A vehicle-assisted edge digital twin aggregation method based on information age matching

By adopting the vehicle-assisted edge digital twin aggregation method with information age matching in 6G network, the virtual and real mapping error problem caused by data transmission and processing delay of digital twins in 6G network is solved, and high-reliability and low-latency communication is achieved, improving The accuracy and efficiency of digital twins.

CN115499869BActive Publication Date: 2024-11-12GUANGDONG UNIV OF TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202211065754.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-09-01
Publication Date
2024-11-12
Estimated Expiration
2042-09-01

AI Technical Summary

Technical Problem

In 6G networks, due to data transmission and processing delays, there is a large deviation between the actual value and the estimated value of the digital twin, which affects the optimization decision of the digital twin. In addition, the computing and communication resources of the wireless network are limited, making it difficult to ensure high-independence realization Virtual model.

Method used

Using the vehicle-assisted edge digital twin aggregation method based on information age matching, the objective function and optimization target model are constructed, and the information age of the digital twin is minimized as the optimization goal, and converted into a bilateral matching model based on the problem structure to obtain stable matching results. Carry out digital twin aggregation.

Benefits of technology

It effectively reduces the virtual and real mapping error of digital twins, realizes high reliability and low latency communication, and improves the accuracy and efficiency of digital twins.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115499869B_ABST
    Figure CN115499869B_ABST
Patent Text Reader

Abstract

The present invention provides a vehicle auxiliary edge digital twin aggregation method based on information age matching, comprising the following steps: S1: constructing an objective function according to a digital twin communication system; S2: taking minimizing the information age of the digital twin as the optimization goal, constructing an optimization target model according to the objective function; S3: converting the optimization target model into a bilateral matching model based on the problem structure to obtain a stable matching result; S4: performing digital twin aggregation based on the stable matching result. The present invention provides a vehicle auxiliary edge digital twin aggregation method based on information age matching, which can effectively reduce the virtual-real mapping error of the digital twin, realize highly reliable and low-latency communication, and solve the problem that there is a large deviation between the actual value and the estimated value of the digital twin due to the existence of non-negligible data transmission and processing delays.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of digital twin aggregation, and more specifically, to a vehicle auxiliary edge digital twin aggregation method based on information age matching. Background Art

[0002] With the explosive development of the Internet of Things (IoT) and 5G such as smart transportation and smart cities, the research on the sixth generation mobile network (6G) has been put on the agenda. Compared with 5G, 6G shows better performance in data rate, service latency and edge intelligence. However, more than billions of end users in the Internet of Things will be connected to the 6G network and generate a large amount of data, which requires edge servers to provide rich computing resources and communication resources for processing. Therefore, the imbalance between user demand and edge server resources is the main challenge for 6G to provide high-quality services for emerging and new applications.

[0003] As a promising technology for achieving high reliability and low latency communications in 6G, digital twins combine physical sensors, machine learning algorithms and software analysis with spatial network graphics to create a real-time digital simulation model of physical entities in the server. This model maintains real-time interactive connection with physical entities in the real environment, while using virtual models to simulate the behavior of physical entities in reality. By building a digital twin model of physical devices on the server, the reliability of the system can be improved, thereby providing more optimized decisions for physical entities.

[0004] Digital twins reflect the mapping relationship between virtual reality and the physical world and the virtual world. The accuracy of the virtual model directly affects the effectiveness of the digital twin. However, due to the non-negligible data transmission and processing delays, there is a large deviation between the actual value and the estimated value of the digital twin, making it difficult to ensure a high-fidelity virtual model. This virtual-real mapping error greatly affects the optimization decisions made by the digital twin. At the same time, in order to synchronize real-time data and establish corresponding models, the maintenance of digital twins requires a large amount of computing and communication resources, but the computing and communication resources in wireless networks are very limited. Summary of the invention

[0005] In order to overcome the current technical defects that there is a large deviation between the actual value and the estimated value of the digital twin due to the existence of non-negligible data transmission and processing delays, the present invention provides a vehicle auxiliary edge digital twin aggregation method based on information age matching.

[0006] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0007] A vehicle-assisted edge digital twin aggregation method based on information age matching includes the following steps:

[0008] S1: Construct objective function based on digital twin communication system;

[0009] S2: Taking minimizing the age of information (AoI) of digital twins as the optimization goal, the optimization target model is constructed according to the objective function;

[0010] S3: Convert the optimization target model into a bilateral matching model based on the problem structure to obtain a stable matching result;

[0011] S4: Digital twin aggregation based on stable matching results.

[0012] In the above scheme, by introducing the age of information (AoI) indicator, an optimization target model is constructed with the minimization of the AoI of the digital twin as the optimization goal, and a matching algorithm is introduced to transform the optimization target model into a bilateral matching model based on the problem structure, and a stable matching result is obtained to perform digital twin aggregation, thereby effectively reducing the virtual-to-real mapping error of the digital twin and achieving high reliability and low-latency communication.

[0013] Preferably, the digital twin communication system includes I terminal users, M base stations and V parked vehicles, and the base stations include overloaded base stations and non-overloaded base stations; wherein the terminal users communicate in association with the base stations, the overloaded base stations communicate in association with the parked vehicles, and the terminal users communicate in association with the parked vehicles.

[0014] Preferably, the information age of the digital twin refers to the time from when the data in the terminal user is collected to when it is called to update the digital twin.

[0015] Preferably, the objective function is:

[0016]

[0017] Among them, β v,i represents a binary dependent variable, when β v,i =1 indicates the digital twin DT of the end user i is selected to be associated with the parked vehicle v when β v,i =0 indicates the digital twin DT of the end user i Not selected to be associated with the parked vehicle v;

[0018] Represents digital twin DT i The age of the information when it is associated with the parked vehicle v,

[0019]

[0020] represents the time required for the physical entity to collect data, Indicates uploading the historical data D of the digital twin to the parked vehicle vi The transmission delay of represents the end-user digital twin communication rate associated with the parked vehicle v, W is the channel bandwidth, and N 0 is the Gaussian white noise power, p v,i Parked Vehicle to Digital Twin DT i Transmitting power, g v,i Parked Vehicle to Digital Twin DT i The channel gain, represents the time delay consumed by updating the digital twin based on the uploaded data, f v is the CPU frequency of the parked vehicle v, r v is the number of CPU cycles required to process a data sample, τ v represents the parking time of the parked vehicle v.

[0021] Preferably, the optimization target model is:

[0022]

[0023] stC1:

[0024] C2:

[0025] C3:

[0026] C4:

[0027] variables:γ m,v ∈{0,1},β v,i ∈{0,1}

[0028] Among them, γ m,v and β v,i is a binary association variable, and constraint C1 ensures that the maximum number of twins associated with a parked vehicle v and an end user is Constraint C2 indicates that the end-user twin can only be associated with one parked vehicle, constraint C3 ensures that the parked vehicle has enough time to process the associated end-user twin, and constraint C4 ensures that the parked vehicle v has sufficient time to process the associated end-user twin. i The minimum transmission rate R 1 .

[0029] Preferably, the matching process of the bilateral matching model based on the problem structure is specifically as follows:

[0030] S3.1: Set all parked vehicles and end users to unmatched status;

[0031] S3.2: For the unmatched terminal user i, find the best preference in its preference list and obtain the corresponding parking vehicle v with the best preference * ,Right now And to the optimal preferred parking vehicle v * Send matching request; F i (v) represents the preferences of end user i;

[0032] S3.3: The parked vehicle v collects all the end users who send matching requests and pairs them to form a requester set Y v ;

[0033] If the number of parked vehicles v exceeds its capacity, then v Find the end user i with the worst preference * ,Right now and cancel the parking vehicle v and the end user i with the worst preference * Pairing (i * ,v);F v (i) represents the preference for parking vehicle v;

[0034] If the matching of the parked vehicle v just reaches its own capacity at this time, for the new requester r, the parked vehicle v in Y v Find the end user i with the worst preference * , if F v (i * ) <F v (r), then cancel the pairing (i * ,v), and establish a pairing (r,v);

[0035] S3.4: For the terminal user whose pairing is cancelled, delete the best preferred parked vehicle in its preference list and reset it to the unmatched state, and return to step S3.2 until a stable match is achieved.

[0036] Preferably, when a valid match does not contain any blocking pairs, it is considered a stable match; wherein,

[0037] A valid match must meet the following conditions:

[0038] A1: Each digital twin can only match one parked vehicle, that is, |Φ(i)|≤1;

[0039] A2: Parked vehicles are associated with multiple digital twins, but the number of matches is limited, i.e.

[0040] A3: For matching relations, Φ(i) = {v∈V|i→v∈Φ} and Φ(v) = {i∈I|v→i∈Φ};

[0041] When a pair (i, v) blocks the matching Φ, it is considered a blocking pair. The conditions for satisfying the blocking pair are as follows:

[0042] B1: End user i and parked vehicle v have mutual preferences, i∈F v And v∈F i ;

[0043] B2: End user i has no pairing, or prefers parking vehicle v over matching Φ(i) = v';

[0044] B3: The parked vehicle v does not reach the pairing capacity, or relative to the pairing i'∈Φ -1 (v), parked vehicle v is more biased towards end user i;

[0045] Among them, v' and v represent different parked vehicles, and i' and i represent different end users.

[0046] Preferably, the preference of the digital twin is the digital twin DT associated with the parked vehicle v i The corresponding transmission rate:

[0047] F i (v) = R v,i

[0048] The preference of the parked vehicle v is the digital twin DT of interest at the parking time i Information age:

[0049]

[0050] Preferably, the method further includes obtaining the best association between the overloaded base station and the parked vehicle through an exhaustive search algorithm.

[0051] Preferably, the specific process of the exhaustive search algorithm is:

[0052] D1: Construct a feasible solution set γ for the optimization objective;

[0053] D2: For one possible outcome, calculate its result;

[0054] D3: Determine whether the result meets the requirements. If not, return to step D2 to search for the next possible situation. If it meets the requirements, obtain the optimal decision variable γ m,v .

[0055] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0056] The present invention provides a vehicle auxiliary edge digital twin aggregation method based on information age matching. By introducing the age of information (AoI) indicator, an optimization target model is constructed with minimizing the AoI of the digital twin as the optimization goal, and a matching algorithm is introduced to transform the optimization target model into a bilateral matching model based on the problem structure, so as to obtain a stable matching result for digital twin aggregation, thereby effectively reducing the virtual-to-real mapping error of the digital twin and realizing high-reliability and low-latency communication. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a flowchart of the implementation steps of the technical solution of the present invention;

[0058] Figure 2 This is a schematic diagram of the scenario of vehicle-assisted edge digital twin aggregation in the present invention. DETAILED DESCRIPTION

[0059] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;

[0060] In order to better illustrate the present embodiment, some parts in the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product;

[0061] It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0062] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0063] Example 1

[0064] like Figure 1 As shown, a vehicle-assisted edge digital twin aggregation method based on information age matching includes the following steps:

[0065] S1: Construct objective function based on digital twin communication system;

[0066] S2: Taking minimizing the age of information (AoI) of digital twins as the optimization goal, the optimization target model is constructed according to the objective function;

[0067] S3: Convert the optimization target model into a bilateral matching model based on the problem structure to obtain a stable matching result;

[0068] S4: Digital twin aggregation based on stable matching results.

[0069] In the specific implementation process, by introducing the age of information (AoI) indicator, an optimization target model is constructed with the minimization of the AoI of the digital twin as the optimization goal, and a matching algorithm is introduced to transform the optimization target model into a bilateral matching model based on the problem structure, and a stable matching result is obtained to perform digital twin aggregation, thereby effectively reducing the virtual-to-real mapping error of the digital twin and achieving high reliability and low-latency communication.

[0070] Example 2

[0071] A vehicle-assisted edge digital twin aggregation method based on information age matching includes the following steps:

[0072] S1: Construct objective function based on digital twin communication system;

[0073] More specifically, the digital twin communication system includes I terminal users (such as devices and mobile devices in the Internet of Things, etc.), M base stations and V parked vehicles, and the base stations include overloaded base stations and non-overloaded base stations; wherein the terminal users communicate in association with the base stations, the overloaded base stations communicate in association with the parked vehicles, and the terminal users communicate in association with the parked vehicles.

[0074] S2: Taking minimizing the age of information (AoI) of digital twins as the optimization goal, the optimization target model is constructed according to the objective function;

[0075] More specifically, the information age of the digital twin refers to the time from when the data in the terminal user is collected to when it is called to update the digital twin.

[0076] More specifically, the objective function is:

[0077]

[0078] Among them, β v,i represents a binary dependent variable, when β v,i =1 indicates the digital twin DT of the end user i is selected to be associated with the parked vehicle v when β v,i =0 indicates the digital twin DT of the end user i Not selected to be associated with the parked vehicle v;

[0079] Represents digital twin DT i The age of the information when it is associated with the parked vehicle v,

[0080]

[0081] represents the time required for the physical entity to collect data, Indicates uploading the historical data D of the digital twin to the parked vehicle vi The transmission delay of represents the end-user digital twin communication rate associated with the parked vehicle v, W is the channel bandwidth, and N 0 is the Gaussian white noise power, p v,i Parked Vehicle to Digital Twin DT i Transmitting power, g v,i Parked Vehicle to Digital Twin DT i The channel gain, represents the time delay consumed by updating the digital twin based on the uploaded data, f v is the CPU frequency of the parked vehicle v, r v is the number of CPU cycles required to process a data sample, τ v represents the parking time of the parked vehicle v.

[0082] More specifically, the optimization target model is:

[0083]

[0084] stC1:

[0085] C2:

[0086] C3:

[0087] C4:

[0088] variables:γ m,v ∈{0,1},β v,i ∈{0,1}

[0089] Among them, γ m,v and β v,i is a binary association variable, and constraint C1 ensures that the maximum number of twins associated with a parked vehicle v and an end user is Constraint C2 indicates that the end-user twin can only be associated with one parked vehicle, constraint C3 ensures that the parked vehicle has enough time to process the associated end-user twin, and constraint C4 ensures that the parked vehicle v has sufficient time to process the associated end-user twin. i The minimum transmission rate R 1 .

[0090] S3: Convert the optimization target model into a bilateral matching model based on the problem structure to obtain a stable matching result;

[0091] More specifically, the matching process of the bilateral matching model based on the problem structure is as follows:

[0092] S3.1: Set all parked vehicles and end users to unmatched status;

[0093] S3.2: For the unmatched terminal user i, find the best preference in its preference list and obtain the corresponding parking vehicle v with the best preference * ,Right now And to the optimal preferred parking vehicle v * Send matching request; F i (v) represents the preferences of end user i;

[0094] To achieve bilateral matching, each digital twin must build its preference list by ranking the parked vehicles on the other side according to preference;

[0095] More specifically, the preference of the digital twin is the digital twin DT associated with the parked vehicle v. i The corresponding transmission rate:

[0096] F i (v) = R v,i

[0097] S3.3: The parked vehicle v collects all the end users who send matching requests and pairs them to form a requester set Y v ;

[0098] If the number of parked vehicles v exceeds its capacity, then v Find the end user i with the worst preference * ,Right now and cancel the parking vehicle v and the end user i with the worst preference * Pairing (i * ,v);F v (i) represents the preference for parking vehicle v;

[0099] The parked vehicle v is more concerned about the quality of the digital twin. At the same time, the parked vehicle v considers its own parking time. Therefore, the preference of the parked vehicle v is defined as the digital twin DT concerned under the parking time. i Age of Information AoI:

[0100]

[0101] If the matching of the parked vehicle v just reaches its own capacity at this time, for the new requester r, the parked vehicle v in Y v Find the end user i with the worst preference * , if F v (i * ) <F v (r), then cancel the pairing (i * ,v), and establish a pairing (r,v);

[0102] S3.4: For the terminal user whose pairing is cancelled, delete the best preferred parked vehicle in its preference list and reset it to the unmatched state, and return to step S3.2 until a stable match is achieved.

[0103] More specifically, when a valid match does not contain any blocking pairs, it is considered a stable match;

[0104] A valid match must meet the following conditions:

[0105] A1: Each digital twin can only match one parked vehicle, that is, |Φ(i)|≤1;

[0106] A2: A parked vehicle can be associated with multiple digital twins, but the number of matches is limited, i.e.

[0107] A3: For matching relations, Φ(i) = {v∈V|i→v∈Φ} and Φ(v) = {i∈I|v→i∈Φ};

[0108] When a pair (i, v) blocks the matching Φ, it is considered a blocking pair. The conditions for satisfying the blocking pair are as follows:

[0109] B1: End user i and parked vehicle v have mutual preferences, i∈F v And v∈F i ;

[0110] B2: End user i has no pairing, or prefers parking vehicle v over matching Φ(i) = v';

[0111] B3: The parked vehicle v does not reach the pairing capacity, or relative to the pairing i'∈Φ -1 (v), parked vehicle v is more biased towards end user i;

[0112] Among them, v' and v represent different parked vehicles, and i' and i represent different end users.

[0113] S4: Digital twin aggregation based on stable matching results.

[0114] In the specific implementation process, by reasonably associating parked vehicles or base stations, the AoI of digital twins can be optimized, which can effectively reduce the virtual-real mapping error of digital twins, effectively reduce the deviation or data error in information interaction caused by perceived data transmission errors and delays in the twin network model mapping process, and improve the accuracy of the twin network model. In addition, the aggregation problem of edge digital twins is considered as a many-to-one bilateral matching problem. Through the process of many-to-one matching algorithm, a stable matching result between the base station and the end user is obtained. The base station aggregates the digital twin models of all associated users based on the result. However, considering the load of the base station, the digital twin models aggregated and updated by the base station are limited, and too many digital twins are associated, resulting in overload of the base station. Therefore, by establishing an association between the base station and the vehicle, the parked vehicle is scheduled as a general computing node to assist the overloaded base station in aggregating the digital twin model, and the results of the aggregated digital twin update are fed back to the associated overloaded base station, thereby effectively alleviating the congestion of the overloaded base station during peak hours.

[0115] Example 3

[0116] A vehicle-assisted edge digital twin aggregation method based on information age matching includes the following steps:

[0117] S1: Construct objective function based on digital twin communication system;

[0118] More specifically, Figure 2 As shown, the digital twin communication system includes I terminal users (such as devices and mobile devices in the Internet of Things, etc.), M base stations and V parked vehicles, and the base stations include overloaded base stations and non-overloaded base stations; wherein the terminal users are associated with the base stations for communication, the overloaded base stations are associated with the parked vehicles for communication, and the terminal users are associated with the parked vehicles for communication.

[0119] Figure 2 The scenario shown is mainly divided into two parts. The first part is the association between digital twins and base stations. Each base station covers a certain range. There are a certain number of users within the coverage range. These users can be mobile devices, IoT devices, etc. Each user generates its own digital twin with the help of a nearby AP and associates the twin with the base station. The base station aggregates the uploaded digital twins to optimize each digital twin model, thereby reducing the computational overhead caused by the co-evolution between the physical device and its twin. The second part is the association between digital twins and vehicles. Since the base station needs to handle a large number of services at the same time, its limited resources result in a limited number of base station associations, while parked vehicles have abundant unused resources. For this reason, digital twins without associated users can be associated with parked vehicles. Vehicles aggregate the uploaded digital twins to update and maintain the digital twin model, and at the same time feed back the result to the base station associated with it.

[0120] The data of end user i∈I is:

[0121]

[0122] Among them, D i Indicates the size of the data, x i is the data collected by end user i, y i is x i Data labels.

[0123] DT for digital twins of end users i It is expressed as:

[0124] DT i =(M i , D i ,s t )

[0125] Among them, M i is the behavior model of the digital twin, D i is the operational data required to run the digital twin, s t Indicates the real-time status of the physical entity, D i and t It is the basic data required to run digital twin applications.

[0126] The first part is the communication between the digital twin and the base station, from the end user's digital twin DT i The uplink data rate that can be achieved to base station m is:

[0127] R m,i =α m,i Wlog 2 (1+SINR m,i )

[0128] Among them, SINR m,i is the signal-to-noise ratio, p m,i From base station m to digital twin DT i Transmitting power, g m,i From base station m to digital twin DT i The channel gain, W is the channel bandwidth, N 0 is the Gaussian white noise power. m,i is a binary dependent variable, when α m,i =1, it means selecting digital twin DT i Associated with base station m, when α m,i =0, indicating digital twin DT i Not selected to associate with base station m.

[0129] The second part is the communication between the overloaded base station and the parked vehicle. The parked vehicle has abundant computing resources. In order to relieve the burden of the overloaded base station in processing digital twins, the parked vehicle is associated with the overloaded base station. At the same time, the parked vehicle feeds back the processed digital twin to the overloaded base station, and the base station performs combined processing based on the uploaded digital twin model. 2 The achievable data rate of the parked vehicle v(v∈V) associated with

[0130] R m,v =γ m,v Blog 2 (1+SINR m,v )

[0131] Among them, p m,v is the transmission power from base station m to parked vehicle v, g m,v is the channel gain from base station m to parked vehicle v, B is the channel bandwidth, N 0 is the Gaussian white noise power. m,v is a binary dependent variable, when γ m,v =1, it means that the parked vehicle v is associated with the base station m. m,v =0, it indicates that the parked vehicle v is not selected to be associated with the base station m.

[0132] More specifically, it also includes obtaining the best association between the overloaded base station and the parked vehicle through an exhaustive search algorithm.

[0133] More specifically, the specific process of the exhaustive search algorithm is:

[0134] D1: Construct a feasible solution set γ for the optimization objective;

[0135] D2: For one possible outcome, calculate its result;

[0136] D3: Determine whether the result meets the requirements. If not, return to step D2 to search for the next possible situation. If it meets the requirements, obtain the optimal decision variable γ m,v .

[0137] In a specific implementation process, an exhaustive search algorithm is used to solve the association problem between the base station and the parked vehicles, so as to obtain the best association between the overloaded base station and the parked vehicles.

[0138] The third part is the communication between the digital twin and the parked vehicle. The parked vehicle associated with the overload base station establishes an association with the end user through V2X communication. The communication rate of the end user digital twin associated with the parked vehicle v is:

[0139]

[0140] p v,i Parked Vehicle to Digital Twin DT i Transmitting power, g v,i Parked Vehicle to Digital Twin DT i The channel gain, β v,i is a binary dependent variable, when β v,i =1, it means selecting digital twin DT i Associated with the parked vehicle v, when β v,i =0, indicating digital twin DT i Not selected to be associated with the parked vehicle v.

[0141] S2: Taking minimizing the age of information (AoI) of digital twins as the optimization goal, the optimization target model is constructed according to the objective function;

[0142] More specifically, the information age of the digital twin refers to the time from when the data in the terminal user is collected to when it is called to update the digital twin.

[0143] In the specific implementation process, the age of information (AoI) indicator is introduced to reduce the virtual-reality mapping error of digital twins.

[0144] The latency of terminal user i collecting a set of data is related to its CPU frequency. Therefore, the latency of user i collecting data is:

[0145]

[0146] where f i col Indicates the CPU frequency of user i collecting data, F i col represents the total number of floating-point operations required by user i to collect a set of data each time, C i is the number of floating-point operations per CPU cycle for user i.

[0147] When a parked vehicle is associated with a digital twin, data is directly transferred from the user device to the vehicle. To this end, the historical data D of the digital twin is uploaded to the parked vehicle v. i The transmission delay is:

[0148]

[0149] The time delay consumed by updating the digital twin based on the uploaded data is:

[0150]

[0151] Among them, f v is the CPU frequency of the parked vehicle v, r vThe number of CPU cycles required to process a data sample.

[0152] To this end, the entire AoI takes into account the time required for the physical entity to collect data, the latency of data processing, and the latency of the twin to upload historical data. i When associated with a parked vehicle v, the AoI is defined as:

[0153]

[0154] More specifically, since the parking time of a vehicle is limited, in order to better associate the parked vehicle with the digital twin, the parking time of the parked vehicle cannot be less than the time it takes to transmit data. v represents the parking time of the parked vehicle v. Therefore, in order to better maintain the digital twin, the parking time should satisfy Considering the parking time of the parked vehicle, the parked vehicle focuses on the quality of the digital twin. Therefore, the objective function of the parked vehicle associated digital twin is:

[0155]

[0156] More specifically, in order to improve the timeliness of digital twins, it is necessary to always pay attention to the quality of the digital twin model. To this end, the optimization goal is to minimize the AoI of the digital twin. Therefore, the optimization goal model is:

[0157]

[0158] stC1:

[0159] C2:

[0160] C3:

[0161] C4:

[0162] variables:γ m,v ∈{0,1},β v,i ∈{0,1}

[0163] Among them, γ m,v and β v,i is a binary association variable, and constraint C1 ensures that the maximum number of twins associated with a parked vehicle v and an end user is Constraint C2 indicates that the end-user twin can only be associated with one parked vehicle, constraint C3 ensures that the parked vehicle has enough time to process the associated end-user twin, and constraint C4 ensures that the parked vehicle v has sufficient time to process the associated end-user twin. i The minimum transmission rate R 1.

[0164] S3: Introducing matching theory to transform the optimization target model into a bilateral matching model based on the problem structure, and obtaining a stable matching result;

[0165] More specifically, the matching process of the bilateral matching model based on the problem structure is as follows:

[0166] S3.1: Set all parked vehicles and end users to unmatched status;

[0167] S3.2: For the unmatched terminal user i, find the best preference in its preference list and obtain the corresponding parking vehicle v with the best preference * ,Right now And to the optimal preferred parking vehicle v * Send matching request; F i (v) represents the preferences of end user i;

[0168] To achieve bilateral matching, each digital twin must build its preference list by ranking the parked vehicles on the other side according to preference;

[0169] More specifically, the preference of the digital twin is the digital twin DT associated with the parked vehicle v. i The corresponding transmission rate:

[0170] F i (v) = R v,i

[0171] S3.3: The parked vehicle v collects all the end users who send matching requests and pairs them to form a requester set Y v ;

[0172] If the number of parked vehicles v exceeds its capacity, then v Find the end user i with the worst preference * ,Right now and cancel the parking vehicle v and the end user i with the worst preference * Pairing (i * ,v);F v (i) represents the preference for parking vehicle v;

[0173] The parked vehicle v is more concerned about the quality of the digital twin. At the same time, the parked vehicle v considers its own parking time. Therefore, the preference of the parked vehicle v is defined as the digital twin DT concerned under the parking time. i Age of Information AoI:

[0174]

[0175] If the matching of the parked vehicle v just reaches its own capacity at this time, for the new requester r, the parked vehicle v in Y v Find the end user i with the worst preference * , if F v (i * ) <F v (r), then cancel the pairing (i * ,v), and establish a pairing (r,v);

[0176] S3.4: For the terminal user whose pairing is cancelled, delete the best preferred parked vehicle in its preference list and reset it to the unmatched state, and return to step S3.2 until a stable match is achieved.

[0177] More specifically, when a valid match does not contain any blocking pairs, it is considered a stable match;

[0178] A valid match must meet the following conditions:

[0179] A1: Each digital twin can only match one parked vehicle, that is, |Φ(i)|≤1;

[0180] A2: A parked vehicle can be associated with multiple digital twins, but the number of matches is limited, i.e.

[0181] A3: For matching relations, Φ(i) = {v∈V|i→v∈Φ} and Φ(v) = {i∈I|v→i∈Φ};

[0182] When a pair (i, v) blocks the matching Φ, it is considered a blocking pair. The conditions for satisfying the blocking pair are as follows:

[0183] B1: End user i and parked vehicle v have mutual preferences, i∈F v And v∈F i ;

[0184] B2: End user i has no pairing, or prefers parking vehicle v over matching Φ(i) = v';

[0185] B3: The parked vehicle v does not reach the pairing capacity, or relative to the pairing i'∈Φ -1 (v), parked vehicle v is more biased towards end user i;

[0186] Among them, v' and v represent different parked vehicles, and i' and i represent different end users.

[0187] S4: Digital twin aggregation based on stable matching results.

[0188] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A vehicle-assisted edge digital twin aggregation method based on information age matching, characterized in that: The following steps are involved: S1: constructing an objective function according to a digital twin communication system; the digital twin communication system includes I terminal users, M base stations and V parked vehicles, the base stations include overloaded base stations and non-overloaded base stations; wherein the terminal users are associated with the base stations for communication, the overloaded base stations are associated with the parked vehicles for communication, and the terminal users are associated with the parked vehicles for communication; S2: Taking minimizing the information age of the digital twin as the optimization goal, an optimization target model is constructed according to the objective function; the information age of the digital twin is specifically the time from the completion of data collection in the end user to the time when the data is called to update the digital twin; the objective function is: Among them, β v,i represents a binary dependent variable, when β v,i =1 indicates the digital twin DT of the end user i is selected to be associated with the parked vehicle v when β v,i =0 indicates the digital twin DT of the end user i Not selected to be associated with the parked vehicle v; Represents digital twin DT i The age of the information when it is associated with the parked vehicle v, represents the time required for the physical entity to collect data, Indicates uploading the historical data D of the digital twin to the parked vehicle v i The transmission delay of represents the end-user digital twin communication rate associated with the parked vehicle v, W is the channel bandwidth, N0 is the Gaussian white noise power, and p v,i Parked Vehicle to Digital Twin DT i Transmitting power, g v,i Parked Vehicle to Digital Twin DT i The channel gain, represents the time delay consumed by updating the digital twin based on the uploaded data, f v is the CPU frequency of the parked vehicle v, r v is the number of CPU cycles required to process a data sample, τ v represents the parking time of the parked vehicle v; S3: Convert the optimization target model into a bilateral matching model based on the problem structure to obtain a stable matching result; the optimization target model is: s.t.C1: C2: C3: C4: variables:c m,v ∈{0,1},β v,i ∈{0,1} Among them, γ m,v and β v,i is a binary association variable, and constraint C1 ensures that the maximum number of twins associated with a parked vehicle v and an end user is Constraint C2 indicates that the end-user twin can only be associated with one parked vehicle, constraint C3 ensures that the parked vehicle has enough time to process the associated end-user twin, and constraint C4 ensures that the parked vehicle v has sufficient time to process the associated end-user twin. i The minimum transmission rate R1; The matching process of the bilateral matching model based on the problem structure is as follows: S3.1: Set all parked vehicles and end users to unmatched status; S3.2: For the unmatched terminal user i, find the best preference in its preference list and obtain the corresponding parking vehicle v with the best preference * ,Right now And to the optimal preferred parking vehicle v * Send matching request; F i (v) represents the preferences of end user i; S3.3: The parked vehicle v collects all the end users who send matching requests and pairs them to form a requester set Y v ; If the number of parked vehicles v exceeds its capacity, then v Find the end user i with the worst preference * ,Right now and cancel the parking vehicle v and the end user i with the worst preference * Pairing (i * ,v);F v (i) represents the preference for parking vehicle v; If the matching of the parked vehicle v just reaches its own capacity at this time, for the new requester r, the parked vehicle v in Y v Find the end user i with the worst preference * , if F v (i * ) <F v (r), then cancel the pairing (i * ,v), and establish a pairing (r,v); S3.4: For the terminal user whose pairing is cancelled, the best preferred parked vehicle is deleted from its preference list, and the state is reset to unmatched, and the process returns to step S3.2 until a stable match is achieved; When a valid match does not contain any blocking pairs, it is considered a stable match; A valid match must meet the following conditions: A1: Each digital twin can only match one parked vehicle, that is, |Φ(i)|≤1; A2: Parked vehicles are associated with multiple digital twins, but the number of matches is limited, i.e. A3: For matching relations, Φ(i) = {v∈V|i→v∈Φ} and Φ(v) = {i∈I|v→i∈Φ}; When a pair (i, v) blocks the matching Φ, it is considered a blocking pair. The conditions for satisfying the blocking pair are as follows: B1: End user i and parked vehicle v have mutual preferences, i∈F v And v∈F i ; B2: End user i has no pairing, or prefers parking vehicle v over matching Φ(i) = v'; B3: The parked vehicle v does not reach the pairing capacity, or relative to the pairing i'∈Φ -1 (v), parked vehicle v is more biased towards end user i; Among them, v' and v represent different parked vehicles, i' and i represent different end users; S4: Perform digital twin aggregation based on stable matching results, where the preference of the digital twin is the digital twin DT associated with the parked vehicle v i The corresponding transmission rate: F i (v)=R v,i The preference of the parked vehicle v is the digital twin DT of interest at the parking time i Information age:

2. According to claim 1, a vehicle auxiliary edge digital twin aggregation method based on information age matching is characterized in that: The method also includes obtaining the best association between the overloaded base station and the parked vehicles through an exhaustive search algorithm.

3. The vehicle auxiliary edge digital twin aggregation method based on information age matching according to claim 2 is characterized in that: The specific process of the exhaustive search algorithm is: D1: Construct a feasible solution set γ for the optimization objective; D2: For one possible outcome, calculate its result; D3: Determine whether the result meets the requirements. If not, return to step D2 to search for the next possible situation. If it meets the requirements, obtain the optimal decision variable γ m,v .