Satellite communication network flow management method and system based on big data analysis

Through big data analysis and reinforced learning technology, satellite communication network traffic is dynamically optimized, and the problems of traditional solutions are solved, intelligent management and optimization of network traffic is realized, and the stability and responsiveness of communication services are ensured.

CN119093999APending Publication Date: 2024-12-06KEYIDEA SATCOM INFORMATION TECH (NANJING) CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411246739.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-06
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

Traditional satellite communication traffic management solutions rely on preset thresholds and empirical judgments, resulting in lagging responses and rigid resource allocation, and being unable to effectively adapt to the growing data transmission needs and complex communication environments, resulting in network failure and degradation of service quality.

Method used

Using a method based on big data analysis, the user entity network parameters in the satellite communication network are obtained, behavioral characteristics are extracted, and the timing prediction model is used to predict future traffic demands, and the traffic control strategy is determined through reinforcement learning, including bandwidth allocation adjustment and priority management. and user notifications to dynamically optimize network traffic management.

Benefits of technology

Real-time monitoring and optimization of network traffic is achieved, network congestion is avoided, communication services are ensured to the stability of peak periods, rapid response capabilities to emergencies, and resource utilization efficiency is improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119093999A_ABST
    Figure CN119093999A_ABST
Patent Text Reader

Abstract

The invention provides a satellite communication network flow management method and system based on big data analysis. The method comprises the following steps: acquiring network parameters of each user entity in a satellite communication network; extracting behavior characteristics of the user entity according to the network parameters of the user entity; classifying the user entities according to the behavior characteristics of the user entities; utilizing a pre-trained time sequence prediction model to predict the flow required by the user entity at the future moment; and obtaining a flow control strategy for the user entity according to the predicted flow required by the user entity at the future moment and the category of the user entity, and executing the flow control strategy. According to the method, the network traffic of satellite communication can be monitored in real time, and the network traffic management strategy is dynamically scheduled and intelligently optimized so as to adapt to different traffic requirements.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of satellite communication technology, and in particular to a satellite communication network traffic management method and system based on big data analysis. Background Art

[0002] In the field of satellite communications, traditional traffic management solutions mostly rely on preset threshold rules and operator experience. This solution is highly dependent on personnel experience and cannot make immediate and accurate responses. When emergencies occur, this delayed processing method can easily lead to network paralysis, affecting communication quality and user experience. In addition, the rigid resource allocation problem common in traditional methods also increases unnecessary costs for network operations.

[0003] Since traditional traffic management technology cannot adapt to the growing data transmission needs and complex and changing communication environment, especially when dealing with massive data and complex network scheduling, how to effectively manage and optimize the allocation of satellite communication network traffic resources has become an urgent problem to be solved. Summary of the invention

[0004] Purpose of the invention: The present invention aims to overcome the above-mentioned technical problems and proposes a satellite communication network traffic management method and system based on big data analysis, which can monitor the network traffic of satellite communications in real time, dynamically schedule and intelligently optimize the network traffic management strategy to adapt to different traffic requirements.

[0005] Invention content: To achieve the above objectives, the present invention proposes the following technical solutions:

[0006] In a first aspect, a satellite communication network traffic management method based on big data analysis is provided, the method comprising:

[0007] Obtaining network parameters of each user entity in the satellite communication network;

[0008] extracting behavior characteristics of the user entity according to network parameters of the user entity;

[0009] classifying the user entity according to the behavioral characteristics of the user entity;

[0010] Using a pre-trained time series prediction model to predict the traffic required by the user entity at a future time;

[0011] According to the predicted flow required by the user entity at a future time and the category of the user entity, a flow control strategy for the user entity is obtained, and the flow control strategy is executed.

[0012] As an optional implementation manner of the method of the first aspect, the network parameters of the user entity include signal strength, transmission rate, user request data volume and service quality parameters; extracting the behavior characteristics of the user entity according to the network parameters of the user entity specifically includes:

[0013] The principal component analysis method is used to extract features from the network parameters of the user entity to obtain the behavioral features of the user entity.

[0014] As an optional implementation of the method of the first aspect, using a pre-trained time series prediction model to predict the traffic required by the user entity at a future time specifically includes:

[0015] Obtaining traffic usage history data of the user entity;

[0016] The traffic usage history data of the user entity is input into the time series prediction model to obtain the traffic required by the user entity at a future moment.

[0017] As an optional implementation manner of the method of the first aspect, obtaining a traffic control strategy for the user entity according to the predicted traffic required by the user entity at a future time and the behavior characteristics of the user entity specifically includes:

[0018] Determining a reward mechanism according to the predicted traffic required by the user entity at a future time and the behavior characteristics of the user entity;

[0019] By adopting a reinforcement learning method, taking the traffic usage in the satellite communication network as the environment, a traffic control strategy for the user entity is found under the reward mechanism.

[0020] As an optional implementation manner of the method described in the first aspect, the traffic control strategy includes at least one of bandwidth allocation adjustment, priority management, user notification and policy negotiation.

[0021] In a second aspect, a satellite communication network traffic management system based on big data analysis is provided, the system comprising:

[0022] a data acquisition module configured to obtain network parameters of each user entity in the satellite communication network;

[0023] a data analysis module configured to extract behavior characteristics of the user entity according to the network parameters of the user entity, and classify the user entity according to the behavior characteristics of the user entity;

[0024] A prediction decision module, configured to use a pre-trained time series prediction model to predict the traffic required by the user entity at a future moment;

[0025] The flow control module is configured to obtain a flow control strategy for the user entity according to the predicted flow required by the user entity at a future moment and the category of the user entity, and execute the flow control strategy.

[0026] As an optional implementation manner of the system described in the second aspect, the network parameters of the user entity include signal strength, transmission rate, user requested data volume and service quality parameters; the data analysis module is specifically used to use principal component analysis to extract features of the network parameters of the user entity to obtain behavioral characteristics of the user entity.

[0027] As an optional implementation of the system described in the second aspect, the prediction decision module is specifically used to obtain the traffic usage history data of the user entity; input the traffic usage history data of the user entity into the time series prediction model to obtain the traffic required by the user entity at a future time.

[0028] As an optional implementation manner of the system described in the second aspect, the traffic control module is specifically used to determine a reward mechanism based on the predicted traffic required by the user entity at a future moment and the behavioral characteristics of the user entity; using reinforcement learning, with the traffic usage in the satellite communication network as the environment, to find a traffic control strategy for the user entity under the reward mechanism.

[0029] As an optional implementation manner of the system described in the second aspect, the traffic control strategy adopted by the traffic control module includes at least one of bandwidth allocation adjustment, priority management, user notification and policy negotiation.

[0030] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0031] The satellite communication network traffic management method based on big data analysis proposed by the present invention deeply analyzes the traffic usage behavior characteristics of various user entities in the satellite communication network, and predicts the data traffic trend of the satellite communication network based on the traffic usage behavior characteristics of each type of user entity, so that the corresponding traffic control strategy can be specified in advance. Based on this method, the entire satellite communication network can automatically adjust the data routing and reasonably allocate bandwidth resources, avoiding network congestion and service quality degradation caused by improper resource allocation, ensuring that stable communication services can still be provided during peak hours, and enhancing the satellite communication network's ability to respond quickly to emergencies. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1 It is a flow chart of a satellite communication network traffic management method based on big data analysis involved in an embodiment of this specification.

[0034] Figure 2 It is a structural diagram of a satellite communication network traffic management system based on big data analysis involved in the embodiments of this specification. DETAILED DESCRIPTION

[0035] First of all, it should be noted that 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", "the" 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 otherwise.

[0036] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Therefore, it should be recognized by those skilled in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0037] It should be noted that: in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the steps included in the method may be more or less than those described in this specification. In addition, a single step described in this specification may be decomposed into multiple steps for description in other embodiments; and multiple steps described in this specification may be combined into a single step for description in other embodiments.

[0038] This specification aims to address the problems of delayed response capability and rigid resource allocation in current traditional traffic management solutions, and proposes a satellite communication network traffic management method and system based on big data analysis to at least partially overcome the above-mentioned defects.

[0039] The following will further describe in detail a satellite communication network traffic management method and system based on big data analysis described in one or more embodiments of this specification in combination with the drawings and specific embodiments of the specification, but this detailed description does not constitute a limitation on the embodiments of this specification.

[0040] Please refer to Figure 1 , Figure 1 The following is a schematic diagram of a flow chart of a satellite communication network traffic management method based on big data analysis. Figure 1 As shown, the method includes steps S100 to S108.

[0041] S100: Acquire network parameters of each user entity in the satellite communication network.

[0042] The user entities mentioned above refer to individual users, corporate users or any entity that uses satellite communication networks for data transmission. They can be personal computers, mobile phones, IoT devices, etc. that access the Internet through satellite communication networks.

[0043] The above network parameters include but are not limited to signal strength, transmission rate, user requested data volume, service quality parameters, etc.

[0044] It should be noted that the network parameters obtained in this implementation can be selected according to needs, and this implementation does not impose any restrictions on this.

[0045] S102: Extracting behavior characteristics of the user entity according to the network parameters of the user entity.

[0046] After the required network parameters of the user entity are obtained, the behavior characteristics of the user entity in terms of traffic usage can be analyzed based on the network parameters of the user entity.

[0047] In this embodiment, the analysis method for analyzing the behavioral characteristics of user entities can be adaptively selected according to needs, and this embodiment does not impose any restrictions on this. For example, a neural network model and a feature analysis algorithm can be used to extract features of the network parameters of the above-mentioned user entities, thereby extracting the behavioral characteristics of the user entities.

[0048] The following takes the principal component analysis method as an example to explain the specific steps of extracting the behavioral characteristics of the user entity from the network parameters of the user entity.

[0049] First, the network parameter vector V of the user entity is constructed, V = [S, R, D, Q], where S represents signal strength, R represents transmission rate, D represents the amount of user requested data, and Q represents service quality parameters.

[0050] Next, perform principal component analysis on the above network parameter vector V, that is, calculate:

[0051] C=PCA(V)

[0052] Where C is the principal component score, and PCA(V) is a conversion function used to convert the original data variable V (i.e., the network parameters of the user entity mentioned above) into independent principal components. The principal components represent the key characteristics of the data and provide decision support for traffic management.

[0053] S104: Classify the user entity according to the behavior characteristics of the user entity.

[0054] This step S104 aims to analyze the behavior patterns of these different user entities in the satellite communication network, such as data transmission frequency, data volume, service type, etc., so as to better analyze the actual demand and usage of user entities for network traffic in the satellite communication network, thereby providing more accurate and efficient traffic management strategies for each type of user entity.

[0055] Based on this purpose, any classification method that meets the above purpose can be used to classify user entities, and this embodiment does not limit this.

[0056] For example, the k-means clustering algorithm can be used to classify based on the behavioral characteristics of user entities. The calculation formula of the k-means clustering algorithm is as follows:

[0057] K = kmeans(C,n)

[0058] Where K represents the clustering result, kmeans(C,n) represents the clustering function, C is the principal component score obtained from step S102, and n is the preset number of clusters. This algorithm can divide user behaviors into n groups, providing a basis for traffic management strategies for different groups.

[0059] The choice of n can be determined according to actual needs. For example, if we are looking at a satellite communication network, we know from historical experience that the satellite communication network mainly serves three types of users: personal users, corporate users, and IoT devices. Then, we can set n to 3 and directly classify and cluster the data.

[0060] For another example, for a satellite communication network, you can first use data visualization technology (such as scatter plots) to visualize and analyze the historical traffic usage data in the satellite communication network, and draw a scatter plot of user data request size and time. If several clustered areas can be clearly distinguished in the graph, these areas may represent different user behavior patterns, and a preliminary number of clusters can be set accordingly.

[0061] S106: Use the pre-trained time series prediction model to predict the traffic required by the user entity at a future moment.

[0062] The above time series prediction model is trained using the historical traffic usage data of the satellite communication network. The specific training steps are as follows:

[0063] In the historical traffic usage data, traffic usage data for a continuous period of time is sampled as a training sample, and the traffic usage data at the next moment of the continuous period of time is used as the label of the training sample.

[0064] The training samples are input into the time series prediction model to obtain the predicted traffic usage data at the next moment.

[0065] According to the predicted traffic usage data at the next moment and the actual traffic usage data at the next moment, a loss function is constructed to train the above-mentioned time series prediction model until a time series prediction model that meets the preset conditions is obtained.

[0066] The time series prediction model pre-trained based on the above training process can predict the traffic data required by the user entity at a future moment according to the collected current traffic usage data of the user entity.

[0067] Therefore, in this step S106, for each type of user entity, a pre-trained time series prediction model may be used to predict the traffic required by the type of user entity at a future time.

[0068] The above time series prediction model can be selected according to the needs. For example, an autoregressive moving average (ARMA) model can be used. The expression of the autoregressive moving average (ARMA) model is:

[0069]

[0070] Among them, X(t) is the predicted flow at time t, and θ1…θ q is the model parameter, ε(t) is the white noise term, and the model calculates the weighted sum of past traffic data to predict future traffic.

[0071] S108: Obtain a flow control strategy for the user entity according to the predicted flow required by the user entity at a future time and the category of the user entity, and execute the flow control strategy.

[0072] By analyzing the behavioral characteristics of user entities and the predicted future traffic of this type of user entity, a classification decision algorithm can be used to generate specific traffic control strategies for different user entity groups. This process aims to optimize network resource utilization, improve service quality, and prevent potential traffic congestion problems.

[0073] In some implementations, a reinforcement learning strategy may be used to generate a traffic control strategy for each type of user entity:

[0074] First, the reward mechanism is determined based on the predicted traffic required by the user entity in the future and the behavioral characteristics of this type of user entity.

[0075] Then, reinforcement learning is adopted to find the traffic control strategy for this type of user entity under the reward mechanism, taking the traffic usage in the satellite communication network as the environment.

[0076] The reinforcement learning process can be expressed as follows:

[0077] Q(s,a)=Q(s,a)+α[r+γmaxQ(s′,a′)-Q(s,a)]

[0078] Among them, Q() is the incentive function, which can be determined based on the predicted traffic required by the user entity in the future and the behavioral characteristics of this type of user entity. Q(s, a) is the expected reward of taking action a in state s, r is the immediate reward, α is the learning rate, γ is the decay factor, and maxQ(s', a') is the maximum expected reward for the next state. The reinforcement learning algorithm maximizes the traffic control effect by continuously iteratively updating and optimizing the Q value.

[0079] Specifically, in reinforcement learning, an action a is chosen by a policy π, a function or mapping that decides what action to take based on the current state of the environment s. Since it is determined by the policy, there is no specific formula for the action itself. However, we can express a definition of a policy π, which can be deterministic or stochastic.

[0080] For a deterministic strategy, action a can be directly determined by state s, which can be written in the form:

[0081] a=π(s)

[0082] This means that for each state s, the policy π specifies a certain action a.

[0083] For a stochastic strategy, the choice of action a is based on a certain probability distribution, which can be expressed in the following form:

[0084] P(a|s)=π(a|s)

[0085] Here, π(a|s) represents the probability of choosing action a in state s. Stochastic strategies are often used to explore the environment and find the optimal strategy, because it allows the algorithm to try a variety of different actions and learn the best action strategy from them.

[0086] When using reinforcement learning strategies to find traffic control policies for each type of user entity, a stochastic strategy is usually adopted in the initial stage for exploration, and then gradually shifts to a more deterministic strategy to exploit the discovered optimal actions.

[0087] The above-mentioned traffic control strategy may specifically include at least one of bandwidth allocation adjustment, priority management, user notification and strategy negotiation.

[0088] For example, suppose that through traffic forecasting, it is known that during the peak time period of each day (such as 8pm to 10pm), the data demand of a certain type of device A will suddenly increase. Based on this forecast result, the following traffic control strategy can be generated:

[0089] (1) Bandwidth allocation adjustment: During predicted high-demand periods, more bandwidth resources are allocated in advance to Category A devices to meet their increased data demands.

[0090] (2) Priority management: Traffic is classified and managed to ensure that critical services (such as emergency communications) have sufficient bandwidth at all times. At the same time, some non-emergency service traffic may be temporarily lowered in priority.

[0091] (3) User notification and policy negotiation: Send advance notifications to users of Category A devices to inform them of the possible risk of network congestion and recommended usage times, or provide traffic usage discount policies to encourage users to use the network during off-peak hours and reduce the load during peak hours.

[0092] By continuously monitoring, executing and adjusting traffic control strategies, parameters can be dynamically adjusted and traffic management strategies can be updated according to changes in actual satellite communication network traffic and system performance feedback to ensure network service quality and efficient resource utilization.

[0093] The above is Figure 1 The satellite communication network traffic management method based on big data analysis is shown. The method determines the traffic usage behavior characteristics of different categories of user entities by performing data analysis on the traffic usage in the satellite communication network, and predicts the future traffic required by different categories of user entities based on this, so as to generate targeted traffic management strategies, so that the entire satellite communication network can automatically adjust data routing and reasonably allocate bandwidth resources, avoiding network congestion and service quality degradation caused by improper resource allocation, ensuring that stable communication services can still be provided during peak periods, and enhancing the satellite communication network's rapid response capability to emergencies.

[0094] Corresponding to the above-mentioned satellite communication network traffic management method based on big data analysis, this embodiment also provides a satellite communication network traffic management system based on big data analysis. It should be noted that the satellite communication network traffic management method based on big data analysis described in this embodiment can be implemented by relying on the satellite communication network traffic management system based on big data analysis, but is not limited to the satellite communication network traffic management system based on big data analysis.

[0095] Please refer to Figure 2 , Figure 2 The following schematically shows a structural diagram of a satellite communication network traffic management system based on big data analysis. Figure 2 As shown, the system includes:

[0096] The data acquisition module 201 is configured to obtain network parameters of each user entity in the satellite communication network.

[0097] The data analysis module 202 is configured to extract the behavior characteristics of the user entity according to the network parameters of the user entity, and classify the user entity according to the behavior characteristics of the user entity.

[0098] The prediction decision module 203 is configured to use a pre-trained time series prediction model to predict the traffic required by the user entity at a future moment.

[0099] The traffic control module 204 is configured to obtain a traffic control strategy for the user entity according to the predicted traffic required by the user entity at a future time and the category of the user entity, and execute the traffic control strategy.

[0100] For the above-mentioned data acquisition module 201, the model is specifically used to connect with each data interface in the satellite communication network and collect network parameters of the user entity in real time. The data acquisition module 201 should have high compatibility with satellite communication network equipment and can adapt to different satellite communication protocols and standards. The data acquisition module 201 uses high-precision sensors and advanced data receiving technology to ensure the accuracy and timeliness of data acquisition.

[0101] The user entities mentioned above refer to individual users, corporate users or any entity that uses satellite communication networks for data transmission. They can be personal computers, mobile phones, IoT devices, etc. that access the Internet through satellite communication networks.

[0102] The above network parameters include but are not limited to signal strength, transmission rate, user requested data volume, service quality parameters, etc.

[0103] It should be noted that the network parameters obtained in this implementation can be selected according to needs, and this implementation does not impose any restrictions on this.

[0104] Regarding the above-mentioned data analysis module 202, this module is mainly used to analyze the behavior characteristics of the user entity in terms of traffic usage according to the network parameters of the user entity.

[0105] The data analysis module 202 can adaptively select the analysis method for analyzing the behavioral characteristics of the user entity according to the needs, and this embodiment does not impose any restrictions on this. For example, the data analysis module 202 can use a neural network model and a feature analysis algorithm to extract features of the network parameters of the above-mentioned user entity, thereby extracting the behavioral characteristics of the user entity.

[0106] The following takes the principal component analysis method as an example to explain the specific steps of the data analysis module 202 extracting the behavior characteristics of the user entity from the network parameters of the user entity.

[0107] First, the network parameter vector V of the user entity is constructed, V = [S, R, D, Q], where S represents signal strength, R represents transmission rate, D represents the amount of user requested data, and Q represents service quality parameters.

[0108] Next, perform principal component analysis on the above network parameter vector V, that is, calculate:

[0109] C=PCA(V)

[0110] Where C is the principal component score, and PCA(V) is a conversion function used to convert the original data variable V (i.e., the network parameters of the user entity mentioned above) into independent principal components. The principal components represent the key characteristics of the data and provide decision support for traffic management.

[0111] The above-mentioned data analysis module 202 is also used to classify user entities according to the behavioral characteristics of the user entities. In some embodiments, the data analysis module 202 can use the k-means clustering algorithm to classify based on the behavioral characteristics of the user entities. The calculation formula of the k-means clustering algorithm is as follows:

[0112] K = kmeans(C,n)

[0113] Where K represents the clustering result, kmeans(C,n) represents the clustering function, C is the principal component score obtained from step S102, and n is the preset number of clusters. This algorithm can divide user behaviors into n groups, providing a basis for traffic management strategies for different groups.

[0114] Regarding the above-mentioned prediction decision module 203, this module is specifically used to collect the current traffic usage data of each type of user entity, and then use the pre-trained time series prediction model to predict the traffic data required by the user entity in the future.

[0115] The above time series prediction model can be implemented using an autoregressive moving average (ARMA) model, and the expression of the autoregressive moving average (ARMA) model is:

[0116]

[0117] Among them, X(t) is the predicted flow at time t, and θ1…θ q is the model parameter, ε(t) is the white noise term, and the model calculates the weighted sum of past traffic data to predict future traffic.

[0118] Regarding the above-mentioned traffic control module 204, this module is specifically used to determine the reward mechanism based on the predicted traffic required by the user entity in the future and the behavioral characteristics of the user entity; using reinforcement learning, taking the traffic usage in the satellite communication network as the environment, under the reward mechanism, find the traffic control strategy for the user entity.

[0119] The reinforcement learning process can be expressed as follows:

[0120] Q(s,a)=Q(s,a)+α[r+γmaxQ(s′,a′)-Q(s,a)]

[0121] Among them, Q() is the incentive function, which can be determined based on the predicted traffic required by the user entity in the future and the behavioral characteristics of this type of user entity. Q(s, a) is the expected reward of taking action a in state s, r is the immediate reward, α is the learning rate, γ is the decay factor, and maxQ(s', a') is the maximum expected reward for the next state. The reinforcement learning algorithm maximizes the traffic control effect by continuously iteratively updating and optimizing the Q value.

[0122] Specifically, in reinforcement learning, an action a is chosen by a policy π, a function or mapping that decides what action to take based on the current state of the environment s. Since it is determined by the policy, there is no specific formula for the action itself. However, we can express a definition of a policy π, which can be deterministic or stochastic.

[0123] For a deterministic strategy, action a can be directly determined by state s, which can be written in the form:

[0124] a=π(s)

[0125] This means that for each state s, the policy π specifies a certain action a.

[0126] For a stochastic strategy, the choice of action a is based on a certain probability distribution, which can be expressed in the following form:

[0127] P(a|s)=π(a|s)

[0128] Here, π(a|s) represents the probability of choosing action a in state s. Stochastic strategies are often used to explore the environment and find the optimal strategy, because it allows the algorithm to try a variety of different actions and learn the best action strategy from them.

[0129] When the traffic control module 204 adopts a reinforcement learning strategy to find a traffic control strategy for each type of user entity, it usually adopts a random strategy for exploration in the initial stage, and then gradually turns to a more deterministic strategy to utilize the discovered optimal actions.

[0130] The flow control strategy generated by the flow control module 204 may specifically include at least one of bandwidth allocation adjustment, priority management, user notification, and policy negotiation. After determining the flow control strategy, the flow control module 204 is also used to execute the strategy, and specifically, the flow control strategy may be implemented by data routing adjustment, priority adjustment, and bandwidth allocation.

[0131] The traffic control module 204 can dynamically adjust parameters and update traffic management strategies according to changes in actual traffic in the satellite communication network and system performance feedback by continuously monitoring, executing and adjusting traffic control strategies to ensure network service quality and efficient resource utilization.

[0132] It is to be understood that the structure illustrated in the embodiments of this specification does not constitute a specific limitation on the system of the embodiments of this specification. In other embodiments of the specification, the above system may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0133] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0134] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0135] It should be noted that the above examples are only specific embodiments of the present invention, and the present invention is obviously not limited to the above examples, and there are many similar variations. All variations directly derived or associated from the contents disclosed by the technicians in this field should fall within the protection scope of the present invention.

Claims

1. A satellite communication network traffic management method based on big data analysis, characterized in that: include: Obtaining network parameters of each user entity in the satellite communication network; extracting behavior characteristics of the user entity according to network parameters of the user entity; classifying the user entity according to the behavioral characteristics of the user entity; Using a pre-trained time series prediction model to predict the traffic required by the user entity at a future time; According to the predicted flow required by the user entity at a future time and the category of the user entity, a flow control strategy for the user entity is obtained, and the flow control strategy is executed.

2. The method according to claim 1, characterized in that: The network parameters of the user entity include signal strength, transmission rate, user requested data volume, and service quality parameters; extracting the behavior characteristics of the user entity according to the network parameters of the user entity specifically includes: The principal component analysis method is used to extract features from the network parameters of the user entity to obtain the behavioral features of the user entity.

3. The method according to claim 1, characterized in that The pre-trained time series prediction model is used to predict the traffic required by the user entity at a future time, specifically including: Obtaining current traffic usage data of the user entity; The current traffic usage data of the user entity is input into the time series prediction model to obtain the traffic required by the user entity at a future time.

4. The method according to claim 1, characterized in that: According to the predicted flow required by the user entity at a future time and the behavior characteristics of the user entity, a flow control strategy for the user entity is obtained, which specifically includes: Determining a reward mechanism according to the predicted traffic required by the user entity at a future time and the behavior characteristics of the user entity; By adopting a reinforcement learning method, taking the traffic usage in the satellite communication network as the environment, a traffic control strategy for the user entity is found under the reward mechanism.

5. The method according to claim 1, characterized in that The traffic control strategy includes at least one of bandwidth allocation adjustment, priority management, user notification and policy negotiation.

6. Satellite communication network traffic management system based on big data analysis, characterized in that: include: a data acquisition module configured to obtain network parameters of each user entity in the satellite communication network; a data analysis module configured to extract behavior characteristics of the user entity according to the network parameters of the user entity, and classify the user entity according to the behavior characteristics of the user entity; A prediction decision module, configured to use a pre-trained time series prediction model to predict the traffic required by the user entity at a future moment; The flow control module is configured to obtain a flow control strategy for the user entity according to the predicted flow required by the user entity at a future moment and the category of the user entity, and execute the flow control strategy.

7. The system according to claim 6, characterized in that The network parameters of the user entity include signal strength, transmission rate, user requested data volume and service quality parameters; the data analysis module is specifically used to extract features of the network parameters of the user entity using the principal component analysis method to obtain the behavioral characteristics of the user entity.

8. The system according to claim 6, characterized in that The prediction decision module is specifically used to obtain the traffic usage history data of the user entity; input the traffic usage history data of the user entity into the time series prediction model to obtain the traffic required by the user entity at a future moment.

9. The system according to claim 6, characterized in that The traffic control module is specifically used to determine a reward mechanism based on the predicted traffic required by the user entity at a future time and the behavioral characteristics of the user entity; using reinforcement learning, with the traffic usage in the satellite communication network as the environment, to find a traffic control strategy for the user entity under the reward mechanism.

10. The system according to claim 6, characterized in that The traffic control strategy adopted by the traffic control module includes at least one of bandwidth allocation adjustment, priority management, user notification and policy negotiation.

Citation Information

Patent Citations

  • Network resource allocation method and system, storage medium and electronic equipment

    CN117201308A

  • Intelligent connection scheduling method and system for home optical modem

    CN117579520A

  • Dynamic adaptive network traffic management system and method

    CN118055024A

  • A method and system for resisting weak network for audio and video transmission

    CN118101941A

  • Deploying and updating machine learning models over a communication network

    US20230198855A1