Method and device for determining access network, electronic equipment, storage medium and product

Abstract

The embodiment of the invention provides an access network determination method and device, electronic equipment, a storage medium and a product. Comprising the following steps: sending an edge computing task access request, acquiring network information of a ground cellular network, an unmanned aerial vehicle network and a satellite network according to the edge computing task access request, and determining an information transmission rate and required transmitting power required by a user terminal to be accessed according to the network information of each network, the method comprises the following steps: acquiring current position information of a user terminal to be accessed, determining a target network to which an edge computing task access request is accessed according to an information transmission rate required by the user terminal to be accessed, required transmitting power and the current position information, and accessing the target network, the target network is any one of a ground cellular network, an unmanned aerial vehicle network and a satellite network. According to the method and the device, the to-be-accessed user terminal can access the ground cellular network and can also access the unmanned aerial vehicle network or the satellite network, so that the optimal network is accessed, and the resource allocation efficiency is improved.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a method, device, electronic device, storage medium and product for determining an access network. Background Art

[0002] With the rapid development of wireless communication technology, users' demand for efficient and stable network access is increasing. The network accessed by user terminals will directly affect the user experience and the efficiency of network resource utilization.

[0003] In the related art, a traditional network resource allocation method is usually to allocate access points to user terminals to be accessed based on network resource information of a terrestrial cellular network, wherein the access points may be terrestrial base stations.

[0004] However, the resource allocation efficiency of the above method is low and its applicability is poor. Summary of the invention

[0005] The embodiments of the present application provide a method, device, electronic device, storage medium and product for determining an access network, so as to achieve the technical effect of improving resource allocation efficiency.

[0006] In a first aspect, an embodiment of the present application provides a method for determining an access network, including:

[0007] Send edge computing task access request;

[0008] According to the edge computing task access request, obtain network information of the ground cellular network, the drone network, and the satellite network;

[0009] Determine the information transmission rate and the required transmission power required by the user terminal to be connected according to the network information of each network;

[0010] Acquire the current location information of the user terminal to be accessed;

[0011] Determine a target network for the edge computing task access request according to the information transmission rate required by the user terminal to be accessed, the required transmission power, and the current location information, where the target network is any one of the ground cellular network, the drone network, and the satellite network;

[0012] Access the target network.

[0013] In a possible implementation manner, determining, according to the network information of each network, the information transmission rate and the required transmit power required by the user terminal to be accessed includes:

[0014] Determining an optimized network-side load function according to the network information of each network;

[0015] Determining an optimized user-side load function according to the network information of each network;

[0016] The optimized network-side load function is combined with the optimized user-side load function to obtain a target load function;

[0017] According to the target load function, the information transmission rate and the required transmission power required by the user terminal to be connected are determined.

[0018] In a possible implementation manner, the network information of each network includes: current network resource usage data of each network, and information transmission rate and transmission power of user terminals currently connected to each network;

[0019] The step of determining an optimized network side load function according to the network information of each network includes:

[0020] Acquire an initial network side load function according to the current network resource usage data of each network, and the information transmission rate and transmission power of the user terminal currently connected to each network;

[0021] Obtaining the total network resources of the channels preset by each network and the required information transmission rate preset by the user terminal to be connected;

[0022] The optimized network side load function is determined by taking the total network resources of the channels preset by the networks and the required information transmission rate preset by the user terminal to be accessed as constraint conditions and taking the initial network side load function as the optimization target.

[0023] In a possible implementation manner, the network information of each network includes: the transmission delay, maximum data packet length, and information transmission rate of the user terminal currently connected to each network;

[0024] The step of determining an optimized user-side load function according to the network information of each network includes:

[0025] Obtaining an initial user-side load function according to the transmission delay, maximum data packet length, and information transmission rate of the user terminals currently connected to each network;

[0026] Obtain the total network resources of the channels preset for each network and the required channel network resources preset for the edge computing task;

[0027] Taking the total channel network resources preset by each network and the required channel network resources preset by the edge computing task as constraints, taking the initial user-side load function as the optimization target, the optimized user-side load function is determined.

[0028] In a possible implementation manner, determining, according to the target load function, the information transmission rate and the required transmit power required by the user terminal to be connected includes:

[0029] Obtaining a preset transmit power of the user terminal to be accessed, a preset transmit power of each network, and a preset average calculation delay of each network;

[0030] Inputting the preset transmit power of the user terminal to be accessed, the preset transmit power of each network, the preset average calculation delay of each network and the target load function into a preset optimization algorithm to determine the information transmission rate required by the user terminal to be accessed;

[0031] The information transmission rate required by the user terminal to be accessed, the average calculation delay of each preset network and the target load function are input into a preset optimization algorithm to determine the transmission power required by the user terminal to be accessed.

[0032] In a possible implementation, determining the target network to be accessed by the edge computing task access request according to the information transmission rate required by the user terminal to be accessed, the required transmit power, and the current location information includes:

[0033] The information transmission rate required by the user terminal to be accessed, the required transmission power and the current location information are input into a pre-trained deep reinforcement learning model to obtain the target network accessed by the edge computing task access request.

[0034] In a possible implementation, the training step of the pre-trained deep reinforcement learning model includes:

[0035] Acquire training data, where the training data includes historical networks accessed by the user terminal to be accessed by sending multiple historical edge computing tasks, historical information transmission rates corresponding to accessing each historical network, historical transmission power, and historical location information, where the accessed historical network is a label;

[0036] The training data is used to train the initial deep reinforcement learning model until a preset convergence condition is met to obtain a trained deep reinforcement learning model.

[0037] In a second aspect, an embodiment of the present application provides a device for determining an access network, including:

[0038] A sending module, used to send edge computing task access requests;

[0039] An acquisition module, used to acquire network information of a ground cellular network, a drone network, and a satellite network according to the edge computing task access request;

[0040] A determination module, used to determine the information transmission rate and the required transmission power required by the user terminal to be connected according to the network information of each network;

[0041] The acquisition module is further used to acquire the current location information of the user terminal to be connected;

[0042] The determination module is further used to determine the target network for access by the edge computing task access request according to the information transmission rate required by the user terminal to be accessed, the required transmission power, and the current location information, wherein the target network is any one of the ground cellular network, the drone network, and the satellite network;

[0043] An access module is used to access the target network.

[0044] In a third aspect, an embodiment of the present application provides an electronic device, including: a memory, a processor;

[0045] The memory stores computer-executable instructions;

[0046] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0047] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementations of the first aspect.

[0048] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementation methods of the first aspect.

[0049] The embodiments of the present application provide a method, device, electronic device, storage medium and product for determining an access network, the method comprising: sending an edge computing task access request, and obtaining network information of a ground cellular network, a drone network and a satellite network according to the edge computing task access request. Then, according to the network information of each network, the information transmission rate and the required transmission power required by the user terminal to be accessed are determined, and the current location information of the user terminal to be accessed is obtained. According to the information transmission rate, the required transmission power and the current location information required by the user terminal to be accessed, the target network to be accessed by the edge computing task access request is determined, and the target network is accessed, wherein the target network is any one of a ground cellular network, a drone network and a satellite network. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0051] Figure 1 A schematic diagram of an application scenario provided for this application;

[0052] Figure 2 A flowchart of a method for determining an access network provided in an embodiment of the present application;

[0053] Figure 3 A flowchart of a method for determining an information transmission rate and a required transmission power required by a user terminal to be connected provided in an embodiment of the present application;

[0054] Figure 4 A flowchart of a method for instantiating a method for determining an information transmission rate and a required transmission power required by a user terminal to be connected provided in an embodiment of the present application;

[0055] Figure 5 A flowchart of a method for training a network-side deep reinforcement learning model provided in an embodiment of the present application;

[0056] Figure 6 A schematic diagram of a training method for a deep reinforcement learning model provided in an embodiment of the present application;

[0057] Figure 7 A schematic diagram of the structure of a device for determining an access network provided in an embodiment of the present application;

[0058] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0059] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0060] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0061] In the description of the embodiments of the present application, terms such as "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or component must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be understood as a limitation on the present application.

[0062] In the description of the embodiments of the present application, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0063] In the field of communications, resource allocation is crucial because communication resources (such as spectrum, bandwidth, power, etc.) are usually limited. Resource allocation can optimize resource usage to improve network performance and user experience.

[0064] In the related technology, when a user terminal sends an edge computing task access request, the traditional resource allocation method usually allocates an access point to the user terminal based on the network resource information of the ground cellular network that the user terminal can access and a pre-trained resource allocation model. The access point can be a ground base station, etc.

[0065] However, the above method only considers the ground cellular network and does not take into account other heterogeneous networks, resulting in inefficient resource allocation. It is not suitable for the space-ground integration scenario and has poor applicability.

[0066] The space-ground integration scenario refers to a communication and computing environment that combines ground networks and space networks, where the space network can include drone networks and satellite networks. In the space-ground integration scenario, since the location of user terminals is variable and not limited to the ground, the resource allocation method applicable to cellular networks in the above-mentioned related technologies is difficult to effectively adapt.

[0067] Therefore, in response to the above technical problems in the related technology, the inventors found in the process of research that when the user terminal sends an edge computing task access request, in addition to being able to access the ground cellular network, if it can also access the drone network or satellite network, then the resources of the drone network or satellite network can be fully utilized, thereby greatly improving the efficiency of resource allocation. Specifically, after the user terminal sends an edge computing task access request, it can obtain the network information corresponding to the three networks of the ground cellular network, the drone network and the satellite network, and determine the information transmission rate and the required transmission power required by the user terminal to be accessed according to the network information of each network. The user terminal can also obtain the current location information of the user terminal to be accessed, and then determine the target network accessed by the above edge computing task access request according to the information transmission rate, the required transmission power and the current location information required by the user terminal to be accessed, and finally access the target network, wherein the target network is any one of the ground cellular network, the drone network and the satellite network. Based on this, the present application proposes a method, device, electronic device, storage medium and product for determining an access network.

[0068] In order to facilitate the understanding of this application, Figure 1 An example of an application scenario diagram of the method for determining access network in this application is illustrated, such as Figure 1 As shown, Figure 1 A schematic diagram of an application scenario provided for the present application. The specific application scenario of the present application may include multiple user terminals 01, a ground cellular network 02, a drone network 03 and a satellite network 04. Each user terminal 01 can be wirelessly connected to the ground cellular network 02, the drone network 03 and the satellite network 04 respectively, wherein the user terminal 01 may also be referred to as a mobile station (Mobile Station, MS).

[0069] For any user terminal 01, after sending an edge computing task access request, the user terminal 01 will obtain the network information of the ground cellular network 02, the drone network 03 and the satellite network 03. According to the network information corresponding to the three networks, the information transmission rate and the required transmission power required by the user terminal 01 can be determined. The user terminal 01 can obtain its current location information, which can be a remote suburb, a suburb or a central city, and then the user terminal 01 determines one of the above networks to access the edge computing task access request based on the required information transmission rate, the required transmission power and the current location information.

[0070] In the present application, a core network controller 05 may also be included. The core network controller 05 may be connected to the ground cellular network 02, the drone network 03 and the satellite network 04 by wired communication or wireless communication. The core network controller 05 may obtain communication data of the ground cellular network 02, the drone network 03 and the satellite network 04.

[0071] It can be understood that the present application does not limit the number, type or function of user terminals 01, ground cellular networks 02, drone networks 03 and satellite networks 04, which can be limited according to actual application conditions.

[0072] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0073] See also Figure 2 , Figure 2 A flowchart of a method for determining an access network provided in an embodiment of the present application, the execution subject of the method may be a device for determining an access network, and the device for determining an access network may be implemented by a computer program; it may also be implemented by a medium storing relevant computer programs, such as a USB flash drive and / or an optical disk, or it may also be implemented by a physical device integrated or installed with relevant computer programs, such as a chip or an electronic device, and the electronic device may be a server, a server cluster, a computer, an intelligent terminal, etc. The method may include the following steps:

[0074] S201. Send an edge computing task access request.

[0075] In this embodiment, the execution subject is explained by taking a smart terminal (referred to as a user terminal) as an example. The user terminal can also be called a mobile station. The user terminal sends an edge computing task access request.

[0076] Among them, edge computing tasks refer to the use of edge computing technology to provide various computing and data processing services. Edge computing is a distributed computing architecture that moves computing resources and data storage from traditional centralized data centers to the edge of the network, that is, closer to the data generation source or user terminal.

[0077] Optionally, edge computing tasks can be real-time image and video processing tasks, speech recognition and natural language processing tasks, augmented reality applications, personalized content recommendation tasks, and game optimization tasks, etc.

[0078] S202. Obtain network information of the ground cellular network, the drone network, and the satellite network according to the edge computing task access request.

[0079] After the user terminal sends an edge computing task access request, the edge devices corresponding to the ground cellular network, drone network and satellite network will respond to the edge computing task access request and send their respective network information to the user terminal, so that the user terminal can obtain the network information of the ground cellular network, the network information of the drone network and the network information of the satellite network.

[0080] Among them, the terrestrial cellular network refers to a wireless communication network structure that provides mobile communication services by deploying a large number of base stations (also called cellular towers) on the ground.

[0081] A drone network is a system consisting of multiple drones that are connected to each other through wireless communication technology and work together to provide mobile communication services.

[0082] A satellite network is a system consisting of multiple artificial satellites that provides mobile communication services by forming a widely covered communication network.

[0083] S203: Determine the information transmission rate and the required transmission power required by the user terminal to be connected according to the network information of each network.

[0084] The user terminal can determine the required information transmission rate and the required transmission power based on the network information of the ground cellular network, the network information of the drone network and the network information of the satellite network.

[0085] Optionally, the network information of each network may be various parameters, including but not limited to: the current network resource usage data of each network, the information transmission rate and transmission power of the user terminals currently connected to each network, the transmission delay of the user terminals currently connected to each network, the maximum data packet length and information transmission rate, etc.

[0086] S204: Obtain current location information of the user terminal to be connected.

[0087] Optionally, the user terminal can obtain the current location information through a satellite navigation system, can obtain the current location information through cellular network positioning, can also obtain the current location information through a wireless network Wi-Fi, can also obtain the current location information through sensors installed inside the user terminal, etc.

[0088] It is understandable that the above method for obtaining the current location information is only for illustration and does not limit the present application.

[0089] S205. Determine the target network for the edge computing task access request to access based on the information transmission rate, the required transmission power, and the current location information required by the user terminal to be accessed.

[0090] The user terminal determines the target network for the edge computing task access request based on the required information transmission rate and the required transmission power determined in step S203 and the current location information obtained in step S204, based on a preset optimization algorithm or a pre-trained optimization model.

[0091] The target network is any one of a ground cellular network, a drone network and a satellite network.

[0092] S206: Access the target network.

[0093] After the user terminal determines the target network, it accesses the target network.

[0094] In the above-mentioned embodiment of the present application, an edge computing task access request is sent, and the network information of each network of the ground cellular network, the drone network and the satellite network is obtained according to the edge computing task access request. Then, according to the network information of each network, the information transmission rate and the required transmission power required by the user terminal to be accessed are determined, and the current location information of the user terminal to be accessed is obtained. According to the information transmission rate, the required transmission power and the current location information required by the user terminal to be accessed, the target network to be accessed by the edge computing task access request is determined, and the target network is accessed, wherein the target network is any one of the ground cellular network, the drone network and the satellite network. In the method of this embodiment, the user terminal to be accessed can determine the information transmission rate and the required transmission power required by the user terminal to be accessed according to the network information of each network, and then determine the target network according to the required information transmission rate, the required transmission power and the current location information, so that the user terminal to be accessed can access the ground cellular network in addition to the drone network or the satellite network, thereby accessing the optimal network and improving the efficiency of resource allocation.

[0095] Further, based on the above embodiment, the following embodiment describes a process of determining the information transmission rate and the required transmission power required by the user terminal to be connected according to the network information of each network.

[0096] See also Figure 3 , Figure 3 A flow chart of a method for determining an information transmission rate and a required transmission power required by a user terminal to be connected provided in an embodiment of the present application may include the following steps:

[0097] S301. Determine an optimized network-side load function according to network information of each network.

[0098] Optionally, the network information of each network includes, but is not limited to: current network resource usage data of each network, and information transmission rate and transmission power of user terminals currently connected to each network.

[0099] One possible implementation is:

[0100] The initial network side load function is obtained according to the current network resource usage data of each network, and the information transmission rate and transmission power of the user terminals currently connected to each network.

[0101] The initial network side load function can be shown as formula (1), and the constraints of some parameters in formula (1) are shown in formulas (2) to (4):

[0102]

[0103]

[0104]

[0105]

[0106] in, represents the initial network side load function; Represents the network type matrix, the network types are ground cellular network, drone network and satellite network; Represents the information transmission rate matrix of the user terminals currently connected to each network; Indicates the current network resource usage data of each network, i.e., the current channel orthogonal resource usage data, which may change over time; Indicates channel slices; The total number of resource blocks representing the channel slice resource blocks; Indicates that each network is The first of the channel slices The transmit power matrix of the currently connected user terminal on the resource block may change over time; represents the first system sensitivity factor; represents the second system sensitivity factor; represents the third system sensitivity factor; represents the fourth system sensitivity factor, to It can be adjusted by the user terminal based on the learning of the local data pool; Indicates the total network resources of the channel; Indicates Networks include ground cellular networks, drone networks, and satellite networks; It represents the smallest resource allocation unit in any of the ground cellular networks, drone networks and satellite networks. Indicates the total number of resource allocation units.

[0107] The total network resources of the channels preset by each network and the preset demand information transmission rate of the user terminal to be connected are obtained, and the total network resources of the channels preset by each network and the preset demand information transmission rate of the user terminal to be connected are used as constraints, and the initial network side load function is used as the optimization target to determine the optimized network side load function.

[0108] The optimized network side load function can be expressed as That is, as shown in formula (5), the constraints of some parameters in formula (5) are shown in formulas (6) to (8):

[0109]

[0110]

[0111]

[0112]

[0113] S302: Determine an optimized user-side load function according to network information of each network.

[0114] Optionally, the network information of each network includes, but is not limited to: the transmission delay, maximum data packet length, and information transmission rate of the user terminal currently connected to each network.

[0115] One possible implementation is:

[0116] The initial user-side load function is obtained according to the transmission delay, maximum data packet length and information transmission rate of the user terminals currently connected to each network.

[0117] The initial user side load function can be shown as formula (9), and the constraints of some parameters in formula (9) are shown in formulas (10) to (11):

[0118]

[0119]

[0120]

[0121] in, represents the initial user side load function; Represents the transmission delay matrix of the user terminals currently connected to the ground cellular network, drone network, and satellite network; Represents the current congestion index matrix of each network; Represents the computing power matrix of edge devices corresponding to each network; Represents the maximum data packet length matrix of user terminals currently connected to each network.

[0122] Obtain the total channel network resources preset for each network and the required channel network resources preset for the edge computing task, use the total channel network resources preset for each network and the required channel network resources preset for the edge computing task as constraints, take the initial user-side load function as the optimization target, and determine the optimized user-side load function.

[0123] The optimized user-side load function can be expressed as That is, as shown in formula (12), the constraints of some parameters in formula (12) are shown in formulas (13) to (16):

[0124]

[0125]

[0126]

[0127]

[0128]

[0129] S303: Merge the optimized network-side load function and the optimized user-side load function to obtain a target load function.

[0130] The target load function is obtained by combining the optimized network side load function, i.e., formula (5), with the optimized user side load function, i.e., formula (12). As the formula That is, as shown in formula (17), the constraints of some parameters in formula (17) are shown in formulas (18) to (22):

[0131]

[0132]

[0133]

[0134]

[0135]

[0136]

[0137] S304: Determine the information transmission rate and the required transmission power required by the user terminal to be connected according to the target load function.

[0138] Then, according to the determined target load function, the information transmission rate and the required transmission power required by the user terminal to be connected are determined. The specific implementation method will be described in detail in the following embodiments. Please refer to the following embodiments.

[0139] In the above-mentioned embodiment of the present application, the optimized network side load function is determined according to the network information of each network, and the optimized user side load function is determined according to the network information of each network, the optimized network side load function is merged with the optimized user side load function to obtain the target load function, and the information transmission rate and the required transmission power required by the user terminal to be connected are determined according to the target load function. The method of this embodiment determines the target load function more accurately according to the optimized network side load function and the optimized user side load function, and then the information transmission rate and the required transmission power required by the user terminal to be connected determined based on the accurate target load function are more accurate.

[0140] Further, based on the above embodiment, the following embodiment describes a process of determining the information transmission rate and the required transmission power required by the user terminal to be connected according to the target load function.

[0141] See also Figure 4 , Figure 4 A flow chart of a method for instantiating a method for determining an information transmission rate and a required transmission power required by a user terminal to be connected provided in an embodiment of the present application, the method may include the following steps:

[0142] S401: Obtain a preset transmit power of a user terminal to be connected, a preset transmit power of each network, and a preset average calculation delay of each network.

[0143] Since some parameters and constraints in the target load function are non-convex, it is very difficult to solve. Therefore, in order to reduce the calculation delay, the problem can be converted into solving the information transmission rate sub-problem, solving the transmission power and solving the target network sub-problem.

[0144] For solving the sub-problem of information transmission rate, the transmission power of the user terminal to be accessed and the transmission power of each network are fixed, and the average calculation delay of each network is given, that is, the transmission power of the user terminal to be accessed, the transmission power of each network and the average calculation delay of each network are preset.

[0145] S402: Input the preset transmission power of the user terminal to be connected, the preset transmission power of each network, the preset average calculation delay of each network, and the target load function into a preset optimization algorithm to determine the information transmission rate required by the user terminal to be connected.

[0146] To solve the information transmission rate subproblem, one possible implementation method is:

[0147] Based on the preset optimization algorithm, the preset transmit power of the user terminal to be connected, the preset transmit power of each network, the preset average calculation delay of each network and the target load function are input into the preset optimization algorithm, and the target load function is converted into the following formula: That is, formula (23). The constraints of some parameters in formula (23) are shown in formulas (24) to (26):

[0148]

[0149]

[0150]

[0151]

[0152] in, Indicates the information transmission rate required by the user terminal to be connected; Indicates the number of orthogonal time-frequency resource blocks determined by the user terminal to be connected, and its value can be 0 or 1; Indicated in The first of the channel slices The channel bandwidth of the resource block is determined on the resource blocks; Indicates the total number of channel resources in each network; is the channel capacity per unit spectrum expressed by Shannon’s formula, Indicated in The first of the channel slices The channel state matrix estimated on the resource blocks can change over time; Indicates the preset maximum value of information transmission rate; Indicated in The maximum channel bandwidth preset for each resource block; Indicates the total number of channel slices; Indicates the total amount of resource blocks in the channel slice; represents the average noise power of a preset channel slice; represents the total network resources of the channel of the ith network; Indicates that each network is currently in The first of the channel slices The network resource usage data on each resource block can change over time.

[0153] Thereby, the information transmission rate required by the user terminal to be accessed is obtained.

[0154] S403: Input the information transmission rate required by the user terminal to be connected, the preset average calculation delay of each network, and the target load function into a preset optimization algorithm to determine the transmission power required by the user terminal to be connected.

[0155] For solving the transmission power sub-problem, after determining the information transmission rate required by the user terminal to be connected, the average calculation delay of each network is given.

[0156] One possible way to solve the transmit power subproblem is:

[0157] Based on the preset optimization algorithm, the information transmission rate required by the user terminal to be connected, the preset average calculation delay of each network and the target load function are input into the preset optimization algorithm, and the target load function is converted into the following formula: That is, formula (27). The constraints of some parameters in formula (27) are shown in formula (28):

[0158]

[0159]

[0160] in, Indicates the transmit power required by the user terminal to be connected, Indicates the preset maximum transmit power.

[0161] Thereby, the transmission power required by the user terminal to be accessed is obtained.

[0162] In this embodiment, since the selection of the number of resource blocks and the transmission power is discrete in actual engineering, and the problem after conversion does not have the problem of variable coupling, in addition to using the above method to determine the information transmission rate and the required transmission power required by the user terminal to be connected, the grid search method can also be used, that is, the information transmission rate and the required transmission power required by the possible user terminals to be connected are brought into the target load function one by one until the information transmission rate and transmission power that meet the requirements are found.

[0163] In the above-mentioned embodiment of the present application, by obtaining the preset transmission power of the user terminal to be accessed, the preset transmission power of each network, and the preset average calculation delay of each network, the preset transmission power of the user terminal to be accessed, the preset transmission power of each network, the preset average calculation delay of each network, and the target load function are input into the preset optimization algorithm to determine the information transmission rate required by the user terminal to be accessed, and the information transmission rate required by the user terminal to be accessed, the preset average calculation delay of each network, and the target load function are input into the preset optimization algorithm to determine the transmission power required by the user terminal to be accessed. The method of this embodiment determines the information transmission rate and the required transmission power required by the user terminal to be accessed more accurately through the preset transmission power of the user terminal to be accessed, the preset transmission power of each network, and the preset average calculation delay of each network, reduces the delay required for calculation, and improves calculation efficiency.

[0164] Furthermore, based on the above embodiments, the following embodiments illustrate the process of determining the target network for the edge computing task access request to access based on the information transmission rate, required transmission power and current location information required by the user terminal to be accessed.

[0165] After the information transmission rate and the required transmission power of the user terminal to be connected are determined through the above embodiment, they are substituted into the target load function, and the target load function is converted into the following formula: That is, formula (29). The constraints of some parameters in formula (29) are shown in formulas (30) to (34):

[0166]

[0167]

[0168]

[0169]

[0170]

[0171]

[0172] in, It means that what is stored is the network resource allocation situation at the next moment. The one that changes from 0 to 1 is the target network, which can be any one of the ground cellular network, drone network and satellite network.

[0173] A possible solution to the above formula (33) is:

[0174] The information transmission rate, required transmission power and current location information required by the user terminal to be accessed are input into the pre-trained deep reinforcement learning model to obtain the target network for the edge computing task access request.

[0175] Optionally, the pre-trained deep reinforcement learning model may be a distributed deep reinforcement learning model (Distributed-Deep Q Network, DP-DQN).

[0176] To facilitate understanding of the pre-trained deep reinforcement learning model in this embodiment, the training process is described below. Figure 5 , Figure 5 A flowchart of a method for training a network-side deep reinforcement learning model provided in an embodiment of the present application, the method may include the following steps:

[0177] S501: Obtain training data.

[0178] In this embodiment, the execution subject is a user terminal, which may also be called a mobile station.

[0179] Among them, the training data includes the historical networks accessed by the user terminal to be accessed to send multiple historical edge computing tasks, the historical information transmission rate, historical transmission power and historical location information corresponding to the access to each historical network, and the accessed historical network is the label.

[0180] The historical networks accessed are ground cellular networks, drone networks, or satellite networks.

[0181] S502: Use the training data to train the initial deep reinforcement learning model until a preset convergence condition is met to obtain a trained deep reinforcement learning model.

[0182] The initial deep reinforcement learning model is trained using the training data until the number of training times reaches a preset training times threshold or the loss function reaches a preset loss function threshold, thereby obtaining a trained deep reinforcement learning model.

[0183] The user terminal's training of the initial deep reinforcement learning model is the third training.

[0184] See also Figure 6 , Figure 6 A schematic diagram of a training method for a deep reinforcement learning model provided in an embodiment of the present application is shown below. Figure 5 and Figure 6 Describe the training process of the final trained deep reinforcement learning model.

[0185] Optionally, the initial deep reinforcement learning model can be obtained by training the network devices corresponding to the ground cellular network, the drone network and the satellite network.

[0186] One possible implementation is:

[0187] The edge device corresponding to the ground cellular network trains the first deep reinforcement learning model based on its own network information to obtain the trained first sub-deep reinforcement learning model. Then, the edge device corresponding to the drone network trains the first sub-deep reinforcement learning model based on its own network information to obtain the trained second sub-deep reinforcement learning model. Then, the edge device corresponding to the satellite network trains based on its own network information to obtain the trained third sub-deep reinforcement learning model.

[0188] Among them, the third sub-deep reinforcement learning model can be the initial deep reinforcement learning model in step S502.

[0189] It can be understood that the order of training the edge devices corresponding to the ground cellular network, the edge devices corresponding to the drone network, and the edge devices corresponding to the satellite network is not limited in this application. The training of the first deep reinforcement learning model by the edge devices of each network can be understood as the second model training.

[0190] Optionally, the first deep reinforcement learning model can be trained by the core network controller, and the core network controller obtains the network information of each network by detecting the status of each network. The core network controller uses the preset average network parameters of each network and the historical congestion index of each network as input, and the connected historical network as a label to train the second deep reinforcement learning model until the convergence condition is met to obtain the trained first deep reinforcement learning model, wherein the second deep reinforcement learning model is the original model that has not undergone any training, and the training of the original model by the core network controller can be understood as the first model training.

[0191] Optionally, the trained deep reinforcement learning model deployed on the user terminal in the present application may be a model after the third training, a model after the second training, or a model after the first training.

[0192] After the user terminal obtains the trained deep reinforcement learning model, the required information transmission rate, the required transmission power and the current location information are input into the pre-trained deep reinforcement learning model to obtain the target network for the edge computing task access request.

[0193] The deep reinforcement learning trained in this application has low network overhead and can be applied to situations where the number of user terminal accesses surges in the space-ground integration scenario.

[0194] In the above-mentioned embodiment of the present application, by acquiring training data and using the training data to train the initial deep reinforcement learning model until the preset convergence conditions are met, a trained deep reinforcement learning model is obtained, so that the target network accessed by the edge computing task access request of the user terminal to be accessed obtained according to the trained and accurate deep reinforcement learning model is more accurate, thereby accessing the optimal network and improving resource allocation efficiency.

[0195] Figure 7 A schematic diagram of a structure of a device for determining an access network provided in an embodiment of the present application, such as Figure 7 As shown, the access network determination device provided in this embodiment includes:

[0196] The sending module 701 is used to send an edge computing task access request.

[0197] The acquisition module 702 is used to obtain network information of the ground cellular network, the drone network and the satellite network according to the edge computing task access request.

[0198] The determination module 703 is used to determine the information transmission rate and the required transmission power required by the user terminal to be connected according to the network information of each network.

[0199] The acquisition module 702 is further used to acquire the current location information of the user terminal to be connected.

[0200] The determination module 703 is also used to determine the target network for the edge computing task access request to access based on the information transmission rate, the required transmission power and the current location information required by the user terminal to be accessed. The target network is any one of a ground cellular network, a drone network and a satellite network.

[0201] The access module 704 is used to access the target network.

[0202] In a possible implementation, the determining module 703 is specifically configured to:

[0203] According to the network information of each network, the optimized network side load function is determined.

[0204] According to the network information of each network, an optimized user-side load function is determined.

[0205] The network side load function and the user side load function are combined to obtain the target load function.

[0206] According to the target load function, the information transmission rate and the required transmission power required by the user terminal to be connected are determined.

[0207] In a possible implementation, the network information of each network includes: current network resource usage data of each network, and information transmission rate and transmission power of user terminals currently connected to each network, and the determination module 703 is specifically used to:

[0208] The initial network side load function is obtained according to the current network resource usage data of each network, and the information transmission rate and transmission power of the user terminals currently connected to each network.

[0209] Obtain the total network resources of the channels preset by each network and the required information transmission rate preset by the user terminal to be connected.

[0210] The total network resources of the channels preset in each network and the required information transmission rate preset by the user terminal to be connected are used as constraints, the initial network side load function is used as the optimization target, and the optimized network side load function is determined.

[0211] In a possible implementation, the network information of each network includes: the transmission delay, the maximum data packet length, and the information transmission rate of the user terminal currently connected to each network, and the determination module 703 is specifically used to:

[0212] The initial user-side load function is obtained according to the transmission delay, maximum data packet length and information transmission rate of the user terminals currently connected to each network.

[0213] Obtain the total network resources of the channels preset for each network and the required channel network resources preset for the edge computing task.

[0214] Taking the total channel network resources preset by each network and the required channel network resources preset by the edge computing task as constraints, and the initial user-side load function as the optimization target, the optimized user-side load function is determined.

[0215] In a possible implementation, the determining module 703 is specifically configured to:

[0216] Obtain the preset transmit power of the user terminal to be accessed, the preset transmit power of each network, and the preset average calculation delay of each network.

[0217] The preset transmission power of the user terminal to be accessed, the preset transmission power of each network, the preset average calculation delay of each network and the target load function are input into the preset optimization algorithm to determine the information transmission rate required by the user terminal to be accessed.

[0218] The information transmission rate required by the user terminal to be connected, the preset average calculation delay of each network and the target load function are input into the preset optimization algorithm to determine the transmission power required by the user terminal to be connected.

[0219] In a possible implementation, the determining module 703 is specifically configured to:

[0220] The information transmission rate, required transmission power and current location information required by the user terminal to be accessed are input into the pre-trained deep reinforcement learning model to obtain the target network for the edge computing task access request.

[0221] In a possible implementation, a training module is further included, which is used to:

[0222] Obtain training data, which includes historical networks accessed by the user terminal to be accessed to send multiple historical edge computing tasks, historical information transmission rates corresponding to accessing each historical network, historical transmission power, and historical location information, and the accessed historical network is a label.

[0223] The initial deep reinforcement learning model is trained using the training data until the preset convergence conditions are met to obtain a trained deep reinforcement learning model.

[0224] The device for determining an access network provided in this embodiment can execute the method provided in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be described in detail here.

[0225] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 8 As shown, the electronic device provided in this embodiment includes: at least one processor 801 and a memory 802. Optionally, the electronic device further includes a communication component 803. The processor 801, the memory 802 and the communication component 803 are connected via a bus 804.

[0226] In a specific implementation process, at least one processor 801 executes the computer-executable instructions stored in the memory 802, so that at least one processor 801 executes the above method.

[0227] The specific implementation process of the processor 801 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0228] In the above embodiments, it should be understood that the processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the invention can be directly implemented as a hardware processor, or can be implemented by a combination of hardware and software modules in the processor.

[0229] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (NVM), such as at least one disk storage.

[0230] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of this application is not limited to only one bus or one type of bus.

[0231] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0232] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0233] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0234] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0235] The division of units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0236] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0237] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0238] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0239] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes.

[0240] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art not disclosed by the present invention, are not limited to the precise structure described above and shown in the drawings, and may be modified and changed in various ways without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for determining an access network, characterized in that: include: Send edge computing task access request; According to the edge computing task access request, obtain network information of the ground cellular network, the drone network, and the satellite network; Determine the information transmission rate and the required transmission power required by the user terminal to be connected according to the network information of each network; Acquire the current location information of the user terminal to be accessed; Determine a target network for the edge computing task access request according to the information transmission rate required by the user terminal to be accessed, the required transmission power, and the current location information, where the target network is any one of the ground cellular network, the drone network, and the satellite network; Access the target network.

2. The method according to claim 1, characterized in that The determining, according to the network information of each network, the information transmission rate and the required transmission power required by the user terminal to be accessed includes: Determining an optimized network-side load function according to the network information of each network; Determining an optimized user-side load function according to the network information of each network; The optimized network-side load function is combined with the optimized user-side load function to obtain a target load function; According to the target load function, the information transmission rate and the required transmission power required by the user terminal to be connected are determined.

3. The method according to claim 2, characterized in that The network information of each network includes: current network resource usage data of each network, and information transmission rate and transmission power of user terminals currently connected to each network; The step of determining an optimized network side load function according to the network information of each network includes: Acquire an initial network side load function according to the current network resource usage data of each network, and the information transmission rate and transmission power of the user terminal currently connected to each network; Obtaining the total network resources of the channels preset by each network and the required information transmission rate preset by the user terminal to be connected; The optimized network side load function is determined by taking the total network resources of the channels preset by the networks and the required information transmission rate preset by the user terminal to be accessed as constraint conditions and taking the initial network side load function as the optimization target.

4. The method according to claim 2, characterized in that: The network information of each network includes: the transmission delay, maximum data packet length and information transmission rate of the user terminal currently connected to each network; The step of determining an optimized user-side load function according to the network information of each network includes: Obtaining an initial user-side load function according to the transmission delay, maximum data packet length, and information transmission rate of the user terminals currently connected to each network; Obtain the total network resources of the channels preset for each network and the required channel network resources preset for the edge computing task; Taking the total channel network resources preset by each network and the required channel network resources preset by the edge computing task as constraints, taking the initial user-side load function as the optimization target, the optimized user-side load function is determined.

5. The method according to claim 3 or 4, characterized in that: The determining, according to the target load function, the information transmission rate and the required transmission power required by the user terminal to be connected includes: Obtaining a preset transmit power of the user terminal to be accessed, a preset transmit power of each network, and a preset average calculation delay of each network; Inputting the preset transmit power of the user terminal to be accessed, the preset transmit power of each network, the preset average calculation delay of each network and the target load function into a preset optimization algorithm to determine the information transmission rate required by the user terminal to be accessed; The information transmission rate required by the user terminal to be accessed, the average calculation delay of each preset network and the target load function are input into a preset optimization algorithm to determine the transmission power required by the user terminal to be accessed.

6. The method according to claim 1, characterized in that The determining, according to the information transmission rate required by the user terminal to be accessed, the required transmission power, and the current location information, a target network to be accessed by the edge computing task access request includes: The information transmission rate required by the user terminal to be accessed, the required transmission power and the current location information are input into a pre-trained deep reinforcement learning model to obtain the target network accessed by the edge computing task access request.

7. The method according to claim 6, characterized in that The training steps of the pre-trained deep reinforcement learning model include: Acquire training data, where the training data includes historical networks accessed by the user terminal to be accessed by sending multiple historical edge computing tasks, historical information transmission rates corresponding to accessing each historical network, historical transmission power, and historical location information, where the accessed historical network is a label; The training data is used to train the initial deep reinforcement learning model until a preset convergence condition is met to obtain a trained deep reinforcement learning model.

8. A device for determining access to a network, characterized in that: include: A sending module, used to send edge computing task access requests; An acquisition module, used to acquire network information of a ground cellular network, a drone network, and a satellite network according to the edge computing task access request; A determination module, used to determine the information transmission rate and the required transmission power required by the user terminal to be connected according to the network information of each network; The acquisition module is further used to acquire the current location information of the user terminal to be connected; The determination module is further used to determine the target network for access by the edge computing task access request according to the information transmission rate required by the user terminal to be accessed, the required transmission power, and the current location information, wherein the target network is any one of the ground cellular network, the drone network, and the satellite network; An access module is used to access the target network.

9. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

11. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 7 when being executed by a processor.

Images