Mobility-aware content caching method in satellite internet and related equipment

By utilizing the deterministic mobile characteristics and deep learning models of satellites in the satellite Internet, dynamically adjusting the cache content, solving the problems of limited user connections and insufficient resources, and improving the cache hit rate and user experience.

CN119051718APending Publication Date: 2024-11-29BEIJING UNIV OF POSTS & TELECOMM
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411129362.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-16
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

There are problems in the satellite Internet with limited number of user connections and insufficient resources, which leads to fluctuations in network delays and difficult to optimize edge cache decisions, affecting user experience.

Method used

Through the deterministic mobile characteristics of the satellite, handover information and user clusters are pre-determined, and a one-dimensional convolutional neural network and deep reinforcement learning model is used to dynamically adjust the cache content, and optimize cache decisions to improve cache hit rate and reduce latency.

Benefits of technology

It improves the cache hit rate of satellite Internet edge cache, reduces transmission delay, improves user experience, and solves the main problems of satellite Internet edge cache decision optimization.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119051718A_ABST
    Figure CN119051718A_ABST
Patent Text Reader

Abstract

The invention provides a mobility-aware content caching method in a satellite internet and related equipment. The method comprises the following steps: pre-determining handover information of a satellite in a predictable duration at a current moment according to an operation orbit and an operation speed of the satellite and a position of a user in a service area; dividing a plurality of user clusters according to the content popularity of the plurality of user request contents in the unit time period in the service area; establishing an optimization target with the maximum cache content hit number of the satellite in a unit time period; determining the current popularity of each content in the unit time period and the future popularity of each content in the unit time period according to the handover information; determining the predicted popularity of each content according to the future popularity of each content and the current popularity of each content; and determining the predicted cache content of the satellite according to the predicted popularity of each content and the cache capacity of the satellite. A dynamic cache decision can be made according to a satellite user handover rule by using the deterministic movement characteristic of a satellite.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a mobility-aware content caching method and related equipment in satellite Internet. Background Art

[0002] With the breakthrough of satellite Internet technology and the continuous innovation of satellite communication equipment, global network coverage has become a revolutionary development direction in the field of communication technology. With the global coverage of satellite Internet, users can access the Internet through satellite communication equipment no matter where they are, enjoy seamless network services, and carry out diversified digital life activities such as remote office, online education and real-time data transmission.

[0003] Satellite Internet technology is no longer limited to urban areas, but has also expanded to remote areas, oceans and the air, providing global users with a wider range of network access possibilities. Although satellite Internet technology has great potential, its high cost and complex technical characteristics also bring challenges. Compared with traditional ground networks, the construction and maintenance costs of satellite Internet are relatively high, requiring a large amount of capital investment and technical support, which further leads to a limited number of satellites that users can connect to, and there is a problem of insufficient resources when large-scale access is required. In addition, satellite Internet services are unstable. Factors such as physical obstruction, weather, and solar storms will cause fluctuations in the quality of the channel connecting the satellite and the user, and these problems directly lead to fluctuations in the network latency on the user side. Summary of the invention

[0004] In view of this, the purpose of this application is to propose a mobility-aware content caching method and related equipment in satellite Internet.

[0005] Based on the above purpose, the present application provides a mobility-aware content caching method in satellite Internet, including:

[0006] According to the orbit and speed of the satellite and the location of the user in the service area, the handover information of the satellite within the predictable duration is pre-determined at the current moment; the handover information includes the user to be switched out and the user to be switched in; the value of the predictable duration is less than the value of the unit time period;

[0007] According to the content popularity of the content requested by multiple users in a unit time period in the service area, the multiple users are divided into multiple user clusters, so that the user to be cut out and the user to be cut in respectively have corresponding typical content popularity; wherein the users in each cluster respectively have similar content popularity; each user cluster respectively has corresponding typical content popularity; the unit time period includes multiple time points;

[0008] Establishing the optimization goal of maximizing the number of cache content hits in a unit time period for the satellite;

[0009] Determining the current popularity of each content in a unit time period and the future popularity of each content in a unit time period according to the handover information;

[0010] Determining predicted popularity of each content according to the future popularity of each content and the current popularity of each content;

[0011] The predicted cache content of the satellite is determined according to the predicted popularity of each content and the cache capacity of the satellite.

[0012] In some embodiments, the optimization goal is: Where t is the time; T is the unit time period; k is the index of the content; is the indicator function, which is the continuous request content of the user in the time period [0, T]; is the service user group of satellite j at time t; u i For users, is the indicator function, and is the satellite s j With user u i Connection status at time t; is the set of users connected to the satellite Internet in the satellite Internet service area; B is the cache capacity of the satellite.

[0013] In some embodiments, the handover information includes Among them, S t is the state space; C t Request information for the current moment's content and the history window t h Content request information; For content c k The number of requests from time t-1 to the current time; For satellites j The handover information collected at time t; in, For the user to be cut out, For the user to be cut in; t w For a predictable duration;

[0014] Determining the current popularity of the content in a unit time period according to the handover information includes:

[0015] Obtaining the historical request count of each content in the state space;

[0016] Processing the historical request times of each content through a one-dimensional convolutional neural network to obtain a K-dimensional vector of the current popularity of each content in a unit time period;

[0017] The future popularity of the content is a K-dimensional vector, expressed as Determine; among them, is the K-dimensional vector of future popularity; f pop A function for calculating the popularity of switching user content; Among them, P i For user u i The content popularity vector of For user u i Content c k The popularity of

[0018] In some embodiments, determining the predicted popularity of each content according to the future popularity of each content and the current popularity of each content comprises:

[0019] Popularity adds the K-dimensional vector corresponding to the current popularity and the K-dimensional vector corresponding to the future content popularity, inputs them into the neural network model, outputs the K-dimensional vector for estimating the Q value, and obtains multiple estimated Q values; wherein the reward function of the neural network model is The action space of the neural network model is: in, is the action space, is the specific cache behavior; K represents the total number of all content; action space Constraints: Indicates the decision whether to cache the i-th content item; is the subset of the content cached by the satellite at time t; C stands for all contents; represents the subset of content requested by the user at time t;

[0020] In some embodiments, determining the predicted cache content of the satellite according to the predicted popularity of each content and the cache capacity of the satellite includes:

[0021] Sort the obtained multiple estimated Q values ​​in descending order, and select the top N estimated Q values;

[0022] Determine that the N contents corresponding to the first N Q values ​​in the sorting are the predicted cache contents of the satellite; wherein the value of N is smaller than the contents cached in the satellite.

[0023] In some embodiments, the method further comprises determining that users have similar content popularity by:

[0024] pass Determine the similarity of content popularity between two users; Pa is the user u a Pb is the content popularity of user u b Popularity of content;

[0025] In response to determining that the similarity is greater than a similarity threshold, it is determined that the corresponding two users have similar content popularity.

[0026] In some embodiments, the method further comprises:

[0027] The cache content of the satellite is updated according to the cache content of the satellite at the current moment and the predicted cache content of the satellite, so that the satellite caches the predicted cache content.

[0028] The embodiment of the present application also provides a mobility-aware content caching device in satellite Internet, including:

[0029] The handover information determination module is configured to predetermine the handover information of the satellite within a predictable duration at the current moment according to the orbit and speed of the satellite and the location of the user in the service area; the handover information includes the user to be switched out and the user to be switched in; the value of the predictable duration is less than the value of the unit time period;

[0030] The user cluster division module is configured to divide the multiple users into multiple user clusters according to the content popularity of the content requested by the multiple users in the unit time period in the service area, so that the user to be cut out and the user to be cut in respectively have corresponding typical content popularity; wherein the users in each cluster respectively have similar content popularity; each user cluster respectively has corresponding typical content popularity; the unit time period includes multiple time points;

[0031] An optimization target establishment module is configured to establish an optimization target of maximizing the number of cache content hits of the satellite within a unit time period;

[0032] A first popularity determination module is configured to determine the current popularity of each content in a unit time period and the future popularity of each content in a unit time period according to the handover information;

[0033] a second popularity determination module configured to determine the predicted popularity of each content according to the future popularity of each content and the current popularity of each content;

[0034] The predicted cache content determination module is configured to determine the predicted cache content of the satellite according to the predicted popularity of each content and the cache capacity of the satellite.

[0035] An embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the methods described above when executing the program.

[0036] An embodiment of the present application also provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable a computer to execute any of the methods described above.

[0037] An embodiment of the present application further provides a computer program product, comprising computer program instructions, which, when executed on a computer, enable the computer to execute any of the methods described above.

[0038] From the above, it can be seen that the mobility-aware content caching method in satellite Internet provided by the present application can further deduce certain user group changes by pre-arranging the satellite handover behavior and pre-arranging a certain satellite to perform handover at the next moment according to the channel quality, and make caching decisions based on the popularity preferences of these users. It can utilize the deterministic mobility characteristics of the satellite, make dynamic caching decisions based on the user handover rules of the satellite, and use the information of the deterministic changes in the user group to decide the optimal cache content, so as to optimize the transmission delay performance and improve the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the present application or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 A schematic diagram of a flow chart of a method for caching content based on mobility awareness in satellite Internet according to an embodiment of the present application;

[0041] Figure 2 A schematic diagram of satellite deterministic movement and handover behavior in an embodiment of the present application;

[0042] Figure 3 A schematic diagram of a mobility-aware cache decision algorithm according to an embodiment of the present application;

[0043] Figure 4 A schematic diagram of a mobility-aware content caching device in satellite Internet according to an embodiment of the present application;

[0044] Figure 5 A schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0046] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should be the usual meanings understood by people with ordinary skills in the field to which the present application belongs. The "first", "second" and similar words used in the embodiments of the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0047] With the global coverage of satellite Internet, users can access the Internet through satellite communication equipment no matter where they are. Companies such as StarLink and OneWeb have deployed their Low Orbit Satellite (LEO) constellations, providing ubiquitous Internet services to the world. However, satellite Internet technology is characterized by high cost and technical complexity, and requires a lot of capital investment and technical support. As a result, the number of satellites that users can connect to is limited, and there is a problem of insufficient resources during large-scale access. In addition, factors such as physical obstruction, weather, and solar storms cause fluctuations in the quality of the channel connecting satellites to users. These problems directly lead to fluctuations in user-side network latency.

[0048] Edge caching is considered one of the main methods to optimize transmission delay. In ground networks, edge caching has been widely used in network transmission, especially in business transmission with low latency requirements. By caching content that users are more likely to request in network nodes closer to users, such as edge servers, the reduction in physical distance can significantly reduce the latency of content transmission, thereby further improving user experience. The most important issue in edge caching is caching decision, that is, which content should be cached within the limited cache space to achieve the highest cache hit rate, that is, the frequency of the user's requested content being cached at the edge node.

[0049] In satellite Internet, the satellite is the last hop in the path from the source server to the user, so it can be regarded as the edge node closest to the user. Using the idea of ​​edge caching, caching highly popular content at the satellite can significantly reduce transmission delay, thereby reducing the degradation of user experience caused by channel quality fluctuations. However, caching decisions in satellite Internet are more difficult than those in ground networks. First, the cache space of the satellite is more limited, and after being deployed in space, the cache space cannot be expanded by adding hardware. Therefore, it requires more precise cache decisions to achieve a good cache hit rate level. In addition, the satellite moves very fast in a fixed orbit. When the satellite is far away from the user, the channel quality decreases with the decrease of elevation angle and the increase of distance. Therefore, when the channel quality is insufficient, it is necessary to hand over the user far away from the satellite to another satellite for service through "handover". From the perspective of the satellite, the composition of the user group served by a single satellite is highly dynamic. This dynamicity further poses a challenge to the analysis of content popularity because different users have different content preferences.

[0050] Therefore, there is a problem in the related technology that the satellite cache hit rate for the request content of the highly dynamic user group is not high enough, which leads to poor transmission delay and poor user experience. It is urgent to optimize the satellite Internet edge cache decision.

[0051] Based on this, the embodiment of the present application provides a content caching decision method based on satellite deterministic mobility, the method includes: proposing a deterministic satellite user handover and content popularity mapping scheme, formulating future handover actions according to the deterministic mobility characteristics of the satellite, and further deriving content popularity; implementing a mobility-aware cache decision algorithm, adjusting cache content in real time according to dynamic user group changes and content popularity changes, to achieve low-latency content delivery effects. The content popularity mapping scheme includes: satellite handover behavior preset based on deterministic mobility; user group change and content popularity mapping method. The satellite mobility-aware cache decision algorithm includes: satellite edge cache decision problem modeling; content popularity knowledge transfer method between satellites; satellite edge cache decision algorithm based on reinforcement learning. The content cache decision method based on satellite deterministic mobility utilizes the deterministic mobility characteristics of the satellite, formulates dynamic cache decisions according to the user handover rules of the satellite, and uses the information of deterministic changes in the user group to decide the optimal cache content, so as to optimize the transmission delay performance and improve the user experience. It can solve the main problem of "satellite Internet edge cache decision optimization" to a certain extent, and maximize the cache hit rate and minimize the transmission delay based on the accurate analysis of content popularity within the limited cache space of the satellite.

[0052] Figure 1This is a flow chart of a method for mobility-aware content caching in satellite Internet according to an embodiment of the present application.

[0053] like Figure 1 As shown, the mobility-aware content caching method in the satellite Internet of the embodiment of the present application may include:

[0054] S100, predetermining, at the current moment, handover information of the satellite within a predictable duration according to the orbit and speed of the satellite and the location of the user in the service area; the handover information includes the user to be switched out and the user to be switched in; the value of the predictable duration is less than the value of the unit time period;

[0055] S200, dividing the multiple users into multiple user clusters according to the content popularity of the content requested by the multiple users in the unit time period in the service area, so that the users to be cut out and the users to be cut in respectively have corresponding typical content popularity; wherein the users in each cluster respectively have similar content popularity; each user cluster respectively has corresponding typical content popularity; the unit time period includes multiple time points;

[0056] S300, establishing an optimization goal of maximizing the number of cache content hits of the satellite within a unit time period;

[0057] S400, determining the future popularity of each content in a unit time period and the current popularity of each content in a unit time period according to the handover information;

[0058] S500, determining the predicted popularity of each content according to the future popularity of each content and the current popularity of each content;

[0059] S600: Determine the predicted cache content of the satellite according to the predicted popularity of each content and the cache capacity of the satellite.

[0060] The mobility-aware content caching method in the satellite Internet of the embodiment of the present application predetermines the handover information of the satellite within a predictable duration at the current moment according to the orbit and speed of the satellite and the position of the user in the service area; the handover information includes the user to be cut out and the user to be cut in; the value of the predictable duration is less than the value of the unit time period; according to the content popularity of the content requested by multiple users in the unit time period in the service area, the multiple users are divided into multiple user clusters, so that the user to be cut out and the user to be cut in respectively have corresponding typical content popularity; wherein the users in each cluster respectively have similar content popularity; each user cluster respectively has corresponding typical content popularity degree; the unit time period includes multiple moments; establishing an optimization goal of maximizing the number of cache content hits of the satellite within the unit time period; determining the current popularity of each content within the unit time period and the future popularity of each content within the unit time period according to the handover information; determining the predicted popularity of each content according to the future popularity of each content and the current popularity of each content; determining the predicted cache content of the satellite according to the predicted popularity of each content and the cache capacity of the satellite; being able to utilize the deterministic mobility characteristics of the satellite, formulate dynamic cache decisions according to the user handover rules of the satellite, and utilize the information of deterministic changes in the user group to decide the optimal cache content, so as to optimize the transmission delay performance and improve the user experience.

[0061] In step S100, the main purpose of the satellite handover action is to ensure that the user always has the best quality of experience (QoE). Satellite handover can be understood as that before the satellite leaves the user's coverage area, the user needs to be transferred to another satellite that can cover the user. Usually, the elevation angle between the satellite and the user can be used to determine whether the satellite covers the user, that is, whether the user is a user served by the satellite. Since the moving speed of the satellite is much greater than the moving speed of the ground user, usually only the movement of the satellite is considered, and the user's position can be considered to be fixed. According to the orbit and speed of the satellite, as well as the position of the user in the service area, the connection angle between the satellite and the user and the ground at each moment can be obtained. The connection angle obtained at each moment is compared with the minimum elevation angle θ between the user and the satellite that usually guarantees the channel quality and avoids obstacles, so as to determine whether the user is a service user of the satellite. Specifically, when the comparison result is less than, it can be determined that the user is not a service user of the satellite; and when the comparison result is greater than, it can be determined that the user is a service user of the satellite. In this way, the satellite can clearly perceive the set of satellites that the user can access in the future, and pre-arrange which satellite to hand over at the next moment according to the channel quality. When the handover situation in the future is predictable, the satellite can further deduce the definite changes in the user group, that is, which users will leave their service user group and which new users will join. It can further make caching decisions based on the popularity preferences of these users.

[0062] In some of these embodiments, at time t, satellite s j A user group can be defined as The user who is switched in the user group Indicates that the user who is switched from the user group Therefore, the new user group in the next moment can be expressed as

[0063] In this way, the deterministic movement of satellites, that is, the characteristic that satellites move at a fixed speed on a fixed orbit, can be used to determine the position of satellites at every moment. Combined with the fixed position of users, the elevation angle between users and satellites can be determined. After comparing with the minimum elevation angle, the satellites covering specific users at each moment can be predicted in advance, and which users will join or leave their user groups at some time in the future can be determined. Conditional handover instructions (CHO) can also be specified based on channel conditions and other relevant information. At the same time, it is convenient for the satellite ground station as the control center to send handover instructions to the satellite.

[0064] In some of the embodiments, the handover information of the satellite within a predictable duration can be predetermined at the current moment. The predictable duration can be understood as multiple moments, whose duration is greater than a single moment and less than a unit time period. The predictable moment can be understood as the moment when the satellite can predict the channel quality. Since the satellite's wireless signal transmission is significantly affected by the environment, including factors such as obstruction, weather, and other signal interference that will affect the channel quality, the change in channel quality is often difficult to predict over a long period of time and can only be predicted within a certain period of time (that is, the predictable duration). By identifying the user group at time t in a predictable duration (for example, a future window t w ) can avoid the poor user QoE that may result from premature determination of the satellite's manual switching plan.

[0065] In step S200, in the service area M, users continuously request content through the satellite network. Generally, the service area can be an area covered by satellite Internet, and can have a corresponding latitude range and a corresponding longitude range. Generally, all users in the service area request content according to a certain content popularity. The popularity of content may be different for different users. In some embodiments, users with similar content popularity are grouped into the same cluster, and then all users can be divided into different clusters, and different clusters represent different typical content popularity. In this way, the typical content popularity corresponding to the user can be quickly analyzed according to the cluster to which the user belongs. It can improve the efficiency of analyzing the changes in content popularity corresponding to the changes in users in the subsequent handover process.

[0066] In some embodiments, the configuration may request that all content contained in the collection User u i Content c k The popularity of And need to order User i The popularity of the content can be For user u a Content c k The popularity of For user u b Content c k The popularity of User a The popularity of the content can be User b The popularity of the content can be User a and user u b The content similarity between them can be calculated as follows: That is, you can use Determine the similarity of content popularity between two users; Pa is the user u a Pb is the content popularity of user u b content popularity; in response to determining that the similarity is greater than a similarity threshold (e.g., 0.9), determining that the corresponding two users have similar content popularity; and dividing the users with similar content popularity into the same cluster. In this way, all users can be divided into different clusters, and different clusters represent different typical content popularity. Therefore, the user group of the satellite service may include users from different clusters.

[0067] In some embodiments, the content of the user's continuous request in the time period [0, T] can be set using the indicator function To describe this process. i Request content c at time t k hour, is equal to 1, otherwise the result is 0. At any given moment, a single user can only make one request for content, which can be expressed as

[0068] In this way, during the satellite service process, the user group may contain users from different clusters, that is, users with different content popularity, but their typical content popularity can be quickly analyzed according to the cluster they are in. Then during the handover process, the changes in the user group can be analyzed in detail as each user's joining and leaving behavior, and further infer the overall content popularity changes through their typical content popularity.

[0069] In step S300, in some embodiments, as Figure 2 As shown, within the satellite Internet service area, terminal users connected to the satellite Internet can A constellation can usually consist of many low-Earth orbit satellites to provide network services to ground users. Each of these s jEach represents a satellite. Satellites move along fixed orbits at a predetermined speed, so they appear regularly in a specific area (that is, a small area obtained by further dividing the service area) and provide connections for users. Multiple satellites are connected to each other through inter-satellite links (ISL) to achieve data transmission between satellites. Satellites are also connected to ground stations in the coverage area through ground station links (GSL). The ground station is the gateway between the terrestrial Internet and the satellite network. On the one hand, the ground station is connected to the terrestrial Internet; on the other hand, the ground station transmits data and control signals to the satellite. When a user initiates a request, the data will first be transmitted from the Internet to the ground station, and then sent to the designated service satellite in the final jump of the satellite, and finally the service satellite will transmit the data to the user.

[0070] Typically, in a specific service time period (e.g., unit time period [0, T]), although multiple satellites cover the area, a user can only establish a connection with one satellite at each time t. i At time t and satellite s j The connection status can be indicated by the indicator function The value of this function is 1, which means there is a connection between the satellite and the user; while the value is 0, which means there is no connection between the satellite and the user. Correspondingly, at time t, each satellite has its service user group, which can be expressed as:

[0071]

[0072] In some of the embodiments, the optimization goal of maximizing the number of cache content hits of the satellite within a unit time period (i.e., the cache decision algorithm of satellite mobility awareness) may be: Where t is the time; T is the unit time period; k is the index of the content; is the indicator function, which is the continuous request content of the user in the time period [0, T]; is the service user group of satellite j at time t; u i For users, is the indicator function, and is the satellite s j With user u i Connection status at time t; is the set of users connected to the satellite Internet in the satellite Internet service area; B is the cache capacity of the satellite.

[0073] Thus, the formula can be defined on the satellite s j In the user group, at time t, the content requested by the user is available on satellite s jThe number of cached situations in can also represent the cache hit rate of the satellite. At the same time, the given constraints can ensure that the number of cached contents of the satellite does not exceed its cache space at any given time. Since the caches between satellites are decoupled, the cache decisions of other satellites will not affect the cache decisions of the current satellite. In this way, solving the optimization problem for each satellite separately is equivalent to maximizing the cache hit rate of the current service area, which can improve the user experience by reducing transmission delays.

[0074] In step S400, in some embodiments, the handover information may include Among them, S t is the state space; C t Request information for the current moment's content and the history window t h The content request information of the historical window th can be understood as the user's request content that has not changed within a predictable time period. For content c k The number of requests from time t-1 to the current time; For satellites j The handover information collected at time t; in, For the user to be cut out, For the user to be cut in; t w The handover information can indicate the upcoming changes in the user group. This facilitates the subsequent analysis of satellite s by combining the popularity of each user's content after the user group changes. j The development of the popularity of the content. By analyzing the content requests in a short time window in the past and the content requests in the future, and analyzing the content popularity (such as frequency) information therein, the current popularity and future popularity of the content can be obtained.

[0075] Generally, the popularity of content is a key factor in the satellite's caching decision process. In this way, in the satellite caching scenario, both past information (such as content) and future switching information (such as content) can be considered. And when the satellite performs the switching action, the typical content popularity of the user to be switched can be informed to the next service satellite. Compared with only considering past information, the changes in content popularity (that is, popularity) can be better analyzed.

[0076] In some of these embodiments, the predicted popularity of content can be determined by a mobility-aware edge cache decision algorithm based on deep reinforcement learning, which can solve the dynamic problem of satellite edge cache. Figure 3As shown in the figure, the satellite's deterministic movement and switching actions can be informed to the agent as part of the environmental state, so that the agent can better adapt to changes in the state when making decisions. The use of deep reinforcement learning enables the satellite to quickly perceive environmental changes and enhance its adaptability to different environments. Compared with strategic algorithms, its dynamic adjustment ability is stronger. The satellite's caching decision can be further modeled as a Markov decision process (MDP).

[0077] In some embodiments, determining the current popularity of the content in the unit time period according to the handover information may include:

[0078] Obtaining the historical request count of each content in the state space;

[0079] The historical request times of each content are processed by a one-dimensional convolutional neural network to obtain a K-dimensional vector of the current popularity of each content in a unit time period. That is, in, is a K-dimensional vector of the current popularity of the content. The one-dimensional convolutional neural network may be a 1D-CNN convolutional neural network, which can analyze the current popularity of the content, that is, can analyze the content popularity corresponding to the users who still remain in the user group of the satellite in the future handover action.

[0080] In some embodiments, the future popularity of the content is a K-dimensional vector, which can be expressed as Determine; among them, is the K-dimensional vector of future popularity; f pop A function for calculating the popularity of switching user content; Among them, P i For user u i The content popularity vector of For user u i Content c k The popularity of

[0081] In this way, a K-dimensional vector for each content is calculated through a moving average window, which can combine users (including users who are switching and users representing their typical popularity information) and the typical popularity information they represent to calculate the changes in the popularity of each content that may occur in the short term (such as within a predictable duration).

[0082] In some embodiments, the deep learning network of the agent can be a DQN algorithm based on a natural reinforcement learning architecture, using a target network and experience replay to optimize the training process. The neural network can be composed of a three-layer fully connected (FC) network. In the context of satellite caching, the action space can be a specific cache behavior. K can be the total number of all content items, and B can be the cache capacity of the satellite. It is represented as a binary vector of length K, where at most B elements are 1 and the rest are 0. Formally, the action space can be defined as: Here, a i ∈{0,1} indicates whether the tth content item is cached (1) or not cached (0), provided that the total number of cached items does not exceed the cache capacity.

[0083] In some embodiments, the behavior space at time i Constraints can include: Here a i ∈{0, 1} indicates whether to cache the content item, and is allowed to be 1 only if the item has been cached or requested and does not violate the cache capacity limit. Usually, 1 means caching is allowed and 0 means caching is not allowed. represents the subset of the contents cached by the satellite at time t, Represents the subset of content requested by the user at time t. Action Must satisfy only from Only the items in the collection can be cached. That is, the action space of the network model is: in, is the action space, is the specific cache behavior; K represents the total number of all content; action space Constraints: Indicates the decision whether to cache the i-th content item; is the subset of the content cached by the satellite at time t; C stands for all contents; represents the subset of content requested by the user at time t;

[0084] In this way, by normalizing the behavior space and making it consistent with the actual cache behavior, the action can be made to conform to the constraints. Next, if the content requested by the user is already in the satellite cache at that time, it can be directly transmitted to the user. Otherwise, the satellite must request the content from the source server and transmit it to the user. At time t, the content that the satellite can choose to cache or not cache is the union of the downloaded content and the current cached content.

[0085] In some embodiments, for the reward function, the optimization goal is to maximize the number of cache hits of a specific satellite over time. A cache hit means that the content requested by the user is already stored in the cache of the satellite. j , the reward for a given action a at time t can be defined as The reward function is the sum of the cache hit counts for all users and content items in , divided by the total number of requests at time t. Where t is the time; T is the unit time period; k is the index of the content; is the indicator function, which is the continuous request content of the user in the time period [0, T]; is the service user group of satellite j at time t; u i For users, is the indicator function, and is the satellite s j With user u i Connection status at time t; is the set of users connected to the satellite Internet in the satellite Internet service area; B is the cache capacity of the satellite. This reward function can encourage the agent to learn a strategy to maximize the probability of storing the most frequently requested content, thereby improving the cache hit rate and improving the overall performance of the satellite communication system. By adding the K-dimensional vector of the current popularity of each content in the unit time period to the corresponding K-dimensional vector of future popularity, the final K-dimensional input vector of the deep learning network for each content in the unit time period can be obtained, and another K-dimensional vector is output as the Q value estimate.

[0086] In this way, the agent is able to analyze short-term historical request data to estimate the popularity of content and predict future changes by integrating handoff information.

[0087] In some embodiments, determining the predicted popularity of each content according to the future popularity of each content and the current popularity of each content comprises:

[0088] Popularity adds the K-dimensional vector corresponding to the current popularity and the K micro-vector corresponding to the future content popularity, inputs the result into a neural network model, outputs a K-dimensional vector for estimating the Q value, and obtains multiple estimated Q values.

[0089] In some embodiments, determining the predicted cache content of the satellite according to the predicted popularity of each content and the cache capacity of the satellite includes:

[0090] The obtained multiple estimated Q values ​​are sorted in descending order, and the top N estimated Q values ​​are selected; the N contents corresponding to the top N Q values ​​are determined as the predicted cache contents of the satellite; wherein the value of N is less than the contents cached in the satellite. Usually N can be a positive integer.

[0091] In this way, the first N Q values ​​can be used to determine the action space. The first N strategy can enhance the scalability of the network structure while reducing the adverse training effects of too large action space dimensions. HACR can still operate normally despite the increase in the satellite's cache capacity and the amount of content in the scene. Applying a mask layer can make the generated actions legal, forcing the action to be a binary vector with at most B elements set to 1 under the cache capacity constraint.

[0092] In some embodiments, the method may further include: updating the cache content of the satellite according to the cache content of the satellite at the current moment and the predicted cache content of the satellite, so that the satellite caches the predicted cache content. Specifically, in response to determining that there is uncached content in the predicted cache content, the satellite may request the content from the source server and cache it. In response to a certain content in the cache content at the current moment not being present in the predicted cache content, the cache content is deleted.

[0093] Experimental example

[0094] Experimental purpose: Conduct experimental simulation and actual system testing to verify the superior optimization performance of the solution proposed in this application in terms of improving cache hit rate, reducing overall latency, and improving user experience.

[0095] Experimental Method: The simulation platform uses the configuration of the Starlink constellation to build a service constellation. Starlink is one of the largest non-terrestrial network (NTN) service providers in the world today. In order to optimize the simulation running time, a simpler scenario is considered, that is, using the first layer in the Starlink constellation to build a service constellation. This layer consists of 22 orbits, the altitude of these orbits is 550 kilometers, and the orbital period is 5730 seconds. Each orbit contains 72 satellites, and there are 1584 satellites in total, which is sufficient for the simulation scenario. The experiment selects a region in northern China as the simulated service area. The region is a suburban area with a low population density and is very suitable for satellite Internet services. The latitude range of the service area is 41 degrees to 45 degrees, and the longitude range is 116 degrees to 120 degrees. The service area is divided into nine small areas, each with 100 users. The geographical locations of the users are randomly generated and dispersed throughout the service area. The minimum elevation angle θ is set to 25 degrees. When the elevation angle of the user pointing to the satellite is less than θ, it can be considered that the satellite is no longer visible to the user.

[0096] The satellite mobility-aware caching decision algorithm of the embodiment of the present application (that is, the mobility-aware content caching method in the satellite Internet of the embodiment of the present application) is compared with the traditional LFU (most frequently requested content priority caching), LRU (most recently requested content priority caching) and another reinforcement learning-based caching decision (ACORN) suitable for ground networks. The results show that the proposed caching strategy algorithm improves the cache hit rate by 28.69% compared with the traditional method and by 16.74% compared with the existing reinforcement learning method. In terms of algorithm stability, the proposed caching strategy algorithm improves by 48.9% compared with the existing reinforcement learning method.

[0097] It is understandable that before using the technical solutions of each embodiment of the present disclosure, the type, scope of use, usage scenarios, etc. of the personal information involved will be informed to the user in an appropriate manner, and the user's authorization will be obtained.

[0098] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly remind the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can independently choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0099] As an optional but non-limiting implementation, in response to receiving the user's active request, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0100] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0101] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present application, and the multiple devices will interact with each other to complete the described method.

[0102] It should be noted that the above describes some embodiments of the present application. 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 above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying 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.

[0103] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a mobility-aware content caching device in satellite Internet.

[0104] refer to Figure 4 , the mobility-aware content caching device 800 in the satellite Internet comprises:

[0105] The handover information determination module 810 is configured to predetermine the handover information of the satellite within a predictable duration at the current moment according to the orbit and speed of the satellite and the location of the user in the service area; the handover information includes the user to be switched out and the user to be switched in; the value of the predictable duration is less than the value of the unit time period;

[0106] The user cluster division module 820 is configured to divide the multiple users into multiple user clusters according to the content popularity of the content requested by the multiple users in the unit time period in the service area, so that the user to be cut out and the user to be cut in respectively have corresponding typical content popularity; wherein the users in each cluster respectively have similar content popularity; each user cluster respectively has corresponding typical content popularity; the unit time period includes multiple time points;

[0107] The optimization target establishment module 830 is configured to establish an optimization target of maximizing the number of cache content hits of the satellite within a unit time period;

[0108] A first popularity determination module 840 is configured to determine the current popularity of each content in a unit time period and the future popularity of each content in a unit time period according to the handover information;

[0109] A second popularity determination module 850 is configured to determine the predicted popularity of each content according to the future popularity of each content and the current popularity of each content;

[0110] The predicted cache content determination module 860 is configured to determine the predicted cache content of the satellite according to the predicted popularity of each content and the cache capacity of the satellite.

[0111] In some embodiments, the optimization goal is: Where t is the time; T is the unit time period; k is the index of the content; is the indicator function, which is the continuous request content of the user in the time period [0, T]; is the service user group of satellite j at time t; u i For users, is the indicator function, and is the satellite s j With user u i Connection status at time t; is the set of users connected to the satellite Internet in the satellite Internet service area; B is the cache capacity of the satellite.

[0112] In some embodiments, the handover information includes Among them, S t is the state space; C t Request information for the current moment's content and the history window t h Content request information; For content c k The number of requests from time t-1 to the current time; For satellites j The handover information collected at time t; in, For the user to be cut out, For the user to be cut in; t w For a predictable duration.

[0113] In some embodiments, the first popularity determination module 840 is configured to:

[0114] Obtaining the historical request count of each content in the state space;

[0115] Processing the historical request times of each content through a one-dimensional convolutional neural network to obtain a K-dimensional vector of the current popularity of each content in a unit time period;

[0116] The future popularity of the content is a K-dimensional vector, expressed as Determine; among them, is the K-dimensional vector of future popularity; f pop A function for calculating the popularity of switching user content; Among them, P i For user u i The content popularity vector of For user u i Content c k The popularity of

[0117] In some embodiments, the second popularity determination module 850 is configured to:

[0118] Popularity adds the K-dimensional vector corresponding to the current popularity and the K-dimensional vector corresponding to the future content popularity, inputs them into the neural network model, outputs the K-dimensional vector for estimating the Q value, and obtains multiple estimated Q values; wherein the reward function of the neural network model is The action space of the neural network model is: in, is the action space, is the specific cache behavior; K represents the total number of all content; action space Constraints: Indicates the decision whether to cache the i-th content item; is the subset of the content cached by the satellite at time t; C stands for all contents; represents the subset of content requested by the user at time t;

[0119] In some embodiments, the predicted cache content determination module 860 is configured to:

[0120] Sort the obtained multiple estimated Q values ​​in descending order, and select the top N estimated Q values;

[0121] Determine that the N contents corresponding to the first N Q values ​​in the sorting are the predicted cache contents of the satellite; wherein the value of N is smaller than the contents cached in the satellite.

[0122] In some embodiments, the user clustering module 820 is configured to:

[0123] pass Determine the similarity of content popularity between two users; Pa is the user u a Pb is the content popularity of user u b Popularity of content;

[0124] In response to determining that the similarity is greater than a similarity threshold, it is determined that the corresponding two users have similar content popularity.

[0125] In some of the embodiments, the mobility-aware content caching device 800 in satellite Internet may further include a cache content updating module configured to:

[0126] The cache content of the satellite is updated according to the cache content of the satellite at the current moment and the predicted cache content of the satellite, so that the satellite caches the predicted cache content.

[0127] For the convenience of description, the above device is described in terms of functions divided into various modules. Of course, when implementing the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0128] The device of the above embodiment is used to implement the corresponding mobility-aware content caching method in satellite Internet in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0129] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the mobility-aware content caching method in satellite Internet described in any of the above embodiments is implemented.

[0130] Figure 5 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other through the bus 1050 in the device.

[0131] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0132] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.

[0133] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0134] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0135] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0136] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.

[0137] The electronic device of the above-mentioned embodiment is used to implement the corresponding mobility-aware content caching method in satellite Internet in any of the above-mentioned embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0138] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the mobility-aware content caching method in the satellite Internet as described in any of the above embodiments.

[0139] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0140] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the mobility-aware content caching method in satellite Internet as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0141] Based on the same inventive concept, corresponding to the method for caching content in satellite Internet with mobility awareness described in any of the above embodiments, the present disclosure also provides a computer program product, which includes computer program instructions. In some embodiments, the computer program instructions can be executed by one or more processors of a computer so that the computer and / or the processor execute the method for caching content in satellite Internet with mobility awareness. Corresponding to the execution subject corresponding to each step in each embodiment of the method for caching content in satellite Internet with mobility awareness, the processor that executes the corresponding step may belong to the corresponding execution subject.

[0142] The computer program product of the above embodiment is used to enable the computer and / or the processor to execute the mobility-aware content caching method in satellite Internet as described in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0143] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0144] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present application difficult to understand, the known power supply / ground connection with the integrated circuit (IC) chip and other components may or may not be shown in the provided drawings. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present application difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform to be implemented in the embodiments of the present application (that is, these details should be fully within the scope of understanding of those skilled in the art). In the case of elaborating specific details (e.g., circuits) to describe exemplary embodiments of the present application, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details or when these specific details are changed. Therefore, these descriptions should be considered to be illustrative rather than restrictive.

[0145] Although the present application has been described in conjunction with specific embodiments of the present application, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0146] The embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of the present application.

Claims

1. A method for caching content with mobility awareness in satellite Internet, characterized in that: include: Based on the orbit and speed of the satellite and the location of users in the service area, the handover information of the satellite within a predictable time is pre-determined at the current moment; The handover information includes users to be switched out and users to be switched in; the value of the predictable duration is less than the value of the unit time period; According to the content popularity of the content requested by multiple users in a unit time period in the service area, the multiple users are divided into multiple user clusters, so that the user to be cut out and the user to be cut in respectively have corresponding typical content popularity; wherein the users in each cluster respectively have similar content popularity; each user cluster respectively has corresponding typical content popularity; the unit time period includes multiple time points; Establishing the optimization goal of maximizing the number of cache content hits in a unit time period for the satellite; Determining the current popularity of each content in a unit time period and the future popularity of each content in a unit time period according to the handover information; Determining predicted popularity of each content according to the future popularity of each content and the current popularity of each content; The predicted cache content of the satellite is determined according to the predicted popularity of each content and the cache capacity of the satellite.

2. The content caching method according to claim 1, characterized in that: The optimization goal is: Among them, t is the time; T is the unit time period; k is Indexing of content; is the indicator function, which is the continuous request content of the user in the time period [0, T]; is the service user group of satellite j at time t; u i For users, is the indicator function, is the satellite s j With user u i Connection status at time t; is the set of users connected to the satellite Internet in the satellite Internet service area; B is the cache capacity of the satellite.

3. The content caching method according to claim 2, characterized in that: The handover information includes Among them, S t is the state space; C t Request information for the current moment's content and the history window t h Content request information; For content c k The number of requests from time t-1 to the current time For satellites j The handover information collected at time t; in, For the user to be cut out, For the user to be cut in; t w For a predictable duration; Determining the current popularity of the content in a unit time period according to the handover information includes: Obtaining the historical request count of each content in the state space; Processing the historical request times of each content through a one-dimensional convolutional neural network to obtain a K-dimensional vector of the current popularity of each content in a unit time period; The future popularity of the content is a K-dimensional vector, expressed as Determine; among them, is the K-dimensional vector of future popularity; f pop A function for calculating the popularity of switching user content; Among them, P i For user u i The content popularity vector of For user u i Content c k The popularity of 4. The content caching method according to claim 3, characterized in that: Determining the predicted popularity of each content according to the future popularity of each content and the current popularity of each content includes: Popularity adds the K-dimensional vector corresponding to the current popularity and the K-dimensional vector corresponding to the future content popularity, inputs them into the neural network model, outputs the K-dimensional vector for estimating the Q value, and obtains multiple estimated Q values; wherein the reward function of the neural network model is The action space of the neural network model is: in, is the action space, is the specific cache behavior; K represents the total number of all content; action space Constraints: a i ∈{0, 1} represents the decision of whether to cache the i-th content item; is the subset of the content cached by the satellite at time t; C stands for all contents; represents the subset of content requested by the user at time t; 5. The content caching method according to claim 4, characterized in that: Determining the predicted cache content of the satellite according to the predicted popularity of each content and the cache capacity of the satellite includes: Sort the obtained multiple estimated Q values ​​in descending order, and select the top N estimated Q values; Determine that the N contents corresponding to the first N Q values ​​in the sorting are the predicted cache contents of the satellite; wherein the value of N is smaller than the contents cached in the satellite.

6. The content caching method according to claim 1, characterized in that: The method also includes determining that users have similar content popularity by: pass Determine the similarity of content popularity between two users; Pa is the user u a Pb is the content popularity of user u b Popularity of content; In response to determining that the similarity is greater than a similarity threshold, it is determined that the corresponding two users have similar content popularity.

7. The content caching method according to claim 1, characterized in that: The method further comprises: The cache content of the satellite is updated according to the cache content of the satellite at the current moment and the predicted cache content of the satellite, so that the satellite caches the predicted cache content.

8. A mobility-aware content caching device in satellite Internet, characterized in that: include: The handover information determination module is configured to predetermine the handover information of the satellite within a predictable duration at the current moment according to the orbit and speed of the satellite and the location of the user in the service area; the handover information includes the user to be switched out and the user to be switched in; the value of the predictable duration is less than the value of the unit time period; The user cluster division module is configured to divide the multiple users into multiple user clusters according to the content popularity of the content requested by the multiple users in the unit time period in the service area, so that the user to be cut out and the user to be cut in respectively have corresponding typical content popularity; wherein the users in each cluster respectively have similar content popularity; each user cluster respectively has corresponding typical content popularity; the unit time period includes multiple time points; An optimization target establishment module is configured to establish an optimization target of maximizing the number of cache content hits of the satellite within a unit time period; A first popularity determination module, configured to determine the current popularity of each content in a unit time period and the future popularity of each content in the unit time period according to the handover information; a second popularity determination module configured to determine the predicted popularity of each content according to the future popularity of each content and the current popularity of each content; The predicted cache content determination module is configured to determine the predicted cache content of the satellite according to the predicted popularity of each content and the cache capacity of the satellite.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 7 when executing the program. 10 . A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method according to claim 1 .

Citation Information

Cited By

  • Active cache routing method and system for satellite content distribution network

    CN121309446A

  • Satellite network cache migration method based on regional feature prediction

    CN122496095A