Layered relay cost flow method for resource scheduling of Internet of Vehicles with sudden local communication traffic

Abstract

The invention discloses a layered relay cost flow method for resource scheduling of the Internet of Vehicles in a local communication traffic burst, and the method comprises the steps: enabling a road side unit to collect the target communication rate, time delay and priority parameters of an access vehicle, building an intra-domain vehicle communication quality demand representation matrix, and calculating the intra-domain comprehensive communication demand target amount; judging a local communication traffic burst state through a congestion early warning threshold value, and initiating a resource rescheduling request to a central cloud; a fast dynamic bandwidth allocation algorithm is constructed, and the central cloud reallocates a bearing budget for the administered road side unit by using the algorithm and issues the bearing budget; the road side unit generates a vehicle feasible transmission path set by adopting a transmission path generation algorithm oriented to link stability, and constructs a source point-vehicle-relay vehicle-resource block-sink layered relay cost flow network diagram; and constructing an enhanced dynamic cost flow algorithm under the constraint of an upper-layer central cloud allocation bearing budget, and traversing the layered relay cost flow network diagram to obtain a final allocation result of each vehicle wireless resource block under the road side unit.

Description

Technical Field

[0001] This invention belongs to the field of vehicle-to-everything (V2X) communication and mainly relates to a hierarchical relay cost flow method for V2X resource scheduling under local traffic bursts. Background Technology

[0002] In recent years, vehicle-to-everything (IoV) communication technology has developed rapidly. Vehicles can communicate directly with base stations via vehicle-to-infrastructure (V2I) links or via vehicle-to-vehicle (V2V) collaboration for relay transmission to support high-bandwidth, low-latency services such as autonomous driving. With the increase in vehicle density and the dynamic fluctuations in service demands over time and space, the limited budget available for base station capacity management has become a bottleneck, necessitating an efficient dynamic resource allocation scheme to improve overall system performance.

[0003] However, existing solutions face challenges in dynamic and complex vehicle-to-everything (V2X) scenarios. Firstly, it's difficult to balance fair resource allocation with system efficiency. Some solutions, aiming to maximize system throughput, tend to concentrate resources on vehicles with excellent channel conditions, resulting in vehicles with weaker channel conditions not receiving basic service guarantees for extended periods, leading to severe resource unfairness. While introducing minimum service rate constraints can alleviate this problem to ensure fairness, static guarantee mechanisms often prematurely and excessively occupy scarce spectrum resources, thus limiting the space for further improvements in system throughput and efficiency. Secondly, when introducing V2V cooperative relays to expand coverage and increase capacity, there is a lack of effective adaptation and resource coordination for dynamic links. In high-speed vehicle scenarios, vehicle location and channel quality change rapidly. If the vehicle link transmission path and resource allocation mechanism fail to fully consider the dynamics and stability of the link, it can easily lead to frequent link interruptions, node switching oscillations, and inefficient or repeated use of spectrum resources. This not only fails to achieve the expected cooperative gains but may also lead to decreased system throughput and resource waste. Furthermore, in scenarios with multiple roadside units coexisting, a single roadside unit's resource scheduling based solely on local information fails to reflect the differences in vehicle demand and service gaps across different roadside units, easily leading to uneven allocation of carrying capacity across roadside units. Conversely, relying entirely on a central node to uniformly handle fine-grained scheduling across the entire network results in excessive computational and signaling overhead, making it difficult to meet the practical constraints of time-slot-level real-time performance. Therefore, existing methods struggle to achieve efficient, stable, and adaptive allocation of global spectrum resources while ensuring basic vehicle service needs. There is an urgent need for a new mechanism that integrates centralized optimization and distributed collaboration, accurately models dynamic links and resource constraints, to systematically address the challenge of synergistic optimization of basic service guarantees, throughput improvement, and collaborative stability. Summary of the Invention

[0004] To overcome the shortcomings of the prior art, this invention proposes a hierarchical relay cost flow method for vehicular network resource scheduling under local traffic bursts. This method aims to achieve cross-roadside unit bearer budget coordination and fine-grained collaborative scheduling within roadside units in vehicular network scenarios where vehicle service demands and link states change dynamically. This improves the basic communication guarantee capability of vehicles with weak coverage, enhances the stability and resource utilization efficiency of vehicle communication, and maximizes the overall system throughput and spectrum utilization under budget constraints. Thus, it provides a new approach to bandwidth resource allocation in dynamic vehicular networks.

[0005] To solve the above problems, the present invention adopts the following technical solution: Step 1: In the scheduling time slot Within the system, the roadside unit set is as follows: ; For any The collection of vehicles within the coverage area is ,exist Internal vehicle indexing ,and ; Building a vehicle matrix , of which Corresponding vehicle In the time slot The vehicle state vector; The vehicle state vector includes the vehicle position. Vehicle speed Vehicle target communication rate Priority Delay ; From vehicle matrix Extract columns relevant to requirement modeling to construct a requirement feature matrix. And construct a weight vector Thus, the target vector of vehicle integrated communication requirements is obtained. Then, the overall communication demand target within the domain is obtained by summarizing. And reported it to the central cloud; Step 2: Based on the demand information reported by each roadside unit, the central cloud dynamically and adaptively allocates the base station bearer budget for each roadside unit. To provide differentiated resources for each roadside unit to meet the communication needs of a sharp increase in local vehicles; Step 3: After receiving the carrying budget, the roadside unit constructs a set of feasible transmission paths within the roadside unit. ; Step 4: Calculate the vehicle quota based on the target amount of comprehensive vehicle communication needs, and then calculate the cost of each feasible transmission path. Step 5: Under capacity and cost constraints, construct a hierarchical relay cost flow network diagram based on V2V vehicle relays. A solution model is constructed, which is the hierarchical relay cost flow network diagram. This includes vehicle nodes, relay vehicle entry nodes, relay vehicle exit nodes, and resource block nodes; Step 6: Based on the network graph The optimal resource allocation scheme is solved using the enhanced dynamic cost flow algorithm, and the allocation variables are output. ; Step 7: Based on the allocation variables Generate and distribute scheduling decisions within roadside units; update historical vehicle rate statistics. In relation to the scheme history, fairness gaps are used for subsequent time slots. With switching penalty calculate.

[0006] In summary, the hierarchical relay cost flow method for vehicular network resource scheduling under localized traffic bursts in this invention collects localized communication demands within the scope of each roadside unit within the scheduling time slot and reports them to the central cloud. The central cloud dynamically and adaptively allocates the bearer budget based on the proportion of communication demands of each roadside unit, achieving resource tilting towards roadside units on the demand burst side. Roadside units, based on pre-configured resource blocks and the real-time location and speed of vehicles, filter and construct a set of candidate transmission schemes, including V2V relay transmission. Roadside units calculate the resource block quota and transmission path edge cost for each vehicle based on its demand information within the scope, constructing a vehicular network resource cost flow network graph based on V2V vehicle relay. Roadside units solve for the optimal resource allocation scheme using a dynamic augmented cost flow algorithm. This invention, by combining a cloud-edge collaborative hierarchical resource orchestration architecture with cost flow network modeling and dynamic augmented cost flow algorithm solving, achieves efficient, fair, and stable scheduling of spectrum resources, improving the overall throughput and resource utilization of the vehicular network system.

[0007] Compared with existing technologies, the beneficial effects of the present invention are as follows: 1. This invention proposes a demand modeling and hierarchical budget allocation mechanism for dynamic services in the Internet of Vehicles (IoV). Roadside units calculate and report the target quantity of comprehensive vehicle communication demand based on vehicle target rate requirements, priority, and time delay parameters within the scheduling time slot. The central cloud scores the demand of each roadside unit based on historical vehicle support gaps and allocates the carrying budget for each roadside unit under the constraint of the base station's overall carrying budget. This allows the carrying budget to adaptively adjust with changes in demand, thereby improving the fairness and utilization efficiency of cross-roadside unit resource allocation.

[0008] 2. This invention proposes a feasible transmission path construction method that considers link quality thresholds and link duration. The roadside unit calculates the achievable rates of vehicle-to-roadside unit and vehicle-to-vehicle links on a single resource block, and determines whether a vehicle uses direct connection or relay transmission based on the single resource block transmission rate threshold. Simultaneously, the V2V link duration is estimated based on the relative vehicle speed, and a minimum duration constraint is applied, thereby filtering out unstable relay links, reducing frequent link interruptions and switching oscillations, and improving the stability and effectiveness of V2V cooperative communication.

[0009] 3. This invention proposes a hierarchical relay cost flow network graph construction method based on V2V vehicle relays, employing a unified modeling approach of "quota capacity + cost function". Roadside units calculate the vehicle reference resource block quota capacity based on the vehicle's overall communication demand target and carrying budget, ensuring interpretable capacity constraints in resource block allocation among vehicles. Simultaneously, an edge cost function is constructed that integrates the average single resource block transmission path rate, the vehicle's overall communication demand target, historical guarantee gaps, and path switching penalties. This ensures that resource allocation can both address the guarantee gaps of vehicles with weak links and suppress instability and additional overhead caused by frequent switching.

[0010] 4. This invention proposes an enhanced dynamic cost flow algorithm, introducing hot start and periodic reset to improve time slot-level solution efficiency and long-term stability. Roadside units construct a hierarchical relay cost flow network graph under capacity and cost constraints, solving for the resource allocation scheme that maximizes resource block occupancy and minimizes total cost under maximum occupancy, obtaining the allocation variables of vehicle-relay vehicle-resource block. Simultaneously, the scheduling results of the previous time slot are used to map and form a hot start initial solution to accelerate the solution, and periodic cold start resets eliminate historical accumulated deviations, thereby maintaining the stability and robustness of scheduling decisions while ensuring real-time performance. Attached Figure Description

[0011] Figure 1 This is a basic flowchart of the hierarchical relay cost flow resource scheduling method for vehicle-to-everything (V2X) networks oriented to local traffic bursts. Figure 2 This is a flowchart of the enhanced dynamic cost flow algorithm based on hot start and periodic reset of the present invention; Figure 3 A schematic diagram comparing the average transmission rate of vehicles within a single RSU as the number of vehicles varies between the average allocation scheme and the scheme of the present invention. Figure 4 A schematic diagram comparing the total throughput of the average allocation scheme and the scheme of the present invention; Figure 5 This diagram illustrates a comparison of the Jain fairness index, which shows the rate requirement satisfaction between the average allocation scheme and the scheme of this invention. Detailed Implementation

[0012] In this embodiment, the hierarchical relay cost flow method for vehicular network resource scheduling under local traffic bursts includes: as follows Figure 1 As shown, the procedure is as follows: Step 1: In the scheduling time slot Within the system, the roadside unit set is as follows: For any The collection of vehicles within its coverage area is ,exist Internal vehicle indexing ,and . Building a vehicle matrix , of which Corresponding vehicle In the time slot The vehicle state vector includes the vehicle position. Vehicle speed Vehicle speed Priority Delay . From vehicle matrix Extract columns relevant to requirements modeling to construct a domain vehicle communication quality requirements representation matrix. And construct a weight vector Thus, the target vector of vehicle integrated communication requirements is obtained. Then, the data is aggregated and reported to the central cloud. In this embodiment, the weight vector selected is based on the different importance levels of the data in different dimensions of the demand feature matrix. .

[0013] Step 1.1: Obtain the vehicle matrix using equation (1). (1) Step 1.2: Use equation (2) to obtain the domain vehicle communication quality requirement representation matrix. (2) Step 1.3: Obtain the target vector of vehicle integrated communication requirements using equation (3). (3) in This indicates the normalization process for column vectors. In this embodiment, the normalization process is Min-Max normalization.

[0014] Step 1.4: Calculate the target quantity of integrated communication demand within the roadside unit domain using equation (4) and report it to the central cloud. (4) in, It is a vector consisting entirely of 1s.

[0015] Step 2: The central cloud is based on each roadside unit. The reported demand information is for each roadside unit. Dynamically and adaptively allocate base station bearer budget For each roadside unit Provide differentiated resources to meet the communication needs of a sharp increase in local vehicles; Step 2.1: Define the base station in the time slot The total available carrying capacity budget is Use equation (5) to calculate the vehicle resource gap; (5) in, , For time slots The actual rate is calculated from the acquired resource blocks and the selected scheme. For smoothing coefficients; , The current minimum guaranteed rate preset by the system; Step 2.2: Calculate the roadside unit using equation (6). Demand rating; (6) in, , , These are the weighting coefficients. This indicates normalization processing.

[0016] Step 2.3: Allocate roadside units using formula (7) Budgetary support; (7) satisfy In this embodiment, the selected weights The system's default minimum guaranteed speed is 10Mbps. Step 3: After receiving the carrying budget, the roadside unit constructs a set of feasible transmission paths within the roadside unit. ; Step 3.1: Construct distance, relative speed, and single resource block rate parameters based on vehicle status.

[0017] Step 3.1.1: From the vehicle matrix Extract the position column vector and velocity column vector, and then combine the roadside units. As the node with index 0, it is incorporated into the first term of the vector. Let the equivalent position and equivalent velocity be respectively... , Then there is (8) (9) Position and velocity are expressed as scalars , Perform equivalent representation, where nodes Vehicle Index ;make It is an all-1 vector. In this embodiment, the vehicle-to-everything (V2X) scenario adopts a single-segment or corridor-type road modeling method, and the vehicle position is represented by arc-length coordinates along the road's travel direction. Therefore, the position variable... It is represented in scalar form. For two-dimensional road networks or more complex road topology scenarios, location variables... It can be extended to two-dimensional or multi-dimensional coordinate vectors accordingly, and the method of the present invention is equally applicable.

[0018] Step 3.1.2: Construct the position difference matrix and velocity difference matrix using equations (10) and (11): (10) (11) Step 3.1.3: Construct the distance matrix and relative velocity matrix using equations (12) and (13): (12) (13) Step 3.1.4: Obtain the matrix elements using equations (14) and (15): (14) (15) Step 3.1.5, Set up roadside units The pre-configured resource block set is Its elements are Calculate a single resource block using equation (16) At any two nodes Direct transmission rate between And summarize them into a direct connection rate matrix. Its matrix elements are: (16) in, Represents resource block bandwidth, Indicates from node arrive The transmission power, Indicates from node arrive The overall channel gain, This represents the equivalent noise power within the resource block bandwidth. For resource blocks The equivalent interference power term within the bandwidth originating from co-channel reused links in neighboring areas or other external transmitting nodes, and related to the distance between nodes; since the resource blocks within the roadside unit are orthogonally allocated in this embodiment, co-channel interference within the same roadside unit is not considered; if neighboring cell reuse interference is ignored, then let .

[0019] Step 3.1.6, for the vehicle With relay vehicles Define a unified single resource block transmission rate notation. ,when When, it indicates the vehicle Use direct connection to occupy resource blocks The transmission rate; when When, it indicates the vehicle Using vehicles Relay occupied resource blocks The transmission rate is calculated using equation (17): (17) in, Indicates vehicle With roadside units Between resource blocks Direct transmission rate on the device Indicates vehicle To the vehicle Between resource blocks V2V transmission rate on Indicates vehicle With roadside units Between resource blocks V2I transmission rate on The time-division coefficient for two-hop relays on the same resource block is used to characterize the effective throughput reduction caused by two-stage forwarding, and is set to (0,1). In this embodiment, .

[0020] Step 3.2: Calculate the single resource block transmission rate threshold using equation (18); (18) in, This is the threshold adjustment factor, used to compensate for transmission loss in relays. It is selected as... , Indicates vehicle Use direct connection to occupy resource blocks The transmission rate at that time. In this embodiment, .

[0021] like Then the vehicle Determined to pass resource block A relay vehicle is needed; if and Then in the roadside unit Continue to provide services for vehicles within the coverage area Looking for a relay vehicle; Step 3.3: Construct a relay link stability indicator matrix using the distance matrix and the relative velocity matrix. Its elements Indicates vehicle As The feasibility of relaying is demonstrated by calculating the matrix elements using equation (19): (19) in For indicator functions, This is the threshold for V2V communication distance. To prevent division by zero of extremely small positive numbers; if Then the vehicle will be retained. To the vehicle The transmission path; if If so, the transmission path is directly eliminated. In this embodiment, , .

[0022] Step 3.4: Compile the last remaining transmission paths into a set of feasible transmission paths. Among them, set elements Indicates vehicle In the time slot Through vehicle nodes Transmitted with roadside units; when When, it indicates the vehicle Using a direct connection method; when When, it indicates the vehicle Using vehicles The relay method.

[0023] In this embodiment, the link and scenario parameters are shown in Table 1, and the resource scheduling and budget parameters are shown in Table 2. Table 1 Link and Scenario Parameters

[0024] Among them, 3GPP TR38.901: Umi-StreetCanyon, LOS is the path loss reference protocol for V2I links; 3GPP TR37.885: Urban, LOS / NLOSv is the path loss reference protocol for V2V links; Table 2 Resource Scheduling and Budget Parameters

[0025] Step 4: Calculate the vehicle quota based on the target amount of comprehensive vehicle communication needs, and then calculate the cost of each feasible transmission path. Step 4.1, Roadside Unit Based on the load budget obtained in step 2 Based on the target amount of comprehensive vehicle communication demand, a proportional mapping is performed, and the rate quota vector of vehicles within the range is calculated using equation (20): (20) in, This is the vehicle speed quota vector.

[0026] Step 4.2, for each vehicle Based on its location within the set of feasible transmission paths The equivalent single resource block reference rate is calculated using equation (21) for the corresponding transmission path: (twenty one) Step 4.3: Calculate the ideal contiguous resource block requirement for the vehicle using equation (22): (twenty two) To prevent long-distance, high-demand vehicles from excessively consuming resource blocks, a resource block limit is set for each vehicle. Using equation (23), the truncated continuous demand is obtained: (twenty three) Step 4.4: If the sum of the consecutive demands of each vehicle after truncation does not exceed the total number of pre-configured resource blocks... Then let (twenty four) If the sum of the consecutive demands of each vehicle after truncation exceeds the total number of pre-allocated resource blocks. Then, the normalized contiguous resource block quota is obtained by proportionally compressing it: (25) The basic quota is obtained by rounding down the continuous quota using equation (26). (26) Calculate the number of remaining resource blocks using equation (27) (27) Define the decimal part using equation (28) (28) make for Select the first ones in descending order of value Given a set of indexes, the final resource block quota for a vehicle is given by equation (29). (29) Where the decimal parts are the same, they are randomly sorted. Indicates vehicle In the time slot The upper limit of the reference resource block quota.

[0027] Step 4.5: For each feasible transmission path The average single resource block transmission path rate is calculated using equation (30): (30) Define the edge cost from the vehicle node to the relay vehicle entry node. Calculate using equation (31): (31) in, This indicates normalization processing. These are the weighting coefficients. The path switching penalty is used to characterize the vehicle. In the current time slot Select transmission path Compared to the previous time slot, the transmission path has changed; when the vehicle In the time slot The transmission path used is the same as the current transmission path. When they are the same, take When the two are different, take the one that is different. In this embodiment, the normalization process is Min-Max normalization, and the weights are selected according to the different importance levels of the data. .

[0028] Step 5: Under capacity and cost constraints, construct a hierarchical relay cost flow network diagram based on V2V vehicle relays. A solution model is constructed, which is the hierarchical relay cost flow network diagram. This includes vehicle nodes, relay vehicle entry nodes, relay vehicle exit nodes, and resource block nodes; Step 5.1: Construct a hierarchical relay cost flow network diagram based on V2V vehicle relays. , where the set of nodes Including source point S, vehicle node Relay vehicle entry node Export nodes Resource block node and sink T. Where, when the transmission path middle When, it indicates the vehicle Using a direct connection method, the relay vehicle node is the vehicle. Its own carrying node, when When, it indicates the vehicle Using vehicles relay methods; Step 5.2: Construct a set of directed edges ,include :capacity = Cost = 0; When the transmission path When adding this edge, the capacity = Cost = ; Capacity = Cost = 0; Capacity = 1, Cost = , Capacity = 1, Cost = 0; in, For the relay vehicle carrying capacity side, its capacity Indicates vehicle In the time slot The maximum communication capacity within the vehicle, which can be used for both vehicles and other applications. It can be used for direct access to its own business, or to carry other vehicles via... The relay forwarding service. In this embodiment, .

[0029] Step 5.3: The solution model is defined in the hierarchical relay cost flow network graph. Determine the allocation variable above ,in Indicates vehicle via transmission path Occupying resource blocks .

[0030] The optimization criterion is to first maximize the number of resource blocks occupied, and then minimize the total cost of occupancy. This prioritizes maximizing the number of resource blocks occupied. (32) Further minimize the total cost of resource block allocation: (33) And it satisfies the following constraints: (34) (35) (36) (37) (38) Step 6: Based on the hierarchical relay cost flow network diagram The optimal resource allocation scheme is solved using the enhanced dynamic cost flow algorithm, and the allocation variables are output. .like Figure 2 As shown, the steps are as follows.

[0031] Step 6.1, in the scheduling time slot Initially, the roadside unit reads the time slot. Resource scheduling results The initial allocation variables for hot start are then mapped to the vehicle set, feasible transmission path set, and resource block set of the current time slot, and equation (39) is used to obtain the initial allocation variables for hot start. : (39) If the initial allocation variable for hot start obtained by equation (39) violates constraints (34)-(38), then a feasibility repair is performed on the conflicting allocation item. The feasibility repair is: among the allocation items that violate the constraints, the allocation item with the larger cost is deleted first until the constraints (34)-(38) are satisfied. Step 6.2, the above Mapping to the initial flow of the corresponding edge in the graph yields the initial flow. ; Step 6.3: Construct the residual network and find the minimum cost based on the shortest path. Augmenting path, iteratively augmented until the termination condition is met, outputting the optimal flow. ; Step 6.3.1, based on the current flow Constructing residual networks Using equation (40) for any side Define residual capacity (40) And set a residual capacity for each fallback reverse edge. ; Step 6.3.2, when the edge is Then, the cost is obtained using equation (41). (41) The cost of all other edges is 0, and the cost of the opposite edge is the opposite. Step 6.3.3, in all residual capacity In the available network composed of edges, according to edge cost Calculate the cumulative cost of the path and use the shortest path algorithm to find the path with the minimum cumulative cost. The through path is the minimum cost path. The path cost is calculated using equation (42). (42) Step 6.3.4: Calculate the augmentable flow of the path using equation (43). (43) Step 6.3.5: Perform augmenting update on the edges of the path: for forward edges Calculate its flow rate using equation (44) (44) The flow rate of the reverse side is calculated using equation (45). (45) Step 6.3.6, when there is no [from] Available paths Or the inflow at the sink reaches the total number of pre-configured resource blocks. When the iteration terminates and the optimal flow is output, the iteration is terminated. ; Step 6.4: Based on the optimal flow Determine the allocation variables The value of , for any If its corresponding complete path In optimal flow If the unit flow rate is below the limit, then let Otherwise, it is 0. This ultimately forms the roadside unit. In the time slot The final resource scheduling scheme within Simultaneously set a periodic reset threshold. ; Step 6.5, when completed continuously After the hot start solution for each time slot is obtained, the next time slot is forced to... A cold start is performed, and historical accumulated deviations are eliminated and scheduling stability is maintained through periodic resets. In this embodiment, the periodic reset threshold is... .

[0032] Step 7: Based on the allocation variables Generate roadside units Internal scheduling decisions are made and issued; historical vehicle statistics are updated. In relation to the scheme history, fairness gaps are used for subsequent time slots. With switching penalty calculate.

[0033] The following are examples: 1000 vehicles are evenly deployed along a 6000m road. Vehicle communication needs are randomly generated. Three Remote Units (RSUs) are positioned at 1000m, 3000m, and 5000m. A 1500m congestion section is established between 4500m and 6000m. The average vehicle speed within the congestion section is 5m / s with a standard deviation of 1m / s. The vehicle speed range within the non-congestion section is (15, 25)m / s. The system operates for 300 time slots. The average allocation scheme means that within each scheduling time slot, the bearer budget is allocated equally to the RSUs within the current base station's coverage area, and available resource blocks are allocated equally to the vehicles within the current RSU's coverage area. Historical guarantee gaps, demand scoring, handover penalties, and hot start mechanisms are not introduced.

[0034] The average transmission rate of vehicles within the RSU coverage area at 5000m was measured as a function of the number of vehicles. Figure 3 As shown; The total throughput of the two schemes was measured as follows: Figure 4 As shown, this scheme is superior to the comparison scheme throughout the entire process.

[0035] To measure the fairness of resource allocation under different vehicle communication needs, this embodiment introduces the Jain fairness index for rate demand satisfaction. Let the vehicles... In the time slot The target rate requirement is The actual transmission rate obtained is First, define the vehicle. In the time slot The degree of demand satisfaction is:

[0036] in, This is a preset minimum value greater than 0, used to avoid a denominator of 0. Further, let the total number of vehicles within the current statistical range be... The Jain Fairness Index for rate demand satisfaction is defined as follows:

[0037] in, The closer the value of is to 1, the more balanced the satisfaction of each vehicle under its respective demand constraints, and therefore the better the system fairness. It should be noted that the comprehensive vehicle demand in this embodiment considers factors such as rate demand, latency constraints, and historical guarantee gaps, and is used for resource block budget allocation and transmission path scheduling. To facilitate a quantitative comparison of the balance in satisfying vehicle service transmission demands under different schemes, this embodiment further uses the Jain fairness index for rate demand satisfaction. This index mainly reflects the fairness of satisfying rate demands for each vehicle, and does not directly characterize the fairness in the latency dimension. Based on this index, the average allocation scheme and the scheme of this invention are compared and evaluated.

[0038] like Figure 5 As shown, the traditional Jain fairness index is calculated based solely on the actual speed obtained by vehicles, primarily reflecting the balance of speed allocation, but failing to reflect the fairness of the degree to which the speed requirements of each vehicle are met under heterogeneous business conditions. Therefore, this embodiment uses the aforementioned Jain fairness index for speed requirement satisfaction for comparative evaluation. The results show that the proposed solution is significantly higher than the average allocation solution throughout the entire operation, indicating that the proposed solution has better fairness in ensuring differentiated speed requirements of vehicles.

[0039] In summary, this invention, in dynamic vehicular networks where vehicle link conditions vary significantly and service requests change rapidly over time, transforms the resource scheduling problem into a cost-flow network optimization model through a hierarchical mechanism of cloud-based dynamic bearer budget allocation and roadside unit local fine-grained scheduling. By utilizing a multi-dimensional edge cost function that integrates rate, demand, guarantee gap, and handover penalty, combined with an enhanced dynamic cost-flow algorithm, the invention achieves the dual objectives of maximizing system throughput and ensuring service fairness. This provides a hierarchical relay cost-flow method for vehicular network resource scheduling under localized traffic bursts.

[0040] In another aspect, the present invention also discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the method described above.

[0041] In another aspect, the present invention also discloses a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the method described above.

[0042] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute the hierarchical relay cost flow method for vehicle network resource scheduling under any local traffic burst in the above embodiments.

[0043] It is understood that the systems, devices, and storage media provided in the embodiments of the present invention correspond to the methods provided in the embodiments of the present invention, and the explanations, examples, and beneficial effects of the relevant content can be referred to the corresponding parts of the above methods.

[0044] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0045] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0046] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0047] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A hierarchical relay cost flow method for vehicular network resource scheduling under localized traffic bursts, characterized in that, Includes the following steps: Step 1: In the scheduling time slot Within the system, the roadside unit set is as follows: ; For any roadside unit The collection of vehicles within the coverage area is In the roadside unit Internal vehicle indexing ,and Roadside unit Building a vehicle matrix vehicle matrix No. Corresponding vehicle In the time slot Vehicle state vector; roadside unit From vehicle matrix Columns related to demand modeling are extracted from the data to construct a domain-specific vehicle communication quality demand representation matrix. And construct a weight vector Thus, the target vector of vehicle integrated communication requirements is obtained. Then, the overall communication requirements of the vehicles are summarized and reported to the central cloud; T represents the transpose matrix; Step 2: The central cloud is based on each roadside unit. The reported demand information is for each roadside unit. Dynamically and adaptively allocate base station bearer budget For each roadside unit Provide differentiated resources to meet the communication needs of a sharp increase in local vehicles; Step 3, Roadside Unit Upon receiving the carrying budget Then, in the roadside unit Internally, a set of feasible transmission paths is constructed using a transmission path generation algorithm oriented towards link stability. ; Step 4: Calculate the vehicle quota based on the target amount of comprehensive vehicle communication needs, and then calculate the cost of each feasible transmission path. Step 5: Under capacity and cost constraints, construct a hierarchical relay cost flow network diagram based on V2V vehicle relays. A solution model is constructed, which is the hierarchical relay cost flow network diagram. This includes vehicle nodes, relay vehicle entry nodes, relay vehicle exit nodes, and resource block nodes; Step 6: Based on the hierarchical relay cost flow network diagram The optimal resource allocation scheme is solved using the enhanced dynamic cost flow algorithm, and the allocation variables are output. ; Step 7: Based on the allocation variables Generate roadside units Internal scheduling decisions are made and issued. Update vehicle historical statistics rate In relation to the scheme history, fairness gaps are used for subsequent time slots. With switching penalty calculate.

2. The hierarchical relay cost flow method for vehicular network resource scheduling under localized traffic bursts as described in claim 1, characterized in that, The vehicle state vector includes the vehicle position. Vehicle speed Vehicle target communication rate Priority Delay .

3. The hierarchical relay cost flow method for vehicular network resource scheduling under localized traffic bursts as described in claim 2, characterized in that, Step 1 involves matrix construction as follows: Step 1.1, Roadside Unit Based on the location of each vehicle within the coverage area ,speed Target communication rate Priority Delay Building a vehicle matrix ; Step 1.2, from the vehicle matrix Extract the target communication rate, priority, and latency-related columns to form a domain-specific vehicle communication quality requirement representation matrix. : Step 1.3: Based on the normalized intra-domain vehicle communication quality requirement representation matrix With weight vector By performing weighted calculations, the overall communication requirement target vector of the vehicle is obtained. : Step 1.4: Target vector of the vehicle's comprehensive communication requirements Scalarization is performed to obtain the target quantity of comprehensive communication demand within the domain. It was then reported to the central cloud.

4. The hierarchical relay cost flow method for vehicular network resource scheduling under localized traffic bursts as described in claim 3, characterized in that, Step 2 is performed as follows: Step 2.1: Define the base station in the time slot The total available carrying capacity budget is ; The vehicle resource gap is calculated based on the current actual vehicle speed, historical statistical speed, and minimum guaranteed speed. ; Step 2.2, based on roadside units The weighted calculation of the roadside unit is based on the resource gap of each vehicle within the area, the target amount of comprehensive communication demand within the region, and the number of vehicles. Demand rating ; Step 2.3: Allocate roadside units using formula (7) Budgetary support; (7) satisfy .

5. The hierarchical relay cost flow method for vehicular network resource scheduling under localized traffic bursts as described in claim 4, characterized in that, Step 3 is performed as follows: Step 3.1: Construct distance, relative speed, and single resource block rate parameters based on vehicle status; Step 3.1.1: From the vehicle matrix Extract the position column vector and velocity column vector, and then combine the roadside units. The node at index 0 is incorporated into the first term of the vector. Based on the position and velocity columns, the roadside unit is obtained. Distance parameters and relative speed parameters between and between vehicles; Step 3.1.5, Set up roadside units The pre-configured resource block set is Its elements are resource blocks Based on resource blocks Given the bandwidth, inter-node transmit power, integrated channel gain, equivalent noise power, and equivalent interference power, calculate the bandwidth between any two nodes in the resource block. The direct transmission rate on the network is denoted as . , indicating time slot internal nodes arrive Direct transmission rate on the device; Step 3.1.6, for the vehicle With relay vehicles Define a unified single resource block transmission rate notation. ,when When, it indicates the vehicle Use direct connection to occupy resource blocks The transmission rate; when When, it indicates the vehicle Using vehicles Relay occupied resource blocks The transmission rate is calculated using equation (17): (17) in, Indicates vehicle With roadside units Between resource blocks Direct transmission rate on the device Indicates vehicle To the vehicle Between resource blocks V2V transmission rate on Indicates vehicle With roadside units Between resource blocks V2I transmission rate on The time-division coefficient for two-hop relays on the same resource block is used to characterize the effective throughput reduction caused by two-stage forwarding, and is set to (0,1). Step 3.2: Calculate the single resource block transmission rate threshold using equation (18); (18) in, This is the threshold adjustment factor, used to compensate for transmission loss in relays. It is selected as... , Indicates vehicle Use direct connection to occupy resource blocks Transmission rate at that time; like Then the vehicle Determined to pass resource block A relay vehicle is needed; if and Then in the roadside unit Continue to provide services for vehicles within the coverage area Looking for a relay vehicle; Step 3.3: Construct a relay link stability indicator matrix using the distance matrix and the relative velocity matrix. Its elements Indicates vehicle As The feasibility of relaying is demonstrated by calculating the matrix elements using equation (19): (19) in For indicator functions, This is the threshold for V2V communication distance. To prevent division by zero of extremely small positive numbers; if Then the vehicle will be retained. To the vehicle The transmission path; if If so, the transmission path will be directly eliminated; Step 3.4: Compile the last remaining transmission paths into a set of feasible transmission paths. , where set elements Indicates vehicle In the time slot Through vehicle nodes With roadside units Transmission is performed; when When, it indicates the vehicle Using a direct connection method; when When, it indicates the vehicle Using vehicles The relay method.

6. The hierarchical relay cost flow method for vehicular network resource scheduling under localized traffic bursts as described in claim 5, characterized in that, Step 4 is performed as follows: Step 4.1, Roadside Unit Based on the load budget obtained in step 2 Based on the target amount of comprehensive vehicle communication demand, a proportional mapping is performed, and the rate quota vector of vehicles within the range is calculated using equation (20): (20) in, For vehicle speed quota vector; Step 4.2, for each vehicle Based on its location within the set of feasible transmission paths The equivalent single resource block reference rate is calculated using equation (21) for the corresponding transmission path: (21) Step 4.3: Calculate the ideal contiguous resource block requirement for the vehicle using equation (22): (22) To prevent long-distance, high-demand vehicles from excessively consuming resource blocks, a resource block limit is set for each vehicle. Using equation (23), the truncated continuous demand is obtained: (23) Step 4.4: When the total resource block demand of all vehicles does not exceed the total number of pre-configured resource blocks, retain the current resource block demand of each vehicle and integerize it to obtain the upper bound of the vehicle's reference resource block quota. When the total resource block demand of all vehicles exceeds the total number of pre-allocated resource blocks, the resource block demand of each vehicle is normalized and compressed before being integerized to obtain the upper bound of the vehicle reference resource block quota. ; Step 4.5: For each feasible transmission path The average single resource block transmission path rate is calculated using equation (30): (30) Define the edge cost from the vehicle node to the relay vehicle entry node. Calculate using equation (31): (31) in, This indicates normalization processing. These are the weighting coefficients. The path switching penalty is used to characterize the vehicle. In the current time slot Select transmission path Compared to the previous time slot, the transmission path has changed; when the vehicle In the time slot The transmission path used is the same as the current transmission path. When they are the same, take When the two are different, take the one that is different. .

7. The hierarchical relay cost flow method for vehicular network resource scheduling under localized traffic bursts as described in claim 6, characterized in that, Step 5 is performed as follows: Step 5.1: Construct a hierarchical relay cost flow network diagram based on V2V vehicle relays. , where the set of nodes Including source point S, vehicle node Relay vehicle entry node Export nodes Resource block node and sink T; where, when the transmission path middle When, it indicates the vehicle Using a direct connection method, the relay vehicle node is the vehicle. Its own carrying node, when When, it indicates the vehicle Using vehicles relay methods; Step 5.2: Construct a set of directed edges ,include :capacity = Cost = 0; When the transmission path When adding this edge, the capacity = Cost = ; Capacity = Cost = 0; Capacity = 1, Cost = , Capacity = 1, Cost = 0; in, For the relay vehicle carrying capacity side, its capacity Indicates vehicle In the time slot The maximum communication capacity within the vehicle, which can be used for both vehicles and other applications. It can be used for direct access to its own business, or to carry other vehicles via... Relay forwarding services; Step 5.3: The solution model is defined in the hierarchical relay cost flow network graph. Determine the allocation variable above ,in Indicates vehicle via transmission path Occupying resource blocks ; The solution model employs a lexicographical optimization criterion of first maximizing resource block occupancy and then minimizing total occupancy cost, and satisfies the following constraints: Each resource block At most one transmission path can be allocated within the same time slot; each vehicle The number of resource blocks acquired shall not exceed the upper bound of its reference resource block quota. It does not belong to the set of feasible transmission paths. The transmission path does not participate in resource allocation; each relay vehicle The amount of traffic carried does not exceed its maximum communication capacity. Roadside unit The total transmission rate corresponding to all allocated resource blocks within the cloud shall not exceed the bearer budget allocated by the central cloud. .

8. The hierarchical relay cost flow method for vehicular network resource scheduling under localized traffic bursts as described in claim 7, characterized in that, Step 6 is performed as follows: In step 6, the roadside unit Read the resource scheduling result of the previous time slot This is then mapped onto the current time slot's vehicle set, feasible transmission path set, and resource block set to obtain the initial allocation result for hot start; If the initial allocation result of the hot start violates the constraints, the conflicting allocation items with higher costs are deleted first to complete the feasibility repair. The repaired initial allocation result is mapped to the initial flow in the hierarchical relay cost flow network graph. The minimum cost augmenting path is iteratively searched based on the residual network and the flow is updated until the termination condition is met to obtain the optimal flow. Based on optimal flow Determine the allocation variables The value of , for any If its corresponding complete path In optimal flow If the unit flow rate is below the limit, then let Otherwise, it is 0; when the continuous warm start reaches the periodic reset threshold, a cold start reset is performed.

9. The hierarchical relay cost flow method for vehicular network resource scheduling under localized traffic bursts as described in claim 8, characterized in that, The shortest path algorithm includes the Bellman-Ford algorithm or the SPFA algorithm.

Images