P(l i = 1) represents the node label probability of the node label l i = 1, and P(l i = 0) represents the node label probability of the node label l i = 0, and P(l i = 0) = 1 - P(l i = 1), W min and W max represent the minimum and maximum values of the node weight.

[0072] In an optional embodiment, the penalty coefficient of each node in the network undirected graph is calculated based on the node label probability, including:

[0073] A1, constructing a node association function based on the node weight, wherein the node association function is as follows:

[0074]

[0075] wherein, AC(i, j) represents the node correlation degree between two nodes in the network undirected graph, η represents the distance coefficient, and Rt represents the node communication range, and σ represents the standard deviation of the node weight;

[0076] A2, constructing a node correlation penalty function based on the node correlation function, wherein the correlation penalty function is as follows:

[0077]

[0078] wherein, ACP i represents the node correlation penalty coefficient of the i-th node in the network undirected graph, V represents the node set of the network undirected graph, δ(l i ,l j ) represents the correlation indicator function, and

[0079] A3, constructing a node self-penalty function based on the node label probability, wherein the node self-penalty function is as follows:

[0080] SCP i = -ln(P(l i = l j ));

[0081] wherein, SCP i represents the node self-penalty coefficient of the i-th node in the network undirected graph, and P(l i = l j ) represents the probability that the node label of the i-th node in the network undirected graph belongs to the real node label;

[0082] A4, constructing a penalty function of the network undirected graph based on the node self-penalty function and the node correlation penalty function, wherein the penalty function of the network undirected graph is as follows:

[0083] LF i = β·SCP i + ACP i ;

[0084] wherein, LF i represents the penalty coefficient of the i-th node in the network undirected graph, and β represents a preset important factor.

[0085] In an optional embodiment, the expression of the state data corresponding to the sub-target ad hoc network is as follows:

[0086] SD = {RTT avg , Ack num , Thr avg , cwnd cl , v, RTT, MinRTT, Jittere ,loss};

[0087] Among them, SD represents the state data corresponding to the sub-target self-organizing network, RTT avg Indicates the average packet delay during the monitoring period, Ack num Indicates the number of valid confirmation messages within the monitoring period, Thr avg Indicates the average throughput during the monitoring period, cwnd cl Indicates the current congestion window size during the monitoring period, v indicates the data sending rate at the end of the monitoring period, RTT indicates the last RTT during the monitoring period, MinRTT indicates the minimum RTT within the recent preset time, and Jitter e It represents the round-trip delay jitter value, and loss represents the packet loss rate of the communication link.

[0088] In an optional embodiment, congestion prediction is performed on the state data based on a pre-trained congestion prediction model to obtain a congestion level label corresponding to the sub-target ad hoc network, including:

[0089] B1. Extract features from the state data to obtain a congestion feature set, where the congestion feature set is expressed as follows:

[0090]

[0091] Among them, FS represents the congestion feature set;

[0092] B2. Perform congestion prediction on the congestion feature set based on a pre-trained congestion prediction model to obtain a congestion level label corresponding to the sub-target ad hoc network, wherein the congestion prediction model adopts a support vector machine model.

[0093] It should be noted that support vector machines are a common supervised learning algorithm widely used in classification and regression problems. In congestion prediction models, support vector machines are used to perform pattern recognition and classification based on training data. The basic idea of ​​support vector machines is to separate data points of different categories by constructing an optimal hyperplane. In congestion prediction, support vector machines can classify network status into different categories (such as normal, light congestion, and heavy congestion) based on extracted congestion features. Congestion level labels are classification labels for network congestion status, indicating the current congestion state of the network (such as no congestion, light congestion, moderate congestion, and heavy congestion).

[0094] In an optional embodiment, a congestion control model is constructed based on the state data and congestion level labels corresponding to the sub-target ad hoc networks, including:

[0095] C1. Construct the state space of the congestion control model based on the state data corresponding to the sub-target ad hoc network;

[0096] C2, constructing a reward function of the congestion control model based on the state data corresponding to the sub-target ad hoc network and the congestion level label;

[0097] C3, constructing an action space of the congestion control model based on all communication links of the sub-target ad hoc network.

[0098] It should be noted that in reinforcement learning, the action space refers to the set of all possible actions that an agent can take, and in the congestion control model, the action space represents all operations that the model can choose to control network congestion, which are usually defined by control strategies in the network.

[0099] In an optional embodiment, the expression of the state space of the congestion control model is as follows:

[0100] s t =(SD t ,SD t-1 ,...,SD t-k );

[0101] wherein s t represents the state space of the congestion control model, SD t represents the state data corresponding to the sub-target ad hoc network at time t.

[0102] The reward function of the congestion control model is as follows:

[0103]

[0104] wherein reward represents the reward function of the congestion control model, p represents the congestion level corresponding to the sub-target ad hoc network, and p th represents a preset congestion level threshold.

[0105] In an optional embodiment, the congestion control model is solved based on a reinforcement learning algorithm to obtain the congestion control scheme corresponding to the sub-target ad hoc network, including:

[0106] D1, selecting an action from the action space according to the policy π and executing in the state space environment through the monitored state space s t , so as to receive the reward reward and enter the next state space s t+1 , wherein the policy π is a mapping from the state space s t to the action space;

[0107] D2, learning based on the reinforcement learning algorithm to maximize the cumulative reward to obtain the policy π corresponding to the maximum reward, wherein the policy π is the congestion control scheme corresponding to the sub-target ad hoc network, and the learning formula of the cumulative reward is as follows:

[0108]

[0109] wherein R t represents the cumulative reward, and T represents the number of times of iterative execution of the reinforcement learning algorithm.

[0110] It should be noted that the reinforcement learning algorithm is an algorithm for training an agent to learn a strategy in a specific environment. Common reinforcement learning algorithms include Q-learning, deep Q network (DQN), and policy gradient method. The core idea of these algorithms is to learn through trial and error and gradually adjust the strategy in the process of interacting with the environment, so as to maximize the cumulative reward. The strategy refers to a decision rule for network congestion control. In a sub-target ad hoc network, the strategy determines how to optimize network performance by controlling network traffic, routing selection, bandwidth allocation, etc., so as to avoid or reduce network congestion. Reinforcement learning continuously optimizes these strategies to achieve the optimal congestion control effect.

[0111] The embodiments of the present application are described in detail above in combination with the drawings, but the present application is not limited thereto. Various changes can be made within the knowledge of those skilled in the art without departing from the spirit of the present application.

Claims

1. A method for processing information of a self-organizing network of massive IoT devices, characterized in that: include: Acquire a network topology structure of a target ad hoc network, and perform graph structure modeling on the network topology structure to obtain a network undirected graph corresponding to the network topology structure; Calculating a node weight of each node in the network undirected graph based on the node position of each node in the network undirected graph, and calculating a node label probability of each node in the network undirected graph based on the node weight; Calculating a penalty coefficient for each node in the network undirected graph based on the node label probability, and comparing the penalty coefficient of each node with a preset penalty threshold table to divide the target ad hoc network into a plurality of control domains, that is, dividing the target ad hoc network into a plurality of sub-target ad hoc networks, and the sub-target ad hoc networks correspond to the control domains; Taking the control domain as a unit, monitoring the status of the sub-target ad hoc network within the control domain according to a preset monitoring period to obtain status data corresponding to the sub-target ad hoc network; Performing congestion prediction on the state data based on a pre-trained congestion prediction model to obtain a congestion level label corresponding to the sub-target ad hoc network; A congestion control model is constructed based on the state data and congestion level label corresponding to the sub-target self-organizing network, and the congestion control model is solved based on a reinforcement learning algorithm to obtain a congestion control solution corresponding to the sub-target self-organizing network.

2. The method for processing information of a self-organizing network for massive IoT devices according to claim 1, characterized in that: The network undirected graph includes a node set and an edge set, wherein the nodes in the node set correspond to Internet of Things devices, and the edges in the edge set correspond to communication links between two Internet of Things devices.

3. The method for processing information of a self-organizing network for massive IoT devices according to claim 2, characterized in that: The calculation formula of the node weight of each node in the network undirected graph is as follows: Among them, W i Represents the node weight of the i-th node in the network undirected graph, TL i represents the maximum value of the delay between the i-th node and all its neighboring nodes in the network undirected graph, and v j Represents the jth node in the network undirected graph, N S Represents the set of neighbor nodes of the i-th node in the network undirected graph, d ij It represents the distance between the i-th node and the j-th node in the network undirected graph, and ρ represents the propagation speed of electromagnetic waves.

4. The method for processing information of a self-organizing network for massive Internet of Things devices according to claim 3, characterized in that: The calculation formula for the node label probability of each node in the undirected network graph is as follows: Among them, l i represents the node label, and l i =0,1,P(l i =1) indicates node label l i The node label probability is 1, and P(l i =0) indicates node label l i The node label probability is 0, and P(l i =0) = 1-P(l i =1), W min and W max Indicates the minimum and maximum values ​​of the node weight.

5. The method for processing information of a self-organizing network for massive Internet of Things devices according to claim 4, characterized in that: Calculating a penalty coefficient for each node in the network undirected graph based on the node label probability includes: A node association function is constructed based on the node weights, wherein the node association function is as follows: Among them, AC(i,j) represents the node association between two nodes in the network undirected graph, η represents the distance coefficient, and Rt represents the communication range of the node, σ represents the standard deviation of the node weight; A node association penalty function is constructed based on the node association function, wherein the association penalty function is as follows: Among them, ACP i represents the node association penalty coefficient of the i-th node in the network undirected graph, V represents the node set of the network undirected graph, δ(l i ,l j ) represents the correlation indicator function, and A node penalty function is constructed based on the node label probability, wherein the node penalty function is as follows: SCP i =-ln(P(l i =l j )); Among them, SCP i It represents the node penalty coefficient of the i-th node in the network undirected graph, P(l i =l j ) represents the probability that the node label of the i-th node in the network undirected graph belongs to the real node label; A penalty function of the network undirected graph is constructed based on the node self-penalty function and the node association penalty function, wherein the penalty function of the network undirected graph is as follows: LF i =β·SCP i +ACP i ; Among them, LF i It represents the penalty coefficient of the i-th in the network undirected graph, and β represents the preset important factor.

6. The method for processing information of a self-organizing network for massive IoT devices according to claim 1, characterized in that: The expression of the state data corresponding to the sub-target ad hoc network is as follows: SD={RTT avg ,Ack num ,Thr avg ,cwnd cl ,v,RTT,MinRTT,Jitter e ,loss}; Among them, SD represents the state data corresponding to the sub-target self-organizing network, RTT avg Indicates the average packet delay during the monitoring period, Ack num Indicates the number of valid confirmation messages within the monitoring period, Thr avg Indicates the average throughput during the monitoring period, cwnd cl Indicates the current congestion window size during the monitoring period, v indicates the data sending rate at the end of the monitoring period, RTT indicates the last RTT during the monitoring period, MinRTT indicates the minimum RTT within the recent preset time, and Jitter e It represents the round-trip delay jitter value, and loss represents the packet loss rate of the communication link.

7. The method for processing information of a self-organizing network for massive IoT devices according to claim 6, characterized in that: Performing congestion prediction on the state data based on a pre-trained congestion prediction model to obtain a congestion level label corresponding to the sub-target ad hoc network includes: Feature extraction is performed on the state data to obtain a congestion feature set, wherein the expression of the congestion feature set is as follows: Among them, FS represents the congestion feature set; Congestion prediction is performed on the congestion feature set based on a pre-trained congestion prediction model to obtain a congestion level label corresponding to the sub-target ad hoc network, wherein the congestion prediction model adopts a support vector machine model.

8. The method for processing information of a self-organizing network for massive IoT devices according to claim 7, characterized in that: Constructing a congestion control model based on the state data and congestion level label corresponding to the sub-target ad hoc network, including: Constructing the state space of the congestion control model based on the state data corresponding to the sub-target ad hoc network; Constructing a reward function of the congestion control model based on the state data and congestion level label corresponding to the sub-target ad hoc network; The action space of the congestion control model is constructed based on all communication links of the sub-target ad hoc network.

9. The method for processing information of a self-organizing network for massive Internet of Things devices according to claim 8, characterized in that: The state space expression of the congestion control model is as follows: s t =(SD t ,SD t-1 ,...,SD t-k ); Among them, s t Represents the state space of the congestion control model, SD t Represents the state data corresponding to the sub-target ad hoc network at time t; The reward function of the congestion control model is as follows: Among them, reward represents the reward function of the congestion control model, p represents the congestion level corresponding to the sub-target ad hoc network, and p th Indicates the preset congestion level threshold.

10. The method for processing information of a self-organizing network for massive IoT devices according to claim 9, characterized in that: Solving the congestion control model based on a reinforcement learning algorithm to obtain a congestion control solution corresponding to the sub-target ad hoc network includes: Through the monitored state space s t , select an action from the action space according to the strategy π and execute it in the state space environment, thereby receiving a reward and entering the next state space s t+1 , the policy π is the state space s t Mapping to action space; Based on the reinforcement learning algorithm, the accumulated reward is maximized to obtain the strategy π corresponding to the maximum reward, wherein the strategy π is the congestion control scheme corresponding to the sub-goal self-organizing network. The learning formula of the accumulated reward is as follows: Among them, E t represents the cumulative reward, and T represents the number of iterations of the reinforcement learning algorithm.

Images