Vehicle social network caching method based on digital twinning

Abstract

The invention relates to a vehicle social network caching method based on digital twinning, and belongs to the field of vehicle social networks. The method comprises the following steps: S1, extracting a connection link between a vehicle network and a social network; s2, mapping the vehicle network from a physical layer to a twin network layer based on a digital twin technology; s3, extracting relation characteristics of vehicles from the twin network layer and establishing a vehicle social network layer; s4, establishing a cache cloud in a vehicle social network layer; and S5, determining a vehicle cache cooperation strategy based on the vehicle social network layer and the cache cloud in the vehicle social network layer. According to the method, the content distribution effectiveness of the vehicle-mounted network is effectively improved, and the method has certain advantages in the aspects of effectiveness and unloading success rate indexes.

Description

Technical Field

[0001] The present invention belongs to the field of vehicle social networks and relates to a vehicle social network caching method based on digital twins. Background Art

[0002] The rapid development of multimedia application services in vehicles has led to an exponential growth in the volume of traffic in the Internet of Vehicles. In addition, the user demand for high-quality vehicle communications has greatly increased, making the network burden more severe. In order to alleviate the problem of insufficient computing resources of the vehicle itself, edge computing technology has been introduced in the Internet of Vehicles to offload the vehicle's tasks. In the process of edge computing offloading, edge caching technology will be used. Edge caching is to enable the vehicle itself to have the ability to cache and store, which can reduce communication energy consumption and other issues in the edge computing process. In the Internet of Vehicles system with edge caching, the hit rate is an important evaluation indicator of data caching technology, among which the transmission and placement of the cache affect the cache hit rate. Some studies have discussed the impact of content placement in the cache, so the placement of the cache is also worth considering. Edge caching technology stores content at each node with storage function in the network, so that the vehicle can obtain the required content from the surrounding cache nodes. Although it increases the redundancy of the content in the network, it improves the cache hit of the content during the service process, reduces the network load, and shortens the service delay.

[0003] When vehicles collaborate in the Internet of Vehicles, traditional methods generally directly consider nearby vehicles to collaborate, without considering using the social attributes of vehicles to form a social network. In the cache strategy part, it is possible to consider combining the cache content of the cache node with the user request match, and then predict the movement trajectory of the vehicle based on historical information. However, due to the high dynamics of vehicles, the intermittent connection between the content demanders and content providers of the vehicles can affect the dissemination efficiency of social perception content. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a vehicle social network caching method based on digital twins. Since digital twins can use their own technology to obtain social relationships between vehicles and current vehicle traffic distribution, they can allocate storage resources and content placement more specifically, combine users' social attributes with hot content, and thus form a vehicle social network. Digital twins can construct a multi-dimensional mapping from physical systems to virtual networks, and with the help of digital twins' data analysis and modeling capabilities, form a closed-loop system of data perception-efficient analysis-intelligent decision-making. In the Internet of Vehicles, digital twins can help realize adaptive network management from digital simulation to visual evaluation. This technology can be used to achieve full network planning for multiple regions. The introduction of digital twins into the Internet of Vehicles' computational offloading can bring three advantages: first, real-time monitoring of vehicle conditions, bringing comprehensive and accurate network analysis of the physical network; second, mapping the physical network layer and application layer to the twin network layer, realizing cross-layer interaction and feedback of the three-layer structure; third, being able to perceive the dynamic changes of the vehicle's network topology and vehicle location.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] A vehicle social network caching method based on digital twins includes the following steps:

[0007] S1, extract the connection between vehicle network and social network;

[0008] S2. Map the vehicle network from the physical layer to the twin network layer based on digital twin technology;

[0009] S3, extracting the relationship features of vehicles from the twin network layer and establishing a vehicle social network layer;

[0010] S4, establishing a cache cloud in the vehicle social network layer;

[0011] S5. Determine a vehicle cache cooperation strategy based on the vehicle social network layer and the cache cloud therein.

[0012] Furthermore, in step S1, the vehicle is influenced by the social activities of the driver and has social attributes. The social relations, user interests and social activities in the social network are the links between the vehicle network and the social network. Personalized recommendations are made to the drivers in the vehicle network based on the user interests in the social network. The social relations in the social network correspond to the network topology in the vehicle network. The network topology in the vehicle network is predicted and maintained based on the social relations. The social relations in the social network are extended based on the network topology. The mobility model in the vehicle network abstracts the social activities in the social network, and the mobility model is predicted or discovered based on the group nature of the social activities.

[0013] Further, in step S2, in the physical layer, it is assumed that there are N moving vehicles and M roadside units RSU on the road, the set of vehicles is defined as N = {1, 2, ..., N}, and the set of RSUs is defined as M = {1, 2, ..., M};

[0014] The vehicle demand content is divided into G categories, and each type of demand content is expressed as where f g Indicates the size of the content type, Indicates the maximum delay of its content, μ g Indicates the sensitivity coefficient of its delay;

[0015] Each RSU is equipped with an edge cache server, and each edge cache server has caching capabilities; the vehicle's required content is pre-stored in the vehicle or RSU's edge cache server.

[0016] Furthermore, in step S2, the twin network layer includes an information collection module, a control module, an RSU collaboration module, an instruction issuing module and a digital twin data storage module; the information collection module obtains the status of the vehicle through communication between the vehicle and the RSU; the control module is used to determine the update cycle and the interaction frequency and data type during the adjustment process, and the adjustment instruction is transmitted to the corresponding vehicle through the instruction issuing module; the RSU collaboration module is used to exchange data and update the status in a timely manner; the digital twin data storage module is used to store data.

[0017] Further, in step S3, the LSTM network is used to extract the social relationship of the vehicle, including at least two relationship features, which are the content matching of the two parties and the communication contact rate of the vehicle;

[0018] The probability β is used to describe the content matching element. Suppose the vehicle needs G content, and the probability of its demand is β = {β1,β2,β3,...,β G},in The probability that the vehicle's demand content matches the demand content in the communication area is β g ;

[0019] The communication contact rate depends on the distribution density and driving speed of vehicles in the area; assuming that the vehicle is traveling at a speed v, the prescribed speed in this section is v0, and the average distance between two consecutive vehicles is expressed as:

[0020]

[0021] Where d0 represents the safe distance between two vehicles, T0 represents the reaction time of the vehicle, and δ represents the speed change factor of the vehicle; according to the coverage range and average distance of the communication area, the vehicle distribution density λ of the communication area is obtained;

[0022] Then the two relational features of the social relationship of the vehicle are expressed as {β,λ}.

[0023] Further, in step S4, each cache cloud is composed of a number of vehicles in the area, each cache cloud includes a demand content; each vehicle includes a number of demand contents, each demand content corresponds to a cache cloud; thus, a multi-dimensional mapping is formed between the cache cloud and the vehicle;

[0024] When communicating, the number of vehicles that a content demander can obtain from the cache is constrained by two factors: one is the size of the cache content provided by the provider, and the other is the transmission rate between the two.

[0025] The signal-to-noise ratio formula for hybrid-degraded V2X communication is:

[0026]

[0027] Where SINR stands for signal-to-noise ratio, d stands for path length, α stands for path loss factor, and p w Represents the transmission power of the vehicle, P v represents the additive noise power of the vehicle;

[0028] Assumption γ min is the minimum signal-to-noise ratio that can be decoded by the receiver when receiving data. In an ideal environment without co-channel interference, the maximum distance of V2X communication is expressed as:

[0029]

[0030] The average rate of vehicle V2X communication is defined as:

[0031] R v =Blog2(1+SINR)

[0032] Wherein, B represents the channel bandwidth.

[0033] Further, in step S5, cache scheduling considers the balance between communication cost, model accuracy and system utility;

[0034] Let x g represents the probability of pre-storing content of category g on the cache cloud, y g represents the probability of pre-storing class g content on the RSU server; under the constraints of cache size and transmission delay, the optimal edge cache problem that maximizes the utility of the cache system is expressed as:

[0035]

[0036] Where N is the set of vehicles; M is the set of RSUs; ψ g (εg ) represents the accuracy of the vehicle social network layer; ε g The amount of system information collected in the digital twin for training the LSTM network and obtaining the vehicle social network layer; β g The probability of encountering a vehicle that requires Category G content; Indicates the time when the vehicle obtains the g-category content from the cache cloud; represents the time it takes for the vehicle to obtain content from RSUr; g,r represents the probability of pre-storing the g-type content on the RSUr server; v is the transmission cost of vehicle v; r is the transmission cost between vehicle and RSUr; c is the transmission cost requested by the vehicle to the data center; {Z} is the indicator function, which is 1 when {Z} is true and 0 when {Z} is false.

[0037] Further, in step S5, an optimal cache strategy is set in an iterative manner based on reinforcement learning to solve the optimal edge cache problem. In each iteration, a cache scheduling strategy is first obtained according to a given vehicle social network layer, and then the amount of information collected during the construction of the vehicle social network layer is modified according to the determined strategy, and the iteration is continued until the system converges;

[0038] According to the given vehicle social network layer and precision function ψ g (ε g ), the utility function for exploring the optimal decision is expressed as:

[0039]

[0040] Optimal caching strategy π * It is expressed as:

[0041]

[0042] Among them, η∈(0,1) represents the discount factor, which is used to weigh the utility function.

[0043] Further, in step S5, DDPG is applied to the edge cache collaborative management scenario. For a given cache strategy set to π, its state S i and Action A i The average system utility obtained under this condition is expressed as:

[0044]

[0045] The DDPG algorithm includes a main network and a target network; each network has an actor part and a critic part. The target network is regarded as an iteration of the main network, generating target values ​​for training the main network, and finally updating the strategy by calculating the loss function; the experience replay pool of the DDPG algorithm stores the learning experience of action rewards and state changes, which is used to train the actor and critic parameters.

[0046] The beneficial effects of the present invention are:

[0047] Compared with the existing cache placement in the Internet of Vehicles, the social cache mechanism based on digital twins proposed in this invention effectively improves the content distribution utility of the vehicle network, and has certain advantages in terms of utility and unloading success rate indicators. First, the digital twin method is used to map the physical edge network into the virtual space. Users will cache content to form a content cache cloud, making full use of the characteristics of digital twins for real-time updates; secondly, in order to build a social network, social attributes will be classified according to the characteristics of its cache and the behavioral characteristics of the vehicle itself; finally, edge cache placement is performed based on the social network. A social relationship model between vehicles is obtained. The model is further used to explore the cooperation between resource-constrained vehicles in large-scale content delivery, and a learning-based social relationship construction and cache resource scheduling scheme is proposed.

[0048] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:

[0050] Figure 1 A schematic diagram of the connection between the vehicle network and the social network of the present invention;

[0051] Figure 2 A schematic diagram of a digital twin-based Internet of Vehicles social network architecture of the present invention;

[0052] Figure 3 A schematic diagram of the construction process of the vehicle social network layer of the present invention;

[0053] Figure 4 A schematic diagram of a vehicle cache cloud of the present invention;

[0054] Figure 5 It is a schematic diagram of the edge cache strategy structure based on DDPG of the present invention. DETAILED DESCRIPTION

[0055] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0056] Among them, the drawings are only used for illustrative explanations, and they only represent schematic diagrams rather than actual pictures, and should not be understood as limitations on the present invention. In order to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0057] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "front", "rear", etc. indicate the orientation or position relationship, they are based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the terms describing the position relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0058] See also Figure 1 to Figure 5 , which is a vehicle social network caching method based on digital twins, and includes the following specific steps:

[0059] S1, extract the connection between vehicle network and social network;

[0060] S2. Map the vehicle network from the physical layer to the twin network layer based on digital twin technology;

[0061] S3. Extract vehicle relationship features from the twin network layer and establish a vehicle social network layer

[0062] S4, establishing a cache cloud in the vehicle social network layer;

[0063] S5. Determine a vehicle cache collaboration strategy based on the vehicle social network layer and the cache cloud therein.

[0064] Furthermore, in step S1, the driver of the vehicle is a person, and the original vehicle does not have social attributes, but because the person has social attributes and is affected by the driver's social activities, the vehicle also has social attributes. Such social attributes make it easier for nodes with similar interests or social connections to meet in a certain area. This network that considers the social attributes and social relationships between vehicles is defined as a vehicular social network (VSN) in the present invention.

[0065] Specifically, vehicle social networks can share information in real time between vehicles, between vehicles and roadside units, and between vehicles and storage devices. Since the high-speed mobility of vehicles can easily lead to interruptions in communication links, the social attributes between vehicles need to establish a stable connection. Figure 1 As shown in the figure, social relations, user interests and social activities are the links between the two architectures. User interests in social networks can be used to make personalized recommendations in vehicle networks. The interests and hobbies of different users create personalized recommendations centered on the user themselves in the application, which promotes the iteration and update of the application. The social relations in social networks correspond to the network topology in the vehicle network. Social relations can predict the network topology in the vehicle network, maintain the network topology in the social network, and the network topology can extend the social relations in the social network. Social activities can be abstracted from the mobility model of the vehicle network, and the mobility model can be predicted or discovered based on the group nature of social activities.

[0066] Further, in step S2, according to the corresponding relationship between the vehicle network and the social network in step S1, the present invention proposes the following Figure 2 The illustrated vehicle networking social network architecture based on digital twins includes at least a physical layer and a twin network layer. In the physical layer, it is assumed that there are N moving vehicles and M roadside units (RSUs) on the road. The set of vehicles is defined as N = {1, 2, ..., N}, and the set of RSUs is defined as M = {1, 2, ..., M}. The content required for the application implementation of smart vehicles is divided into G categories, and each type of content can be expressed as, where f g It refers to the size of the content type, the second parameter indicates the maximum delay of its content, and the third parameter indicates the sensitivity coefficient of its delay, that is, the utility per unit time obtained by reducing the maximum delay during the transmission process.

[0067] In the mentioned IoV architecture, each RSU is equipped with an edge cache server, and each server has cache capability, so in order to avoid long delays between the data center and the vehicle, the server can store part of the content in the cache. In addition, the content can also be pre-stored in the smart vehicle.

[0068] Further, in the IoV social network architecture proposed in step S2, the digital twin network is a digital mapping of the physical network layer and has control functions, such as Figure 3 As shown in the figure, generally speaking, digital twins mainly include three parts: data storage, twin model mapping and twin management. Data storage is responsible for collecting and storing various network data for virtual modeling, optimization and prediction. Twin management is responsible for managing and updating the twin network layer, and has functions such as model update and state synchronization. Figure 3 The construction method of vehicle social network based on digital twin is demonstrated. The main functions of digital twin network (twin network layer) are embodied into five modules: information collection module, control module, RSU collaboration module, command issuance and digital twin data storage. The information collection module mainly obtains the status of the vehicle through communication between the vehicle and RSU, which generally includes the vehicle's location information, service capabilities, driving status and data requirements. The digital twin network here adopts a distributed structure, which is generally built on RSUs in a region, so it is necessary to use the RSU collaboration module to exchange data update status in time to ensure the consistency of the mapped digital twin model. The control module mainly determines the update cycle and the interaction frequency and data type during the adjustment process. The adjustment command will be transmitted to the intelligent vehicle through the command issuance module, so as to change the data type, transmission status and mode in time.

[0069] Further, in step S3, after the present invention completes the establishment of the digital twin network, it is necessary to extract some key features of the social relationship of the vehicle and build a vehicle social network layer (vehicle social model). Extracting key features The present invention considers using LSTM to mine social features from the collected data. The vehicle social network model is mainly stored in the module for storing digital twin data, and will be updated regularly based on the obtained data information.

[0070] Specifically, a vehicle with a caching function can act as a content carrier and forward the cached data to the requesting vehicle through vehicle-to-vehicle communication (V2V). In collaborative offloading, in order to make full use of V2V content interaction, consider utilizing the social relationships of vehicles. When the content of the requester and the supplier is consistent and the two parties can establish a transmission channel, the present invention refers to it as having a social association. Therefore, the social relationship of the vehicle has two relationship characteristics, one is the content matching of the two supply parties, and the other characteristic is the communication contact rate of the vehicle.

[0071] In the present invention, LSTM is used to extract the relationship features of the vehicle social network, such as Figure 3As shown in the figure, in the LSTM model, there are input layers, hidden layers and output layers, where the output layer obtains vehicle traffic information and content demand information in different areas from the established digital twin, and then passes the entire information to the neurons in the hidden layer, and finally obtains the elements in the vehicle social network layer. The training is carried out according to the collected historical data, and the parameters of the neurons are periodically updated.

[0072] The present invention uses probability β to describe content matching elements. It is assumed that the vehicle in the system needs G content, and the probability of its demand is β = {β1, β2, β3, ..., β G},in When in an area with cached content, the probability of encountering a vehicle that needs this content is β g .

[0073] The communication contact rate generally depends on the distribution density and driving speed of vehicles in the area. Assuming that the vehicle is traveling at a certain speed v, the prescribed speed in this section is v0, and the average distance between two consecutive vehicles can be expressed as:

[0074]

[0075] In the formula, d0 represents the safe distance between two vehicles, T0 represents the reaction time of the vehicle, and δ represents the speed change factor of the vehicle. According to the coverage and average distance of the communication area, the vehicle distribution density λ of the communication area is obtained. Therefore, the output content of an area in the vehicle social network can be considered to be expressed as {β,λ}.

[0076] Further, in step S4, if Figure 4 As shown in the figure, each cache cloud consists of multiple vehicles in this area, which are responsible for carrying a certain type of content. Since each vehicle has different content caches, each vehicle, like a person, has multiple different social circles, that is, there are multi-dimensional mapping relationships. The size of content storage is limited by its own storage space. Therefore, when communicating, the number of vehicles that content demanders obtain caches is restricted by two factors: one is the size of the provider's cache content, and the other is the transmission rate between the two.

[0077] The communication between vehicles, between vehicles and roadside units, and between vehicles and storage devices is expressed as hybrid communication V2X. The signal-to-noise ratio formula of V2X communication is:

[0078]

[0079] Where SINR stands for signal-to-noise ratio, d stands for path length, α stands for path loss factor, and p w Represents the transmission power of the vehicle, P vrepresents the additive noise power of the vehicle. Therefore, assuming that γ min is the minimum signal-to-noise ratio that can be decoded by the receiver when receiving data. In an ideal environment without co-channel interference, the maximum distance of V2X communication can be expressed as:

[0080]

[0081] The average rate of vehicle V2X communication is defined as:

[0082] R v =Blog2(1+SINR)

[0083] Wherein, B represents the channel bandwidth.

[0084] Further, in step S5, the cache arrangement will depend on the vehicle social network layer obtained from the digital twin social network. The more information the digital twin network collects, the higher the accuracy of the vehicle social network model. The V2X hybrid communication method is adopted in the process of information collection, which may incur additional communication costs. Therefore, the balance between communication costs, model accuracy and system utility needs to be considered in the scheduling of cache.

[0085] Let x g represents the probability of pre-storing the g-type content in the vehicle-mounted cache cloud, y g represents the probability of pre-storing class g content on the RSU server. Therefore, according to the proposed optimal edge caching problem, the utility of the cache system is maximized under the constraints of cache size and transmission delay, which can be expressed as:

[0086]

[0087] In the formula, N represents the set of vehicles, M represents the set of RSUs, and ψ g (ε g ) represents the accuracy of the vehicle social network layer, ε g is the amount of system information collected by the digital twin for training the LSTM network and obtaining the vehicle social network layer, β g is the probability of encountering a vehicle that needs category g content, It indicates the time it takes for the vehicle to obtain g-type content from the vehicle cache cloud. It represents the time taken by the vehicle to obtain content from RSUr, y g,r represents the probability of pre-storing the g-type content on the RSUr server, ζ v is the transmission cost of vehicle v, ζ r is the transmission cost between vehicle and RSUr, ζ cis the transmission cost requested by the vehicle to the data center. {Z} is the indicator function, which is 1 when {Z} is true and 0 when {Z} is false.

[0088] In order to solve this problem, the present invention proposes an iterative method based on reinforcement learning. In each iteration, the cache scheduling strategy is first obtained according to a given vehicle social network layer, and then the amount of information collected in the model construction is modified according to the determined strategy, and the iteration continues until the system converges.

[0089] According to the given vehicle social network layer and precision function ψ g (ε g ), explore the optimal caching strategy under this model. The utility function in this environment is expressed as:

[0090]

[0091] In order to maximize the effectiveness of the edge cache system, the present invention seeks the optimal cache strategy π * Since the optimal strategy is affected by many factors, the present invention proposes a vehicle-mounted cache cloud architecture and edge cache solution based on DDPG (deep deterministic policy gradient algorithm).

[0092]

[0093] Among them, η∈(0,1) represents the discount factor, which is used to weigh the utility function.

[0094] The architecture of DDPG is as follows Figure 5 As shown in the figure, the learning system of DDPG mainly consists of two parts, namely the main network and the target network. These two networks have similar parts, each network has an actor part and a critic part, and both networks here are deep neural networks. In the learning process of DDPG, the role of the actor part of the main network is mainly to explore edge caching strategies, while the critic network helps the actor to obtain better strategies through gradient methods. The target network can be regarded as an iteration of the main network, generating target values ​​for training the main network, and finally updating the strategy by calculating the loss function. The experience replay pool stores the learning experience of action rewards and state changes, which is used to train the actor and critic parameters.

[0095] DDPG is applied to the scenario of edge cache collaborative management. The Markov decision process takes the form of a reinforcement learning problem. Set the given cache strategy to π, and its state S i and Action A i The average system utility obtained under this condition can be expressed as:

[0096]

[0097] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. A vehicle social network caching method based on digital twins, characterized by: The method comprises the following steps: S1, extract the connection between vehicle network and social network; S2. Map the vehicle network from the physical layer to the twin network layer based on digital twin technology; S3, extracting the relationship features of vehicles from the twin network layer and establishing a vehicle social network layer; S4, establishing a cache cloud in the vehicle social network layer; S5. Determine a vehicle cache collaboration strategy based on the vehicle social network layer and the cache cloud therein.

2. The vehicle social network caching method based on digital twin according to claim 1 is characterized in that: In step S1, the vehicle is affected by the driver's social activities and has social attributes. The social relations, user interests and social activities in the social network are the links between the vehicle network and the social network; personalized recommendations are made to the drivers in the vehicle network based on the user interests in the social network; the social relations in the social network correspond to the network topology in the vehicle network, and the network topology in the vehicle network is predicted and maintained based on the social relations, and the social relations in the social network are extended based on the network topology; the mobility model in the vehicle network abstracts the social activities in the social network, and the mobility model is predicted or discovered based on the group nature of the social activities.

3. The vehicle social network caching method based on digital twin according to claim 2 is characterized in that: In step S2, in the physical layer, it is assumed that there are N moving vehicles and M roadside units RSU on the road, the set of vehicles is defined as N = {1, 2, ..., N}, and the set of RSUs is defined as M = {1, 2, ..., M}; The vehicle demand content is divided into G categories, and each type of demand content is expressed as where f g Indicates the size of the content type, Indicates the maximum delay of its content, μ g Indicates the sensitivity coefficient of its delay; Each RSU is equipped with an edge cache server, and each edge cache server has caching capabilities; the vehicle's required content is pre-stored in the vehicle or RSU's edge cache server.

4. The vehicle social network caching method based on digital twin according to claim 3 is characterized in that: In step S2, the twin network layer includes an information collection module, a control module, an RSU collaboration module, an instruction issuing module, and a digital twin data storage module; The information acquisition module obtains the vehicle status through communication between the vehicle and the RSU; the control module is used to determine the update cycle and the interaction frequency and data type during the adjustment process, and the adjustment command is transmitted to the corresponding vehicle through the command issuing module; The RSU collaboration module is used to exchange data and update status in a timely manner; The digital twin data storage module is used to store data.

5. The vehicle social network caching method based on digital twin according to claim 4 is characterized in that: In step S3, the LSTM network is used to extract the social relationship of the vehicle, including at least two relationship features, which are the content matching of the two parties and the communication contact rate of the vehicle; The probability β is used to describe the content matching element. Suppose the vehicle needs G content, and the probability of its demand is β = {β1,β2,β3,...,β G },in The probability that the vehicle's demand content matches the demand content in the communication area is β g ; The communication contact rate depends on the distribution density and driving speed of vehicles in the area; assuming that the vehicle is traveling at a speed v, the prescribed speed in this section is v0, and the average distance between two consecutive vehicles is expressed as: Where d0 represents the safe distance between two vehicles, T0 represents the reaction time of the vehicle, and δ represents the speed change factor of the vehicle; according to the coverage range and average distance of the communication area, the vehicle distribution density λ of the communication area is obtained; Then the two relational features of the social relationship of the vehicle are expressed as {β,λ}.

6. The vehicle social network caching method based on digital twin according to claim 5 is characterized in that: In step S4, each cache cloud is composed of a number of vehicles in the area, and each cache cloud includes a demand content; each vehicle includes a number of demand contents, and each demand content corresponds to a cache cloud; thus, a multi-dimensional mapping is formed between the cache cloud and the vehicle; When communicating, the number of vehicles that a content demander can obtain from the cache is constrained by two factors: one is the size of the cache content provided by the provider, and the other is the transmission rate between the two. The signal-to-noise ratio formula for hybrid communication V2X communication is: Where SINR stands for signal-to-noise ratio, d stands for path length, α stands for path loss factor, and p w Represents the transmission power of the vehicle, P v represents the additive noise power of the vehicle; Assumption γ min is the minimum signal-to-noise ratio that can be decoded by the receiver when receiving data. In an ideal environment without co-channel interference, the maximum distance of V2X communication is expressed as: The average rate of vehicle V2X communication is defined as: R v =B log2(1+SINR) Wherein, B represents the channel bandwidth.

7. The vehicle social network caching method based on digital twin according to claim 6 is characterized in that: In step S5, cache scheduling considers the balance between communication cost, model accuracy, and system utility; Let x g represents the probability of pre-storing content of category g on the cache cloud, y g represents the probability of pre-storing class g content on the RSU server; under the constraints of cache size and transmission delay, the optimal edge cache problem that maximizes the utility of the cache system is expressed as: Where N is the set of vehicles; M is the set of RSUs; ψ g (ε g ) represents the accuracy of the vehicle social network layer; ε g The amount of system information collected in the digital twin for training the LSTM network and obtaining the vehicle social network layer; β g The probability of encountering a vehicle that requires Category G content; Indicates the time when the vehicle obtains the g-category content from the cache cloud; represents the time it takes for the vehicle to obtain content from RSUr; g,r represents the probability of pre-storing the g-type content on the RSUr server; v is the transmission cost of vehicle v; r is the transmission cost between vehicle and RSUr; c is the transmission cost requested by the vehicle to the data center; {Z} is the indicator function, which is 1 when {Z} is true and 0 when {Z} is false.

8. The vehicle social network caching method based on digital twin according to claim 7 is characterized in that: In step S5, an optimal cache strategy is set in an iterative manner based on reinforcement learning to solve the optimal edge cache problem. In each iteration, a cache scheduling strategy is first obtained according to a given vehicle social network layer, and then the amount of information collected during the construction of the vehicle social network layer is modified according to the determined strategy, and the iteration is continued until the system converges; According to the given vehicle social network layer and precision function ψ g (ε g ), the utility function for exploring the optimal decision is expressed as: Optimal caching strategy π * It is expressed as: Among them, η∈(0,1) represents the discount factor, which is used to weigh the utility function.

9. The vehicle social network caching method based on digital twin according to claim 8 is characterized in that: In step S5, DDPG is applied to the edge cache collaborative management scenario. For a given cache strategy set to π, its state S i and Action A i The average system utility obtained under this condition is expressed as: The DDPG algorithm includes a main network and a target network; each network has an actor part and a critic part. The target network is regarded as an iteration of the main network, generating target values ​​for training the main network, and finally updating the strategy by calculating the loss function; the experience replay pool of the DDPG algorithm stores the learning experience of action rewards and state changes, which is used to train the actor and critic parameters.

Images