A dynamic routing method for drone swarm networks based on general computing function slicing

Abstract

The present invention relates to a dynamic routing method for a drone cluster network based on general computing function slices, comprising: constructing a unified general computing identifier and efficiency constraint equation for the drone cluster network, a unified general computing identifier for nodes in the drone cluster network, including node identifiers, link identifiers, computing power identifiers, path identifiers, and service identifiers; establishing a general computing demand service graph, combining transmission service identifiers, computing service identifiers, and efficiency constraint equations, analyzing the supply and demand relationship and service constraints between general computing function nodes and users using a knowledge graph method, and constructing a service supply and demand graph; and implementing dynamic routing based on general computing slice selection, implementing general computing fusion routing through slice selection based on the general computing service supply and demand graph, and dynamically adjusting routing to meet on-demand responses for user services. The present invention solves the problem of unclear multilateral collaborative matching relationships between general computing service demands and functional resources in future wireless networks based on drone cluster networks.

Description

Technical Field

[0001] The present invention belongs to the field of wireless communication technology, and in particular relates to a dynamic routing method for drone cluster networks based on general computing function slicing. Background Art

[0002] With the large-scale commercialization of the fifth-generation mobile communication system (5G), the global industry has begun systematic research on the sixth-generation mobile communication system (6G). From mobile interconnection to the Internet of Everything, and then to the Intelligent Connection of Everything, 6G will promote the construction of an inclusive and intelligent human society, meeting the needs of new 6G business scenarios such as super-efficient transportation, holographic interaction, precision medicine, and smart industry. Aiming at the 6G business requirements of full coverage, intelligent scene connection, and information coupling, drone swarm networks, with their advantages such as rapid response, flexible deployment, and multi-functional support, have a natural advantage in user-centric business applications. They have become a key component of the 6G integrated three-dimensional network and embody a key architectural feature for the vigorous development of future communication technologies. The 6G white paper points out that the deep integration of communication and computing functions is a key technical feature of 6G's inherent intelligence. By designing the 6G network architecture to natively support the deep integration of communication and computing, 6G services can be better enabled, providing more powerful and intelligent integrated computing services.

[0003] In the above vision, drone swarm networks, with their unique capabilities and advantages, have added unprecedented flexibility and innovation to the efficient computing integration of 6G. Computing integration empowerment has become a new development trend of drone swarm networks under the 6G business model.

[0004] In drone cluster networks, the specific advantages of implementing 6G integrated computing services are as follows:

[0005] (1) Distributed and flexible deployment of nodes: Because user distribution is difficult to predict, traditional ground base stations are often deployed empirically, resulting in uneven network computing load. In contrast, drone deployment is flexible and decentralized, not restricted by buildings and roads, which is conducive to the distributed scheduling response of 6G computing services.

[0006] (2) Agile computing function guarantee: UAVs have highly dynamic self-organizing characteristics and can provide agile computing services.

[0007] UAV swarm networks, oriented towards efficient computing integration, aim to provide full-area coverage, intelligent flexibility, and rapid response for 6G services. However, the highly dynamic and weakly connected networking characteristics and the differentiated and discrete distribution of computing functions are key challenges that hinder UAV swarm networks from providing high-quality, ultra-low latency services. The main reasons are as follows:

[0008] (1) The drone cluster network is a heterogeneous network with a wide variety of access devices, diverse locations, and high mobility of drone nodes. The network is subject to adverse effects such as wireless interference, Doppler frequency shift, and multipath delay, which makes the network topology time-sensitive and the communication environment complex and changeable.

[0009] (2) In view of the communication characteristics of 6G ubiquitous access and full coverage, the problem of discrete and highly differentiated distribution of physical resources in drone cluster networks has become increasingly prominent. In particular, facing the demand for full coverage, relying solely on resource replacement methods such as limited increases in spectrum resources will inevitably lead to inefficient transmission of the system, which in turn will cause problems such as communication resource redundancy. The relatively limited energy supply of drone nodes greatly limits the node's transmission power, making it difficult to overcome the transmission loss of long-distance space communications. The dispersed deployment of general computing function nodes makes it difficult to uniformly schedule computing functions.

[0010] (3) Due to the heterogeneity of 6G services, user services have diverse computing requirements. The matching and coupling problem between the service requirements of user services and the resource supply of computing function nodes needs to be solved urgently, and the flexible configuration of computing functions is particularly important.

[0011] However, current research on collaborative routing in drone swarm networks primarily focuses on maintaining link connectivity to establish globally reachable routing tables, or on improving existing routing methods for physical transmission links based on certain network performance metrics. There has been no systematic study of functional coupling routing methods and addressing methods for interoperable fusion services from the perspective of interoperability. Therefore, given the heterogeneous demands of 6G services and the differentiated distribution of interoperability functions across network nodes, it is of practical significance to characterize the demand-function matching relationship between users and nodes based on 6G service characteristics, implement routing addressing for interoperability function nodes to meet diverse service needs, and resolve the unclear multilateral collaborative matching relationship between service demands and functional resources. Summary of the Invention

[0012] In response to the shortcomings in related technologies, the present invention provides a dynamic routing method for drone cluster networks based on general computing function slicing to solve the problem of unclear multilateral collaborative matching relationship between general computing service requirements and functional resources in future wireless networks based on drone cluster networks.

[0013] According to one aspect of the present application, a dynamic routing method for a drone cluster network based on general computing function slicing is provided, including: constructing a general computing unified identification and efficiency constraint equation for the drone cluster network, a general computing unified identification for nodes in the drone cluster network, including node identification, link identification, computing power identification, path identification and service identification; establishing a general computing demand service graph, combining transmission service identification, computing service identification and efficiency constraint equation, using the knowledge graph method to analyze the supply and demand relationship and service constraints between general computing function nodes and users, and constructing a service supply and demand graph; and realizing dynamic routing based on general computing slice selection, realizing general computing fusion routing through slice selection based on the general computing service supply and demand graph, and dynamically adjusting routing to meet the on-demand response of user services.

[0014] In a possible implementation, the efficiency constraint equation of the UAV swarm network is:

[0015]

[0016] in, is the total spectrum of the system; is the maximum transmit power of the node; and Respectively represent the maximum CPU computing frequency and maximum storage energy of the node; W n is the spectrum bandwidth available to node n, P n is the transmit power of node n.

[0017] In one possible implementation, establishing a general computing demand service graph further includes:

[0018] Define the user node set and the general function node set;

[0019] Based on user business needs and the service capabilities of the inter-computing function nodes, an interaction matrix is ​​established to represent the supply and demand relationship between users and the inter-computing function nodes; and

[0020] Using the knowledge graph method, the supply and demand relationship is mapped into triples to construct a service supply and demand graph.

[0021] In a possible implementation, the user node set and the general computing function node set are defined as U and V respectively;

[0022] Establishing the interaction matrix between user and general computing function nodes Among them, x u,v =1 indicates that user u and the general computing function node v have generated a general computing interaction; otherwise, x u,v =0.

[0023] In a possible implementation manner, the transmission service identifier and the computing service identifier in the service requirement of user u are respectively represented as T_SIDu and C_SID u , the link identifier and computing power identifier of the computing function node v are represented as LID v and CID v , among which LID v and CID v Satisfy the efficiency constraint equation.

[0024] In one possible implementation, establishing a general computing demand service graph further includes:

[0025] The knowledge graph that defines the supply and demand relationship between users and general computing function nodes is:

[0026] G={([T_SID u ,C_SID u ],e u,v ,[LID v ,CID v ])|u∈U,v∈V,e u,v ∈{0,1}},

[0027] Among them, the triple ([T_SID u ,C_SID u ],e,[LID v ,CID v ]) represents the general computing resource of the general computing function node v [LID v ,CID v ]Through the relationship e u,v The general calculation requirements of the head entity u / user u [T_SID u ,C_SID u ]Service, Relationship u,v Indicates whether the physical distance between the computing function node and the user service is reachable, e u,v =1 means the physical distance is reachable, e u,v =0 means the physical distance is unreachable.

[0028] In one possible implementation, dynamic routing is implemented based on the general computation slice selection, including:

[0029] The general computing function slice serving user u is represented as S u , S u The corresponding general function node set is ψ u , user u in ψ u The steady-state performance function under is expressed as:

[0030]

[0031] And it complies with the following constraint equations to meet the user's general calculation requirements:

[0032]

[0033]

[0034]

[0035] Based on the above technical solution, the dynamic routing method of the drone cluster network based on general computing function slicing of the present invention solves the problem of unclear multilateral collaborative matching relationship between general computing service requirements and functional resources in future wireless networks based on drone cluster networks by constructing a unified general computing identification and efficiency constraint equation for the drone cluster network, establishing a general computing demand service map, and realizing dynamic routing based on general computing slice selection. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0037] Figure 1 This is a flow chart of a method for dynamic routing of a drone cluster network based on a unified computing identifier and a common computing function slice according to an embodiment of the present invention;

[0038] Figure 2 Schematic diagram of the general identification code;

[0039] Figure 3 A diagram of the dynamic routing architecture for slicing nodes for general computing functions. DETAILED DESCRIPTION

[0040] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of the present invention.

[0041] In the description of the present invention, it should be understood that the terms "center", "transverse", "longitudinal", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the present invention.

[0042] The terms "first," "second," and "third" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or to implicitly specify the quantity of the technical features indicated. Therefore, a feature specified as "first," "second," or "third" may explicitly or implicitly include one or more of such features.

[0043] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood broadly. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to direct connections, indirect connections through an intermediary, or internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0044] In order to solve the problem of unclear multilateral coordination matching between computing service requirements and functional resources in future wireless networks based on drone cluster networks, this application provides a dynamic routing method for drone cluster networks based on computing function slicing.

[0045] See also Figure 1 In one possible implementation, a method for dynamic routing of a drone swarm network based on general computing function slicing includes: in step S10, constructing a general computing unified identifier and efficiency constraint equation for the drone swarm network, wherein the general computing unified identifier of nodes in the drone swarm network includes a node identifier, a link identifier, a computing power identifier, a path identifier, and a service identifier; in step S20, establishing a general computing demand service graph, combining transmission service identifiers, computing service identifiers, and efficiency constraint equations, and using a knowledge graph method to analyze the supply and demand relationship and service constraints between general computing function nodes and users, thereby constructing a service supply and demand graph; and in step S30, implementing dynamic routing based on general computing slice selection, implementing general computing integrated routing through slice selection based on the general computing service supply and demand graph, and dynamically adjusting routing to meet on-demand response requirements for user services.

[0046] In this solution, a unified identification system for the drone swarm network, along with efficiency constraint equations, is first established to uniformly identify network node identities, links, computing power, paths, and services. Next, a unified service graph for computing demand is established. Combining transmission and computing service identifiers with efficiency constraint equations, knowledge graphs are used to analyze the supply and demand relationships and service constraints between computing nodes and users, thereby constructing a service supply and demand graph. Finally, dynamic routing is implemented based on computing slice selection. By analyzing the service supply and demand graph, appropriate slices are selected for computing converged routing, enabling dynamic routing adjustments to meet user service needs.

[0047] By constructing a unified universal computing identifier and efficiency constraint equation, this method accurately identifies and manages each node in a drone swarm network, along with its universal computing capabilities and service requirements. By analyzing supply and demand relationships through a knowledge graph and constructing a service supply and demand map, this approach helps optimize resource allocation and service quality. A dynamic routing mechanism based on universal computing slice selection enables flexible routing strategy adjustments, improving network adaptability and responsiveness to meet dynamic user needs.

[0048] In a possible implementation, the efficiency constraint equation of the UAV swarm network is:

[0049]

[0050] in, is the total spectrum of the system; is the maximum transmit power of the node; and Respectively represent the maximum CPU computing frequency and maximum storage energy of the node; W n is the spectrum bandwidth available to node n, P n is the transmit power of node n.

[0051] In the above solution, the efficiency constraint equations for the drone swarm network constrain and optimize the resource usage of each node in the network by defining the total system spectrum, the node's maximum transmit power, the maximum CPU computing frequency, the maximum energy storage, the node's available spectrum bandwidth, and the transmit power. These equations ensure that each node can efficiently perform its computing function while meeting the required efficiency.

[0052] In one possible implementation, establishing a general computing demand service graph further includes:

[0053] Define the user node set and the general function node set;

[0054] Based on user business needs and the service capabilities of the inter-computing function nodes, an interaction matrix is ​​established to represent the supply and demand relationship between users and the inter-computing function nodes; and

[0055] Using the knowledge graph method, the supply and demand relationship is mapped into triples to construct a service supply and demand graph.

[0056] By defining a set of user nodes and a set of general computing nodes, and based on user business needs and the service capabilities of the general computing nodes, an interaction matrix is ​​established to represent the supply and demand relationship between users and the general computing nodes. Then, using knowledge graphs, this supply and demand relationship is mapped into triples, constructing a service supply and demand graph. This process ensures accurate representation and analysis of supply and demand relationships, providing a foundation for the implementation of dynamic routing.

[0057] In a possible implementation, the user node set and the general computing function node set are defined as U and V respectively;

[0058] Establishing the interaction matrix between user and general computing function nodes Among them, x u,v =1 indicates that user u and the general computing function node v have generated a general computing interaction; otherwise, x u,v =0.

[0059] The interaction matrix intuitively represents the interaction relationship between users and general computing function nodes, which facilitates subsequent analysis and processing.

[0060] In a possible implementation manner, the transmission service identifier and the computing service identifier in the service requirement of user u are respectively represented as T_SID u and C_SID u , the link identifier and computing power identifier of the computing function node v are represented as LID v and CID v , among which LID v and CID v Satisfy the efficiency constraint equation.

[0061] In the above solution, the transmission service identifier and computing service identifier in the user u business requirement are defined as T_SID u and C_SID u , the link identifier and computing power identifier of the computing function node v are represented as LID v and CID v and ensure that the LID v and CID v Satisfying the efficiency constraint equation can accurately represent the relationship between user needs and node capabilities.

[0062] This identification method can accurately reflect the relationship between user needs and node capabilities, facilitating analysis and optimization of resource allocation. At the same time, ensuring that these identifications meet the efficiency constraint equation can improve the overall efficiency of the network and meet user needs.

[0063] In one possible implementation, establishing a general computing demand service graph further includes:

[0064] The knowledge graph that defines the supply and demand relationship between users and general computing function nodes is:

[0065] G={([T_SID u ,C_SID u ],e u,v ,[LID v ,CID v ])|u∈U,v∈V,e u,v ∈{0,1}},

[0066] Among them, the triple ([T_SID u ,C_SID u ],e,[LID v ,CID v ]) represents the general computing resource of the general computing function node / tail entity v [LID v ,CID v ]Through the relationship e u,v The general calculation requirements of the head entity u / user u [T_SID u ,C_SID u ]Service, Relationship u,v Indicates whether the physical distance between the computing function node and the user service is reachable, e u,v =1 means the physical distance is reachable, e u,v =0 means the physical distance is unreachable.

[0067] By building a knowledge graph, we can intuitively and efficiently represent the supply and demand relationships and physical accessibility between users and interoperability nodes, facilitating analysis and optimizing resource allocation. This representation method also improves the efficiency of data processing and analysis, providing a foundation for the implementation of dynamic routing.

[0068] In one possible implementation, dynamic routing is implemented based on the general computation slice selection, including:

[0069] The general computing function slice serving user u is represented as S u , S u The corresponding general function node set is ψ u , user u in ψ u The steady-state performance function under is expressed as:

[0070]

[0071] And it complies with the following constraint equations to meet the user's general calculation requirements:

[0072]

[0073]

[0074]

[0075] Based on user needs, the appropriate computing slice is dynamically selected to achieve efficient routing optimization. By meeting certain constraints, the user's computing needs are ensured, improving the overall efficiency and response speed of the network.

[0076] The dynamic routing method of a drone cluster network based on general computing function slicing according to an embodiment of the present invention is described in detail as follows:

[0077] In response to the heterogeneous demands of 6G services and the differentiated distribution of computing functions of network nodes, a computing service supply and demand map is established based on the characteristics of 6G services. The demand-function matching relationship between users and nodes is characterized, and then computing function slicing is used to meet different business needs, realize routing addressing of computing function nodes, and solve the problem of unclear multilateral collaborative matching relationship between service demand and functional resources. A dynamic routing method for drone cluster networks based on computing function slicing is provided.

[0078] See also Figure 1 , the specific steps of the present invention are as follows:

[0079] (1) Step 1

[0080] In response to the heterogeneous needs of 6G services and the heterogeneous functional services of computing nodes, a unified computing identifier for nodes in the UAV cluster network is established. The unified computing identifier is defined as a field carried in the data packet to declare the network node and its computing functions and service requirements. By appropriately defining the computing identifier code

[0081] (Integrated Communication And Computing Identifier, ICAC), design coding rules to provide a unique naming method for computing functions and services, so as to characterize the system performance during the execution of computing tasks.

[0082] Combined with the two-layer service architecture of the Handle identification system, the general computing function-service identification space of the UAV cluster network consists of two layers. The top layer is the node identifier (Nid) and the bottom layer is the function-service identifier (F-Sid). The same layers are fair and equal. The general computing identification code is defined as <icac> ::＝ <nid> / <f-sid>,like Figure 2 shown.

[0083] UTF-8 is the unique encoding of the universal calculation identification code, and the node identifier and function-service identifier are separated by the ASCII character " / ". In addition, the function-service identifier includes the link identifier (Lid), calculation identifier (Cid), path identifier (Pid), and service identifier (Sid).

[0084] 1) Node identification: Identify the distributed nodes in the drone swarm network, assign a unique identifier within the entire network, and represent information such as the node location.

[0085] Nid n ::= <n,loc n >

[0086] Among them, n is the node number, loc n is the three-dimensional geographic coordinate of node n.

[0087] 2) Link identification: Different nodes in a drone cluster network have different transmission capabilities. The communication capabilities of the nodes, including spectrum and transmit power, are uniformly characterized, and the link status of the nodes is communicated with the control plane in real time based on business needs.

[0088] Lid n ::= <W n ,P n >

[0089] Among them, W n is the spectrum bandwidth available to node n, P n is the transmit power of node n.

[0090] 3) Computing power identification: When network nodes exchange status information with the control plane, they also need to notify their own real-time computing power status based on business needs, such as CPU frequency, node energy, etc.

[0091] Cid n ::= <f n ,E n >

[0092] Among them, f n The CPU frequency calculated for node n, E n is the remaining available energy of node n.

[0093] 4) Path identification: The node path for distributed execution of general computing tasks, which manages the global execution of general computing tasks in a fine-grained manner.

[0094]

[0095] in, is the set of nodes that execute the general computing task of node n.

[0096] 5) Service Identification: To address the heterogeneous needs of inter-computing services, service identification is used to characterize the inter-computing service requirements of nodes. Service identification includes the transmission service identification (T_Sid), which focuses on acquiring content, and the computing service identification (C_Sid), which focuses on performing computations.

[0097]

[0098] Among them, Q n Represents the total computing task data volume of node n, Indicates the transmission time limit of the computation task of node n.

[0099]

[0100] in, Indicates the computation time limit for the total computation task of node n.

[0101] In addition, according to the efficiency conditions of the UAV cluster network, the efficiency constraint equation is established as follows:

[0102]

[0103] Among them, the cluster network adopts orthogonal multiple access. is the total spectrum of the system; is the maximum transmit power of the node; and They represent the maximum CPU computing frequency and maximum energy storage of the node respectively.

[0104] (2) Step 2

[0105] Combining the transmission service identifier T_Sid, computing service identifier C_Sid, and efficiency constraint equation of the drone cluster network, the knowledge graph method is used to analyze the supply and demand relationship between the general computing function node and the user and the service constraints, and to construct a service supply and demand graph. The user node set and the general computing function node set (ground base station, drone node, and user itself) are defined as U and V respectively. Based on the implicit feedback based on the business completion, the user-general computing function node interaction matrix is ​​obtained. That is, whether the general computing resources provided by the general computing function node set are used for the general computing needs of users. u,v =1 indicates that user u and the general computing function node v have generated a general computing interaction; otherwise, x u,v =0.

[0106] The transmission service identifier and computing service identifier in the user u business requirement are respectively represented as T_SID u and C_SID u , the link identifier and computing power identifier of the computing function node v are represented as LID v and CID v , among which LID v and CID v Limited by the efficiency constraint equation. Among them, T_SID u and LID v Refers to bandwidth and transmit power; C_SID u and CID v Refers to CPU computing frequency and node energy storage.

[0107] The knowledge graph that defines the supply and demand relationship between users and general computing function nodes is G = {([T_SID u ,C_SID u ],e u,v ,[LID v ,CID v ])|u∈U,v∈V,e u,v ∈{0,1}}. Among them, the triple ([T_SID u ,C_SID u ],e,[LID v ,CID v ]) represents the computing resources of the computing function node v / -v [LID v ,CID v ]Through the relationship e u,v The general calculation requirements of the head entity u / user u [T_SID u ,C_SID u ]Service, Relationship u,v Indicates whether the physical distance between the computing function node and the user service is reachable, e u,v =1 means the physical distance is reachable, e u,v =0, otherwise the opposite is true.

[0108] (3) Step 3

[0109] Based on the supply and demand graph of the general computing service, in order to achieve dynamic response to user business needs, the general computing fusion routing is realized through slice selection, such as Figure 3 shown.

[0110] The general computing function slice serving user u is represented as S u , S u The corresponding general function node set is ψ u , user u in ψ u The steady-state performance function under is expressed as:

[0111]

[0112] Among them, the following constraints need to be met to ensure that the user's general computing needs are met.

[0113]

[0114] The present invention provides a dynamic routing method for drone cluster networks based on general computing function slicing, which has the following beneficial effects:

[0115] The dynamic routing method for drone cluster networks based on general computing function slicing has a wide range of applicability and can provide planning guidance for the functional addressing of complex general computing services in future networks.

[0116] A service supply and demand graph is established based on a unified universal computing identifier for universal computing functions and needs. This effectively depicts the demand-function matching relationship between users and nodes, reducing the complexity of service matching implementation.

[0117] The general computing function slice selection mechanism demonstrates the flexibility of dynamic resource allocation in the network, which will be more meaningful in the distributed execution and service response of future network general computing tasks.

[0118] Finally, it should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0119] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the same. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the specific implementation methods of the present invention may still be modified or some technical features may be replaced by equivalents without departing from the spirit of the technical solutions of the present invention, and all of these should be included in the scope of the technical solutions claimed for protection by the present invention. < / nid> < / icac>

Claims

1. A dynamic routing method for drone cluster networks based on general computing function slicing, characterized in that: include: Construct a unified computing identity and efficiency constraint equation for the UAV swarm network, and a unified computing identity for nodes in the UAV swarm network, including node identity, link identity, computing power identity, path identity, and service identity; Establish a general computing demand service graph, combine the transmission service identifier, computing service identifier and efficiency constraint equation, use the knowledge graph method to analyze the supply and demand relationship and service constraints between the general computing function node v and the user u, and construct a service supply and demand graph; and Dynamic routing is achieved through slice selection based on the supply and demand map of the general computing service. Slice selection is used to achieve general computing converged routing, and routing is dynamically adjusted to meet the on-demand response of user services. Among them, establishing a general computing demand service map also includes: Define the user node set and the general function node set; Based on user business needs and the service capabilities of the inter-computing function nodes, an interaction matrix is ​​established to represent the supply and demand relationship between users and the inter-computing function nodes; and Use the knowledge graph method to map the supply and demand relationship into triples and build a service supply and demand graph; Among them, the transmission service identifier and computing service identifier in the user u business requirement are represented as T_SID respectively u and C_SID u , the link identifier and computing power identifier of the computing function node v are represented as LID v and CID v , among which LID v and CID v Satisfy the efficiency constraint equation; Among them, establishing a general computing demand service map also includes: The knowledge graph that defines the supply and demand relationship between user u and general computing function node v is: G={([T_SID u ,C_SID u ],e u,v ,[LID v ,CID v ])|u∈U,v∈V,e u,v ∈{0,1}}, Among them, the triple ([T_SID u ,C_SID u ],e,[LID v ,CID v ]) represents the general computing resource of the general computing function node v [LID v ,CID v ]Through the relationship e u,v is the total computing demand of user u [T_SID u ,C_SID u ]Service, Relationship u,v Indicates whether the physical distance between the computing function node and the user service is reachable, e u,v =1 means the physical distance is reachable, e u,v =0 means the physical distance is unreachable; Among them, dynamic routing is achieved based on the general calculation slice selection, including: The general computing function slice serving user u is represented as S u , S u The corresponding general function node set is ψ u , user u in ψ u The steady-state performance function under is expressed as: And it complies with the following constraint equations to meet the user's general calculation requirements: in, 2. The method for dynamic routing of drone cluster networks based on general computing function slicing according to claim 1 is characterized in that: in, The efficiency constraint equation of the UAV swarm network is: in, is the total spectrum of the system; is the maximum transmit power of the node; and Respectively represent the maximum CPU computing frequency and maximum storage energy of the node; W n is the spectrum bandwidth available to node n, P n is the transmit power of node n.

3. The method for dynamic routing of drone cluster networks based on general computing function slicing according to claim 2 is characterized in that: Define the user node set and the general function node set as U and v respectively; Establishing the interaction matrix between user and general computing function nodes Among them, x u,v =1 indicates that user u and the general computing function node v have generated a general computing interaction; otherwise, x u,v =0.

Images