A Downlink Efficient Content Delivery Method for B5G / 6G Fully Decoupled Cellular Vehicle-to-Everything Networks

Abstract

An efficient content distribution method for B5G (Beyond 5G) / 6G fully decoupled cellular vehicle-to-everything based on many-to-many matching game. 1) The control base station collects and stores all information such as the positions and resources of base stations, as well as information that does not change with time such as road topologies. The base station obtains vehicle-related information in real time, including vehicle positions, speeds, data requests, etc., and then transmits the vehicle information to the control base station, which integrates and processes the collected information. 2) The control base station predicts the future positions of vehicles based on the obtained vehicle information, then conducts system modeling, executes the vehicle content distribution algorithm VCD, obtains the base station service content distribution plan, transmits control information to the base stations, and assigns multiple base stations to each vehicle requesting content. 3) Each base station obtains the data content requested by the vehicle from the edge cloud according to the control information sent by the control base station, and distributes the content to the corresponding requesting vehicles to complete the content distribution.

Description

Technical Field

[0001] The present invention belongs to the technical field of vehicle networking, and relates to an efficient content distribution method for B5G / 6G fully decoupled cellular vehicle networking based on many-to-many matching games. Background Art

[0002] With the commercial deployment of 5G worldwide, the era of the Internet of Everything has arrived, and it has entered the research and development year of the next-generation mobile communication system (6G). Among them, vehicle networking is considered an important part. Vehicle networking is an extremely large network composed of vehicle information, road information, infrastructure information, etc., which can supervise the vehicles in the vehicle networking and can also provide various services for vehicle networking users. As a key part of the Internet of Everything, vehicle networking will play an important role, including reducing traffic congestion, efficient traffic management, providing infotainment, etc. Many researchers have developed many applications for vehicle networking regarding road safety and entertainment, which can provide better and more convenient services for vehicle networking users, timely issue road safety warning information to vehicle users, avoid traffic congestion or remind drivers to drive carefully; this can also help traffic management departments better grasp the real-time traffic situation and achieve intelligent transportation; in addition to ensuring road driving safety, these applications can also provide leisure and entertainment services for vehicle users, making the journey more relaxed and pleasant. These applications can effectively ensure road safety, greatly improve traffic operation efficiency, and can also improve the comfort and convenience of vehicle networking users. All of these rely on the efficient distribution of data content. As the number of vehicles connected to the vehicle networking increases, the data request volume grows exponentially, and the efficient content distribution in the vehicle network becomes increasingly important.

[0003] After retrieving the existing literature, it was found that Haibo Zhou, Bo Liu, etc. published an article titled "ChainCluster: Engineering a Cooperative Content Distribution Framework for Highway Vehicular Communications" in the "IEEE Transactions on Intelligent Transportation Systems" in 2014. Due to the high mobility of vehicles and the intermittency of channel connections, the data download volume of each vehicle is quite limited and the communication dark zone is relatively long. Therefore, it was proposed that vehicles form chain clusters to cooperate in downloading content, which is mainly divided into three stages: 1) Chain cluster formation stage: The target vehicle requests a large-size file. Before entering the communication coverage area, the target vehicle invites the vehicles traveling in the same direction behind it to form a linear cooperative chain cluster; 2) Content download stage: When the vehicles in the cooperative chain cluster drive into the communication coverage area, each vehicle will sequentially download non-overlapping parts of the file; 3) Content forwarding stage: After the vehicles in the cooperative chain cluster drive out of the communication coverage area, the target vehicle collects the file parts from the cooperative vehicles in the cooperative chain cluster to restore the file.

[0004] It was also found through retrieval that Guiyang Luo, Quan Yuan, Haibo Zhou, etc. published an article titled "Cooperative vehicular content distribution in edge computing assisted 5G-VANET" in "China Communications" in 2018. To adapt to the rapid topological changes and unbalanced traffic volume, the article introduced a hierarchical structure based on edge computing. The upper-layer urban controller schedules data caching from a network-wide view and coordinates the resources of macro base stations. The lower-layer macro base stations manage small base stations, roadside units (RSUs), vehicles, and WiFi (Wireless Fidelity) nodes within their coverage area. On this basis, a multi-location, multi-factor prefetching scheme was proposed, which prefetches data content to RSUs or vehicles with a high degree of social centrality and then distributes it to the vehicles requesting data. Then a neighbor graph containing vehicle information was constructed, the neighbor graph was transformed into a matching graph, and the matching graph was divided into multiple subgraphs. The content distribution problem was transformed into a maximum weighted independent set problem for solution.

[0005] In summary, the problems existing in the prior art are still using vehicle-to-vehicle communication and vehicle-to-roadside unit communication. However, the communication ranges of these two methods are small and unstable, making it difficult to complete the content distribution task.

[0006] After retrieval, it is also found that the fully decoupled radio access network architecture published by Quan Yu, Jiacheng Chen, etc. in the Journal of Communications and Information Networks in 2019 has a very flexible access method, which can greatly promote resource collaboration, improve spectrum utilization, reduce the overall network energy consumption, and optimize the quality of user experience. This network structure transforms the traditional wireless access network structure of small cell base stations and macro base stations collaborating, decouples the functions of small cell base stations and macro base stations into an uplink base station responsible for uplink data transmission, a downlink base station responsible for downlink data transmission, and a control base station responsible for network coordination using control signals. Vehicle network users no longer perform uplink and downlink transmissions with the same base station, but flexibly and dynamically select uplink and downlink base stations respectively for uplink and downlink data communication, and can select multiple base stations to perform data communication simultaneously.

[0007] The existing vehicle network content distribution methods mainly focus on vehicle-to-vehicle communication and vehicle-to-roadside unit communication. However, the communication ranges of these two methods are small and cannot complete the content distribution task.

[0008] Aiming at the problems existing in the prior art, in the B5G / 6G fully decoupled cellular vehicle network, the present invention allocates multiple base stations to serve vehicles, mainly for the content distribution of downlink data, aiming to solve the problem of efficient distribution of massive data content in the vehicle network in urban road scenarios and provide high-quality content distribution services for vehicle network users. Summary of the Invention

[0009] Object of the Invention: Aiming at the problems existing in the prior art, the present invention provides a dynamic and efficient content distribution method for vehicle networks under B5G / 6G fully decoupled cellular vehicle networks.

[0010] Technical Solution: The efficient content distribution method for the B5G / 6G fully decoupled cellular vehicle network includes the following steps:

[0011] Step 1: Considering aspects such as vehicle fairness and base station service capabilities, establish a system optimization model:

[0012]

[0013] Among them, U total represents the total utility of the system. The set of vehicle users is set as V = {V 1 , V 2 ,..., Vi ,..., V M} where i represents the identification number of the vehicle and M is the number of vehicles. The set of base stations is set as B = {B 1 , B 2 ,..., B j ,..., B K}, where j represents the identification number of the base station and K is the number of base stations. The set of the number of sub-channels corresponding to each base station is S = {S 1 , S 2 ,..., S j ,..., S K}, S j represents that the base station B j has S j sub-channels, and b represents the maximum number of base stations that a vehicle can access. A matrix A(t) of size M×K is constructed to represent the relationship between vehicles and base stations at time t. The element a i，j in the matrix A(t) belongs to {0, 1}. a i，j = 1 means that the base station B j serves the vehicle V i , and a i，j = 0 means that the base station B j does not serve the vehicle V i .

[0014] Step 2: Using the many-to-many matching game, transform the optimization problem proposed in Step 1 into a many-to-many matching problem:

[0015] First, define the concept of many-to-many matching as follows: Given two disjoint sets, V = {V 1 , V 2 ,..., V i ,..., V M} is the set of vehicle users, and B = {B 1 , B 2 ,..., B j ,..., B K} is the set of base stations. The many-to-many matching Ψ is a mapping from the set V∪B∪{0} to the set of all subsets of V∪B∪{0}, such that for each V i ∈ V and B j ∈ B, there are:

[0016] (1)

[0017] (2)

[0018] (3)|Ψ(V i )| ≤ b;

[0019] (4)|Ψ(Bj )| ≤ S j ;

[0020] (5)

[0021] where Ψ(V i ) represents the set of base stations paired with vehicle V i , Ψ(B j ) represents the set of vehicles paired with base station B j , represents that the set of base stations paired with vehicle V i is a subset of set B, represents that the set of vehicles paired with base station B j is a subset of set V, |·| represents the number of elements in the set, B j ∈ Ψ(V i ) represents that base station B j is an element in the set of base stations paired with vehicle V i , V i ∈ Ψ(B j ) represents that vehicle V i is an element in the set of vehicles paired with base station B j , represents an equivalence relation.

[0022] Condition (1) means that each vehicle user is matched with a subset of base stations, and condition (2) means that each base station is matched with a subset of vehicle users. Considering the service capacity of the base stations and the fairness of vehicle users, conditions (3) and (4) are set. The above matching game is a many-to-many matching game, and there will be a peer effect.

[0023] Affected by the peer effect, the result of the matching game depends to a large extent on the dynamic interaction between participants. To better describe the competitive behavior and decision-making process of each participant, it is assumed that each participant has preferences for the participants in the other group, and a preference relation > is introduced for vehicle users and base stations.

[0024] Preferences of vehicle users: Specifically, for any vehicle user V i ∈ V, its preference for base stations >V i can be described as follows:

[0025] For any two subsets of base stations H B ≠ H′ B , and any two matchings Ψ and Ψ′, and H B = Ψ(V i ), H′ B = Ψ′(V i ), then

[0026]

[0027] Among them, represents the utility that vehicle V can obtain when matching Ψ i obtains. represents the utility that vehicle V can obtain when matching Ψ′ i obtains. represents an equivalence relation.

[0028] This means that if and only if vehicle user V i obtains a greater utility from base station subset H B than from base station subset H′ B , compared with base station subset H′ B , vehicle V i will prefer base station subset H B .

[0029] Preference of base stations: Similar to the preference of vehicle users, for any base station B j ∈B, its preference for vehicle users can be described as follows:

[0030] For any two vehicle user subsets H V , H V ≠H′ V , and any two matchings Ψ and Ψ′, and H V =Ψ(B j ), H′ B =Ψ′(B j ) then

[0031]

[0032] Among them, represents the utility that base station B can obtain when matching Ψ j obtains. represents the utility that base station B can obtain when matching Ψ′ j obtains. represents an equivalence relation.

[0033] This means that if and only if the utility that base station B j obtains from serving vehicle user subset H is greater than the utility from serving vehicle user subset H′, compared with serving vehicle user subset H′, base station B j will prefer to serve vehicle user subset H.

[0034] The many-to-many matching model with externalities is more complex than the traditional bilateral matching model. Under the traditional definition of stable matching, the existence of stable matching cannot be guaranteed even in many-to-one matching.

[0035] The exchange behavior of vehicle users is considered to be that the control base station arranges every two vehicles to exchange their matches while keeping the allocations of other vehicles the same. To better describe how the interdependence of participants' preference relations, i.e., peer effects, affects the matching, the concepts of exchange matching and exchange blocking pairs are introduced below.

[0036] Given a matching Ψ, in this matching Ψ, B p ∈Ψ(V i ), B q ∈Ψ(V j ), V i and V j exchange their respective base stations B p , B q , then the exchange matching is defined as where

[0037] where, B p ∈Ψ(V i ) means that the base station B p is an element in the set of base stations paired with the vehicle V i , that is, the base station B p serves the vehicle V i , means that the base station B p is not an element in the set of base stations paired with the vehicle V j , that is, the base station B p does not serve the vehicle V j , (V i , B p ) means that the vehicle V i and the base station B p are paired, and \ means removing the elements in the set.

[0038] The above definition of exchange matching means that in the exchange operation, two participants in the same set exchange their matches in the opposite set while keeping the allocations of all other participants the same. In the matching model of the present invention, since the control base station has the information of all vehicles and base stations, and the matching algorithm is executed by the control base station, it is meaningful to assume that participants can exchange information with each other. Participants in the exchange matching are allowed to be unmatched, thus allowing some unplanned participants to be active.

[0039] However, not all exchanges are approved. For their own interests, the parties involved in the exchange may not obtain the approval of the other party. By introducing the concept of exchange blocking pairs, the conditions for the approval of exchange operations are proposed.

[0040] Given a matching Ψ and a pair (V i , V j ) in this matching, if there exist B p ∈Ψ(V i ), B q ∈Ψ(V j ) that satisfy the following conditions:

[0041] (1)

[0042] (2)

[0043] Then the exchange matching is approved, and (V i , V j ) is called an exchange blocking pair in the matching Ψ.

[0044] The above exchange approval conditions mean that if the exchange matching is approved, then the utility of any participating participant will not decrease, and the utility of at least one participant will increase. As long as it is a participant, whether it is a vehicle user or a base station, it can initiate an exchange.

[0045] According to the above definition, the behavior of vehicle users in the matching can be described. Every two vehicle users can be arranged by the control base station to form a potential exchange blocking pair. The control base station judges by calculation whether this exchange operation can benefit the vehicle users from each other without harming the interests of the corresponding base station. The participants continuously execute the approved exchange operations to reach a stable state, which is also called a bilateral exchange stable matching, defined as: the matching Ψ is bilaterally exchange stable if it is not blocked by any exchange blocking pair.

[0046] Step 3: Propose a content distribution algorithm VCD based on many-to-many matching games to solve the optimization problem in Step 1:

[0047] The VCD algorithm can be divided into two stages: the initialization stage and the exchange matching stage. It is described as follows:

[0048] (1) Initialization phase: A priority-based allocation scheme is adopted, where weights are used to represent priorities, that is, the greater the weight of a vehicle user, the higher its priority when selecting an available subset of base stations. From the perspective of vehicle fairness, considering the channel capacity of the vehicle in the past time, if the average channel capacity of the vehicle in the past time is small, then it will have a higher priority to select a match at the current moment. At the initial moment, since the vehicle has no past channel capacity, the vehicle selects in a random order.

[0049] Let vehicle V i The weight ω at time t i Is inversely proportional to the average channel capacity in the past (t - 1) moments, that is

[0050]

[0051] (2) Exchange matching phase: The base station is controlled to continuously search for two vehicle users to form an exchange blocking pair, and then if approved, they perform exchange matching and update the current matching. This operation is continuously iterated until no previously undetected exchange blocking pair can be found.

[0052] Beneficial effects. Based on the content distribution of the vehicle-to-everything (V2X) network, the present invention discloses an efficient content distribution method for B5G (Beyond 5G) / 6G fully decoupled cellular V2X network based on matching game, which provides high-quality content distribution services for V2X network users. The present invention uses vehicles to communicate with cellular base stations for content distribution under the B5G / 6G fully decoupled cellular V2X network. The present invention comprehensively considers issues such as vehicle fairness and the maximum service capacity of base stations, and allocates multiple base stations to serve each vehicle, which can meet the requests of vehicle users for large-capacity data. Description of the drawings

[0053] Figure 1 Is the urban road scenario diagram of the efficient content distribution of B5G / 6G fully decoupled cellular V2X network adopted in the embodiment of the present invention.

[0054] Figure 2 Is the schematic diagram of the base station and vehicle in the embodiment of the present invention.

[0055] Figure 3 Is the implementation block diagram of the VCD algorithm for the efficient content distribution of B5G / 6G fully decoupled cellular V2X network adopted in the embodiment of the present invention.

[0056] Figure 4 Is the graph of the total system utility changing with the algorithm iteration at a random moment in the embodiment of the present invention.

[0057] Figure 5 Is the graph of the average channel capacity of all vehicles at all moments under three loads in the embodiment of the present invention. Detailed implementation manners

[0058] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the following provides a detailed description of the embodiments of the present invention with reference to the accompanying drawings: These embodiments are implemented on the premise of the technical solutions of the present invention, and detailed implementation manners and specific operation processes are given. It should be understood that the specific examples described herein are only used to explain the present invention, but the protection scope of the present invention is not limited to the following embodiments.

[0059] The content distribution method of the present invention includes: controlling a base station to collect and store all information such as base station locations, resources, etc., as well as information that does not change with time such as road topologies. The base station obtains vehicle-related information in real time, including vehicle locations, speeds, data requests, etc., and then transmits the vehicle information to the control base station. The control base station integrates and processes the collected information; the control base station predicts the future locations of vehicles based on the obtained vehicle information, then performs system modeling, executes the Vehicle Content Distribution Algorithms (VCD), obtains a base station service content distribution plan, transmits control information to the base stations, and allocates multiple base stations to each vehicle requesting content; each base station obtains the data content requested by the vehicle from the edge cloud according to the control information sent by the control base station, and distributes the content to the corresponding requesting vehicles to complete the content distribution.

[0060] Embodiment

[0061] This embodiment adopts Figure 1 the content distribution scenario diagram of urban roads, and proposes a vehicle-to-everything (V2X) content distribution method based on many-to-many matching games. In this scenario, considering two intersecting two-way six-lane urban roads, simulating the situation of vehicle-requested content, it is divided into three cases: low load is 80 vehicles per kilometer requesting content, medium load is 110 vehicles per kilometer requesting content, and high load is 140 vehicles per kilometer requesting content. Figure 2 is the schematic diagram of base stations and vehicles in the embodiment of the present invention.

[0062] Content distribution mainly consists of a control base station, an edge cloud, downlink small base stations, and vehicles. Content distribution is divided into three stages:

[0063] (1) The control base station collects and stores all information such as base station locations, resources, etc., as well as information that does not change with time such as road topologies. The base station obtains vehicle-related information in real time, including vehicle locations, speeds, data requests, etc., and then transmits the vehicle information to the control base station. The control base station integrates and processes the collected information.

[0064] (2) The control base station predicts the future position of the vehicle based on the obtained vehicle information, then performs system modeling, executes the VCD algorithm, obtains the base station service content distribution plan, transmits control information to the base station, and allocates multiple base stations to each vehicle requesting content.

[0065] (3) Each base station obtains the data content requested by the vehicle from the edge cloud according to the control information sent by the control base station, and distributes the content to the corresponding requesting vehicle to complete the content distribution.

[0066] The set of vehicle users is set as V = {V 1 , V 2 ,..., V i ,..., V M}, where i represents the identification number of the vehicle, and M is the number of vehicles. The set of base stations is set as B = {B 1 , B 2 ,..., B j ,..., B K}, where j represents the identification number of the base station, and K is the number of base stations. The set of the number of sub-channels corresponding to each base station is S = {S 1 , S 2 ,..., S j ,..., S K}, and S j represents that the base station B j has S j sub-channels.

[0067] Construct a matrix A(t) of size M×K to represent the relationship between vehicles and base stations at time t. The element a i , j ∈ {0, 1}, and a i，j = 1 means that the base station B j serves the vehicle V i , and a i，j = 0 means that the base station B j does not serve the vehicle V i .

[0068] The base station B j has S j sub-channels and can serve at most S j vehicles simultaneously, that is

[0069]

[0070] To make full use of the channel resources of the base station, when the number of vehicles served by the base station B j is less than the number of sub-channels, the base station evenly distributes the sub-channels to the served vehicles, then the base station B j distributes to the vehicle V iThe number m of allocated sub-channels i，j is

[0071]

[0072] wherein represents rounding down.

[0073] For user fairness, the maximum number of vehicles connected to the base station is b, that is

[0074]

[0075] Vehicle V i on base station B j the channel capacity C i，j is

[0076] C i，j = W i，j log 2 (1 + SNR i，j ) (4)

[0077] wherein, W i，j is the channel bandwidth obtained by vehicle V i on base station B j , and SNR i，j is the signal-to-noise ratio.

[0078] Let the sub-channel unit bandwidth under base station B j be W j , and the channel bandwidth W i obtained by vehicle V j on base station B i，j is

[0079] W i，j = m i，j W j (5)

[0080] The signal-to-noise ratio SNR i，j is

[0081]

[0082] wherein, P i，j is the power allocated by base station B j to vehicle V i , g i，j is the channel gain between vehicle V i and base station B j , and n i，j is additive white Gaussian noise (AGWN), i.e., n i，j satisfies a Gaussian distribution with a mean of 0 and a variance of Gaussian distribution.

[0083] Assume that all base stations evenly distribute their power to each sub-channel. For base station B j with a total power of P j the power P j allocated to vehicle V i is: i，j

[0084]

[0085] At time t, the sum of the channel capacities of vehicle V i connected to multiple base stations is

[0086]

[0087] When performing content distribution, considering the high-speed mobility of vehicles, when designing the utility function of vehicles in the present invention, the positions of vehicles in the future for a period of time (for example, in the future N time intervals, and the prediction time interval is denoted as τ) are predicted, the channel capacities between the future vehicles and the base stations are calculated, assuming that the content distribution scheme of the base station services remains unchanged in the future N time intervals, and the average transmission rate of the vehicle at the current time and in the future N time intervals is used as the utility of the vehicle at time t That is:

[0088]

[0089] At time t, the total sum of the achievable channel capacities of base station B j is

[0090]

[0091] Similar to the vehicle utility, assuming that the base station allocation remains unchanged in the future N time intervals, the average channel capacity at the current time and in the future N time intervals is used as the utility function of the base station at time t That is:

[0092]

[0093] The sum of the utilities of all vehicle users is used as the utility function U of the system total That is:

[0094]

[0095] According to equations (1), (3), and (12), the efficient content distribution method for the vehicle-to-everything network in the present invention can be modeled as:

[0096]

[0097] ​​That is, under the constraint conditions of satisfying the maximum service quantity of the base station and vehicle fairness, by setting the variable a i，j to optimize U total .

[0098] This paper will use matching games to solve the content distribution problem, turning the base station service content distribution problem into a many-to-many two-sided matching game. The time interval of content distribution is denoted as T 0 , and a matching game is carried out every interval T 0 .

[0099] First, define the concept of many-to-many matching as follows: Given two disjoint sets, V = {V 1 , V 2 ,..., V i ,..., V M} is the vehicle user set, and B = {B 1 , B 2 ,..., B j ,..., B K} is the base station set. The many-to-many matching Ψ is a mapping from the set V ∪ B ∪ {0} to the set of all subsets of V ∪ B ∪ {0}, so that for each V i ∈ V and B j ∈ B, there are:

[0100] (1)

[0101] (2)

[0102] (3)|Ψ(V i )| ≤ b;

[0103] (4)|Ψ(B j )| ≤ S j ;

[0104] (5)

[0105] Among them, Ψ(V i ) represents the set of base stations paired with the vehicle V i , Ψ(B j ) represents the set of vehicles paired with the base station B j , represents that the set of base stations paired with the vehicle V i is a subset of the set B, represents that the set of vehicles paired with the base station B j is a subset of the set V, |·| represents the number of elements in the set, and B j ∈ Ψ(V i ) represents that the base station B j is paired with the vehicle Vi An element in the paired base station set, V i ∈Ψ(B j ) indicates that vehicle V i is an element in the set of base stations paired with base station B j . Indicates an equivalence relation.

[0106] Condition (1) means that each vehicle user is matched with a subset of base stations, and condition (2) means that each base station is matched with a subset of vehicle users. Considering the service capacity of base stations and the fairness of vehicle users, conditions (3) and (4) are set. The above matching game is a many-to-many matching game, and there will be a peer effect.

[0107] Affected by the peer effect, the result of the matching game depends to a large extent on the dynamic interaction between participants. To better describe the competitive behavior and decision-making process of each participant, it is assumed that each participant has preferences for the participants in the other group, and a preference relation > is introduced for vehicle users and base stations.

[0108] Preferences of vehicle users: Specifically, for any vehicle user V i ∈V, its preference for base stations can be described as follows:[[]]

[0109] For any two subsets of base stations H B , H B ≠H′ B , and any two matchings Ψ and Ψ′, and H B =Ψ(V i ), H′ B =Ψ′(V i ), then

[0110]

[0111] where represents the utility that vehicle V i can obtain under the matching Ψ, represents the utility that vehicle V i can obtain under the matching Ψ′.

[0112] This means that if and only if the utility obtained by vehicle user V i from the subset of base stations H B is greater than the utility obtained from the subset of base stations H′ B , compared with the subset of base stations H′ B , vehicle V i will prefer the subset of base stations H B .

[0113] Preferences of the base station: Similar to the preferences of vehicle users, for any base station B j ∈B, its preference for vehicle users can be described as follows: For any two subsets of vehicle users H V , H V ≠H′ V , and any two matchings Ψ and Ψ′, and H V =Ψ(B j ), H ′ B = Ψ′(B j ) then

[0114]

[0115] where represents the utility that base station B j can obtain under the matching Ψ, represents the utility that base station B j can obtain under the matching Ψ′.

[0116] This means that if and only if the utility obtained by base station B j from serving the subset of vehicle users H is greater than the utility obtained from serving the subset of vehicle users H′, compared with serving the subset of vehicle users H′, base station B j will prefer to serve the subset of vehicle users H more.

[0117] The many-to-many matching model with externalities is more complex than the traditional two-sided matching model. Under the traditional definition of stable matching, the existence of a stable matching cannot be guaranteed even in many-to-one matching.

[0118] The exchange behavior of vehicle users is considered to be arranged by the control base station to exchange their matches for every two vehicles while keeping the allocations of other vehicles the same. To better describe how the mutual dependence of the preference relations of participants, that is, the peer effect, affects the matching, the concepts of exchange matching and exchange blocking pairs are introduced below.

[0119] Given a matching Ψ, in this matching Ψ, B p ∈Ψ(V i ), B q ∈Ψ(V j ), V i and V j exchange their respective base stations B p , B q , then the exchange matching is defined as where

[0120] The above swap matching definition means that in a swap operation, two participants in the same set swap their matches in the opposite set while keeping the allocations of all other participants the same. In the matching model of the present invention, since the control base station has information about all vehicles and base stations, and the matching algorithm is executed by the control base station, it makes sense to assume that participants can exchange information with each other. Participants in swap matching are allowed to be unmatched, thus allowing some unplanned participants to be active.

[0121] However, not all swaps are approved. For their own interests, the parties involved in the swap may not get approval from each other. By introducing the concept of swap blocking pairs, the conditions for a swap operation to be approved are proposed.

[0122] Given a matching Ψ and a pair (V i , V j ) in this matching, if there exist B p ∈ Ψ(V i ), B q ∈ Ψ(V j ) that satisfy the following conditions:

[0123] (1)

[0124] (2)

[0125] Then the swap matching is approved, and (V i , V j ) is called a swap blocking pair in the matching Ψ.

[0126] The above swap approval condition means that if a swap matching is approved, then the utility of any participant involved will not decrease, and the utility of at least one participant will increase. As long as they are participants, whether vehicle users or base stations, they can initiate a swap.

[0127] According to the above definition, the behavior of vehicle users in the matching can be described. Every two vehicle users can be arranged by the control base station to form a potential swap blocking pair. The control base station judges by calculation whether this swap operation can make the vehicle users benefit from each other without harming the interests of the corresponding base station. Participants continuously execute approved swap operations to reach a stable state, which is also called a bilateral swap stable matching, defined as: the matching Ψ is bilaterally swap stable if it is not blocked by any swap blocking pair.

[0128] Step 3: Propose a content distribution algorithm VCD based on many-to-many matching game to solve the optimization problem in Step 1:

[0129] The VCD algorithm can be divided into two stages: the initialization stage and the exchange matching stage. The description is as follows:

[0130] (1) Initialization stage: A priority-based allocation scheme is adopted, and weights are used to represent priorities. That is, the greater the weight of a vehicle user, the higher its priority when selecting an available base station subset. From the perspective of vehicle fairness, considering the channel capacity of the vehicle in the past time, if the average channel capacity of the vehicle in the past time is small, then it will have a higher priority to select a match at the current moment. At the initial moment, since the vehicle has no past channel capacity, the vehicle selects in a random order.

[0131] Let vehicle V i have a weight ω i at time t, which is inversely proportional to the average channel capacity in the past (t - 1) moments, that is

[0132]

[0133] (2) Exchange matching stage: The base station is controlled to continuously search for two vehicle users to form an exchange blocking pair. Then, if approved, they perform an exchange match and update the current match. This operation is continuously iterated until no previously undetected exchange blocking pairs can be found.

[0134] Table 1 presents the VCD algorithm for efficient content distribution in B5G / 6G fully decoupled cellular vehicle-to-everything (C-V2X).

[0135] Table 1 VCD Algorithm for Efficient Content Distribution in B5G / 6G Fully Decoupled Cellular Vehicle-to-Everything

[0136]

[0137] The flow chart of this algorithm is as Figure 3 .

[0138] According to the number of requesting vehicles, it is divided into three cases: low load is 80 vehicles requesting content per kilometer, medium load is 110 vehicles requesting content per kilometer, and high load is 140 vehicles requesting content per kilometer. The experiment has a total of 100 moments, with a time interval of 1 s. Figure 4 It shows the change of the total system utility with the algorithm iteration at a randomly selected moment in the experiment. As the algorithm iterates and continuously exchanges matching pairs, the total system utility continuously increases. When the exchange reaches a certain point, the system utility no longer changes and begins to converge. At this time, there are no executable exchange blocking pairs in the system, reaching a state of bilateral exchange stability. When the load is different, the convergence of the algorithm will also change. Especially in the case of high load, the main reason is that there are too many vehicle users and too many exchange blocking pairs, so the convergence will be slower. Figure 5 It shows the average channel capacity of all vehicles at all moments under the three loads in the experiment.

[0139] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An efficient content distribution method for B5G (Beyond 5G) / 6G fully decoupled cellular vehicle-to-everything (C-V2X) based on many-to-many matching game, Characterized in that: The content distribution method includes the following steps: Step 1: The control base station collects and stores information about all base station locations, resources, and road topologies that do not change over time. The base station obtains vehicle-related information in real time, including vehicle location, speed, and data requests, and then transmits the vehicle information to the control base station. The control base station integrates and processes the collected information. Step 2: The control base station predicts the future positions of vehicles based on the obtained vehicle information, then performs system modeling, executes the Vehicle Content Distribution Algorithms (VCD), obtains the base station service content distribution plan, transmits control information to the base stations, and allocates multiple base stations to each vehicle requesting content. Step 3: Each base station obtains the data content requested by the vehicle from the edge cloud according to the control information sent by the control base station, and distributes the content to the corresponding requesting vehicle to complete the content distribution. Considering the current channel capacity between the vehicle and the base station, and by simply predicting the position of the vehicle at a future time, the channel capacity situation of the vehicle at a future time is considered; specifically as follows: The set of vehicle users is set as V = {V 1 , V 2 ,..., V i ,..., V M}, where i represents the identification number of the vehicle and M is the number of vehicles; the set of base stations is set as B = {B 1 , B 2 ,..., B j ,..., B K}, where j represents the identification number of the base station and K is the number of base stations. The set of the number of sub-channels corresponding to each base station is S = {S 1 , S 2 ,..., S j ,..., S K}, and S j represents that the base station B j has S j sub-channels; Construct a matrix \(A(t)\) of size \(M\times K\) to represent the relationship between the vehicle and the base station at time \(t\). The element \(a\) in the matrix \(A(t)\) i，j \(\in\{0, 1\}\), where \(a\) i，j \( = 1\) indicates that the base station \(B\) j serves the vehicle \(V\) i , and \(a\) i,j \( = 0\) indicates that the base station \(B\) j does not serve the vehicle \(V\) i . At time t, vehicle V i The sum of the channel capacities connecting multiple base stations is Among which C i，j is the channel capacity between i vehicle V j and base station B; When performing content distribution, considering the mobility of the vehicle, when designing the utility function of the vehicle, the position of the vehicle in the next period of time is predicted. Here, the next period of time is the next N time intervals, and the prediction time interval is denoted as τ. The channel capacity between the vehicle and the base station in the future is calculated. Assuming that the content distribution scheme served by the base station remains unchanged within the next N time intervals, the average transmission rate of the vehicle at the current time and in the next N time intervals is used as the utility of the vehicle at time t. That is: At time t, base station B j Total achievable channel capacity is Similar to the vehicle utility, assuming that the base station allocation remains unchanged in the next N time intervals, the average channel capacity in the current and the next N time intervals is used as the utility function of the base station at time t That is: Take the sum of the utilities of all vehicle users as the system utility function U total , that is: Focusing on the vehicle fairness issue, the maximum number of base stations that a vehicle can access is set; specifically as follows: For user fairness, each vehicle can connect to at most b base stations, that is The content distribution problem is transformed into a many-to-many matching problem, specifically as follows: Considering the vehicle user set V and the base station set B, both of these two participant sets are selfish and rational, and attempt to maximize their own interests; the base station allocation is determined by the control base station, and the control base station can obtain the base station allocation plan by executing an algorithm. The vehicle users and base stations are regarded as selfish and rational participants. Since the control base station has information about all vehicle users and base stations, and it is the control base station that executes the algorithm, it can be regarded that each participant can exchange information with each other without additional communication costs, that is, each participant has complete information of all participants; if base station B j is assigned to vehicle V i , then V i and B j match each other and form a matching pair; the matching is defined as assigning the base stations in set B to the vehicle users in set V, which is expressed as follows: Given two disjoint sets, V = {V 1 , V 2 , ..., V i , ..., V M} is the set of vehicle users, B = {B 1 , B 2 , ..., B j , ..., B K} is the set of base stations, and the many-to-many matching Ψ is a mapping from the set V ∪ B ∪ {0} to the set of all subsets of V ∪ B ∪ {0}, such that for each V i ∈ V and B j ∈ B, there are: (3)|Ψ(V i )| ≤ b; (4)|Ψ(B j )|≤S j ; where, Ψ(V i ) represents the set of base stations paired with vehicle V i , Ψ(B j ) represents the set of vehicles paired with base station B j , represents that the set of base stations paired with vehicle V i is a subset of set B, represents that the set of vehicles paired with base station B j is a subset of set V, |·| represents the number of elements in the set, B j ∈Ψ(V i ) represents that base station B j is an element of the set of base stations paired with vehicle V i , V i ∈Ψ(B j ) represents that vehicle V i is an element of the set of vehicles paired with base station B j , represents an equivalence relation; Condition (1) means that each vehicle user is matched with a subset of base stations, and condition (2) means that each base station is matched with a subset of vehicle users. Considering the service capacity of the base station and vehicle user fairness, conditions (3) and (4) are set; the above matching game is a many-to-many matching game, and a peer effect will occur. Assume that each participant has preferences for the participants in the other group, and introduce preference relations for vehicle users and base stations ; Preferences of vehicle users: Specifically, for any vehicle user V i ∈ V, its preference for a base station can be described as follows: For any two subsets H of base stations B , H B ≠H′ B , and for any two matchings Ψ and Ψ′, and H B =Ψ(V i ), H′ B =Ψ′(V i ), then Among them, represents the utility that vehicle V i can obtain when matching Ψ, represents the utility that vehicle V i can obtain when matching Ψ′, represents an equivalence relation; This means that if and only if the vehicle user V i obtains a greater utility from the base station subset H B than from the base station subset H′ B will vehicle V B prefer the base station subset H i more than the base station subset H′ B ; Preference of the base station: Similar to the preference of the vehicle user, for any base station B j ∈ B, its preference for the vehicle user can be described as follows: For any two subsets H of vehicle users V , H V ≠ H′ V , and for any two matchings Ψ and Ψ′, and where H V = Ψ(B j ), H′ B = Ψ′(B j ), then Among them, represents the utility that base station B can obtain when matching Ψ j can obtain, represents the utility that base station B can obtain when matching Ψ′ j can obtain, represents an equivalence relation; This means that if and only if the utility obtained by base station B j for the subset H of vehicle users is greater than the utility obtained from serving the subset H' of vehicle users, then compared to serving the subset H' of vehicle users, base station B j will prefer to serve the subset H of vehicle users; First, introduce the concepts of exchange matching and exchange blocking pair. Given a matching Ψ, in this matching Ψ, B p ∈ Ψ(V i ), B q ∈ Ψ(V j ), V i and V j exchange their respective base stations B p , B q , then the exchange matching is defined as where Among them, B p ∈ Ψ(V i ) means that base station B p is an element in the set of base stations paired with vehicle V i , that is, base station B p serves vehicle V i . means that base station B p is not an element in the set of base stations paired with vehicle V j , that is, base station B p does not serve vehicle V j . (V i , B p ) means that vehicle V i and base station B p are paired, and \ means removing elements from the set; The above definition of exchange matching means that in the exchange operation, two participants in the same set exchange their matches in the opposite set while keeping the allocations of all other participants the same; in the matching model, since the control base station has information about all vehicles and base stations, and it is the control base station that executes the matching algorithm, it is assumed that participants can exchange information with each other; participants in the exchange matching are allowed to be unmatched, thus allowing participants not in the plan to be active. By introducing the concept of exchange blocking pair, the conditions for the approval of the exchange operation are proposed. Given a matching Ψ and a pair (V i , V j ) in this matching, if there exist B p ∈ Ψ(V i ), B q ∈ Ψ(V j ) that satisfy the following conditions: Then swap the match is approved, (V i , V j ) is called a swap blocking pair in the match Ψ; The above condition for the approval of exchange matching means that if the exchange matching is approved, then the utility of any participating participant will not decrease, and the utility of at least one participant will increase; as long as it is a participant, whether it is a vehicle user or a base station, it can initiate an exchange. Every two vehicle users are arranged by the control base station to form a potential exchange blocking pair. The control base station judges by calculation whether this exchange operation can benefit the vehicle users mutually without damaging the interests of the corresponding base station. The participants continuously execute the approved exchange operations to reach a stable state, which is called bilateral exchange stable matching and is defined as: the matching Ψ is bilaterally exchange stable if it is not blocked by any exchange blocking pair. A content distribution algorithm based on many-to-many matching game (Vehicle Content Distribution Algorithms, VCD) is proposed, which can be divided into two stages: the initialization stage and the exchange matching stage, described as follows: (1) Initialization stage: A priority-based allocation scheme is adopted, and weights are used to represent priorities, that is, the greater the weight of a vehicle user, the higher its priority when selecting an available base station subset. From the perspective of vehicle fairness, considering the channel capacity of the vehicle in the past time, if the average channel capacity of the vehicle in the past time is small, then it will have a higher priority to select a match at the current moment. At the initial moment, since the vehicle has no past channel capacity, the vehicle selects in a random order. Let vehicle V i have a weight ω at time t i which is inversely proportional to the average channel capacity over the past (t - 1) time instants, i.e., (2) Exchange matching stage: The control base station continuously searches for two vehicle users to form an exchange blocking pair, and then if it is approved, they perform exchange matching and update the current matching. The above exchange matching is continuously iterated until no previously un-found exchange blocking pair can be found.

Images