Service data transmission method and device, electronic equipment and storage medium

Abstract

The invention provides a service data transmission method, and relates to the technical field of communication, in particular to a service data transmission method and device, electronic equipment and a storage medium. According to the specific implementation scheme, the method comprises the steps of determining an unloading position of service data based on current to-be-transmitted service data, determining a first service data unloading proportion and a first bandwidth resource allocation proportion of unloading the service data to a relay end, and determining a second service data unloading proportion and a second bandwidth resource allocation proportion of unloading the service data to a base station end, the unloading position comprises at least one of a base station end and a relay end; and based on the determined unloading position, sending the service data corresponding to the first service data unloading proportion to the relay end through the terminal bandwidth resource corresponding to the first bandwidth resource allocation proportion, and sending the service data corresponding to the second service data unloading proportion to the relay end through the terminal bandwidth resource corresponding to the first bandwidth resource allocation proportion. And sending the terminal bandwidth resource corresponding to the second bandwidth resource allocation proportion to the base station end.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technologies, and in particular, to a method, apparatus, electronic device, and storage medium for transmitting service data. Background Art

[0002] In a communication scenario, the service data of user terminals has increased exponentially and is difficult to be processed quickly. Therefore, computing services need to be offloaded to devices other than user terminals for processing, and it is difficult to determine the bandwidth resource allocation for transmitting service data to devices and the proportion of offloaded service data, resulting in problems such as long transmission time and congestion in the process of transmitting service data. Summary of the Invention

[0003] The present disclosure provides a method, apparatus, electronic device, and storage medium for service data for solving at least one of the above technical problems.

[0004] According to one aspect of the present disclosure, there is provided a method for transmitting service data, which is applied to a user terminal and includes:

[0005] Based on the currently to-be-transmitted service data, determine the offloading location of the service data, and determine a first service data offloading ratio and a first bandwidth resource allocation ratio for offloading the service data to the relay end, and a second service data offloading ratio and a second bandwidth resource allocation ratio for offloading the service data to the base station end, where the offloading location includes at least one of the base station end and the relay end;

[0006] Based on the determined offloading location, send the service data corresponding to the first service data offloading ratio to the relay end through the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio, and send the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio;

[0007] Wherein, the relay end is configured to receive and process the service data corresponding to the first service data offloading ratio, and the base station end is configured to receive and process the service data corresponding to the second service data offloading ratio.

[0008] In some examples, the step of based on the currently to-be-transmitted service data, determine the offloading location of the service data, and determine a first service data offloading ratio and a first bandwidth resource allocation ratio for offloading the service data to the relay end, and a second service data offloading ratio and a second bandwidth resource allocation ratio for offloading the service data to the base station end, includes:

[0009] According to the current communication state of the user terminal, obtain the predicted transmission action of the user terminal based on the reinforcement learning model;

[0010] Wherein, the user terminal sends the service data according to the transmission action of the user terminal; the transmission action of the user terminal at least includes: the offloading location of the service data, the first service data offloading ratio, the first bandwidth resource allocation ratio, the second service data offloading ratio, and the second bandwidth resource allocation ratio.

[0011] In some examples, the communication state of the user terminal at least includes: the data volume of the service data, the remaining terminal bandwidth resources currently, and the number of successful transmissions of the service data.

[0012] In some examples, the offloading location is determined in the following manner: determining the offloading location based on the ε-greedy policy.

[0013] In some examples, the reinforcement learning model is trained in the following manner:

[0014] According to the current communication state of the user terminal, based on the reinforcement learning model to be trained, obtain the predicted transmission action of the user terminal;

[0015] Execute the transmission action of the user terminal to obtain a new communication state of the user terminal and a reward parameter;

[0016] Define a target network based on the offloading location in the transmission action of the user terminal, the current communication state of the user terminal, the new communication state of the user terminal, and the reward parameter to obtain a target vector;

[0017] Calculate a stochastic gradient based on the target vector, the current communication state of the user terminal, and the offloading location in the transmission action of the user terminal, and update the reinforcement learning model according to the calculated stochastic gradient.

[0018] In some examples, the relay end is connected to a first edge computing server; the base station end is connected to a second edge computing server;

[0019] The relay end is configured to receive the service data corresponding to the first service data offloading ratio, and send the service data corresponding to the first service data offloading ratio to the first edge computing server for processing, and, the base station end is configured to receive the service data corresponding to the second service data offloading ratio, and send the service data corresponding to the second service data offloading ratio to the second edge computing server for processing.

[0020] According to one aspect of the present disclosure, there is provided a method for transmitting service data, which is applied to a relay end and includes:

[0021] Receive and process the service data corresponding to the first service data offloading ratio sent by the user terminal through the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio, where the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio and the service data corresponding to the first service data offloading ratio are determined by the above method;

[0022] Wherein, the user terminal sends the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio.

[0023] In some examples, the relay end receives the service data corresponding to the first service data offloading ratio sent by the user terminal by using any in-band full-duplex communication method.

[0024] In some examples, the relay end is connected to a relay auxiliary end, and the transmission rate from the user terminal to the relay end is equal to the transmission rate from the relay end to the relay auxiliary end.

[0025] In some examples, the relay auxiliary end is connected to the base station end;

[0026] The user terminal sending the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio includes:

[0027] The user terminal sends the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio via the link of the user terminal - the relay end - the relay auxiliary end - the base station end.

[0028] In some examples, in the link of the user terminal - the relay end - the relay auxiliary end - the base station end, the sub-link of the user terminal - the relay end - the relay auxiliary end transmits service data in the millimeter wave manner; the sub-link of the relay auxiliary end - the base station end transmits service data in the radio frequency band manner.

[0029] In some examples, in the sub-link of the relay auxiliary end - the base station end, the ratios of the relay auxiliary bandwidth resources allocated to each user terminal are the same and the transmission powers are the same.

[0030] In some examples, the relay end is connected to a first edge computing server;

[0031] The receiving and processing the service data corresponding to the first service data offloading ratio sent by the user terminal through the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio includes:

[0032] Receive the service data corresponding to the first service data offloading ratio, and send the service data corresponding to the first service data offloading ratio to the first edge computing server for processing.

[0033] According to one aspect of the present disclosure, a transmission device for service data is provided, including:

[0034] A determination module, configured to determine the offloading location of the service data based on the currently to-be-transmitted service data, and determine the first service data offloading ratio and the first bandwidth resource allocation ratio for offloading the service data to the relay end, and, determine the second service data offloading ratio and the second bandwidth resource allocation ratio for offloading to the base station end, where the offloading location includes at least one of the base station end and the relay end;

[0035] A sending module, configured to, based on the determined offloading location, send the service data corresponding to the first service data offloading ratio to the relay end through the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio, and, send the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio;

[0036] Wherein, the relay end is configured to receive and process the service data corresponding to the first service data offloading ratio, and, the base station end is configured to receive and process the service data corresponding to the second service data offloading ratio.

[0037] In some examples, the determination module is specifically configured to:

[0038] Based on the current communication state of the user terminal and a reinforcement learning model, obtain the predicted transmission action of the user terminal;

[0039] Wherein, the user terminal sends the service data according to the transmission action of the user terminal; the transmission action of the user terminal at least includes: the offloading location of the service data, the first service data offloading ratio, the first bandwidth resource allocation ratio, the second service data offloading ratio, and the second bandwidth resource allocation ratio.

[0040] In some examples, the communication state of the user terminal at least includes: the data volume of the service data, the currently remaining terminal bandwidth resource, and the number of successful transmissions of the service data.

[0041] In some examples, the offloading location in the determination module is determined in the following manner: determining the offloading location based on the ε-greedy policy.

[0042] In some examples, the reinforcement learning model in the determination module is trained in the following manner:

[0043] Based on the current communication state of the user terminal, a predicted transmission action of the user terminal is obtained based on the reinforcement learning model to be trained.

[0044] Execute the transmission action of the user terminal to obtain a new communication state of the user terminal and a reward parameter.

[0045] Define a target network based on the offloading location in the transmission action of the user terminal, the current communication state of the user terminal, the new communication state of the user terminal, and the reward parameter to obtain a target vector.

[0046] Calculate a stochastic gradient based on the target vector, the current communication state of the user terminal, and the offloading location in the transmission action of the user terminal, and update the reinforcement learning model according to the calculated stochastic gradient.

[0047] In some examples, the relay end is connected to a first edge computing server; the base station end is connected to a second edge computing server;

[0048] The relay end is configured to receive the service data corresponding to the first service data offloading ratio, and send the service data corresponding to the first service data offloading ratio to the first edge computing server for processing, and the base station end is configured to receive the service data corresponding to the second service data offloading ratio, and send the service data corresponding to the second service data offloading ratio to the second edge computing server for processing.

[0049] According to one aspect of the present disclosure, a transmission device for service data is provided, including:

[0050] A receiving module, configured to receive and process the service data corresponding to the first service data offloading ratio sent by the user terminal through the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio, wherein the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio and the service data corresponding to the first service data offloading ratio are determined by the above method;

[0051] Wherein, the user terminal sends the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio.

[0052] In some examples, the receiving module receives the service data corresponding to the first service data offloading ratio sent by the user terminal by using an in-band full-duplex communication method.

[0053] In some examples, the receiving module is specifically configured to: receive the service data corresponding to the first service data offloading ratio, and send the service data corresponding to the first service data offloading ratio to the first edge computing server for processing.

[0054] According to another aspect of the present disclosure, there is provided an electronic device, including:

[0055] at least one processor; and

[0056] a memory communicatively connected to the at least one processor; wherein,

[0057] the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above method.

[0058] According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the above method.

[0059] According to another aspect of the present disclosure, there is provided a computer program product, including a computer program, which implements the above method when executed by a processor.

[0060] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The drawings are used to better understand the solution and do not constitute a limitation to the present disclosure. Among them:

[0062] Figure 1 is a schematic flowchart of a method for transmitting service data provided in the first embodiment of the present disclosure;

[0063] Figure 2 is a schematic flowchart of a method for transmitting service data provided in the second embodiment of the present disclosure;

[0064] Figure 3 is a comparison chart of system throughput in the scenario of introducing semi-following UAV;

[0065] Figure 4 is a graph showing the variation of dual-band and high-band transmission throughput with distance in the scenario of introducing UAV;

[0066] Figure 5 is a schematic diagram of the convergence process of minimizing the energy consumption of the system during the training of the reinforcement learning model;

[0067] Figure 6 is a schematic diagram of the user completion rate during the training of the reinforcement learning model;

[0068] Figure 7It is a schematic structural diagram of a service data transmission device provided by the third embodiment of the present disclosure;

[0069] Figure 8 It is a schematic structural diagram of a service data transmission device provided by the fourth embodiment of the present disclosure;

[0070] Figure 9 It is a block diagram of an electronic device for implementing the embodiments of the present disclosure. Detailed implementation manners

[0071] The following describes exemplary embodiments of the present disclosure with reference to the accompanying drawings. Various details of the embodiments of the present disclosure are included to assist in understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted below.

[0072] Without conflict, the embodiments of the present disclosure and the features in the embodiments can be combined with each other.

[0073] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0074] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a" and "the" are also intended to include the plural forms unless the context clearly indicates otherwise.

[0075] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art. It will also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present disclosure, and will not be interpreted as having an idealized or overly formal meaning unless the present invention is clearly so defined.

[0076] The service data transmission method according to the present disclosure can be executed by an electronic device such as a terminal device or a server. The terminal device can be an in-vehicle device, a user equipment (UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, an in-vehicle device, a wearable device, etc. The method can be implemented by a processor calling computer-readable program instructions stored in a memory. Alternatively, the service data transmission method provided by the present disclosure can be executed by a server.

[0077] In the communication scenario to which the present disclosure is applied, the communication system includes three parties: a user terminal, a relay terminal, and a base station terminal. In this case, the communication link is the user terminal-relay terminal-base station terminal. The user terminal sends part of the service data to the relay terminal through the user terminal-relay terminal link, and sends part of the service data to the base station terminal through the user terminal-relay terminal-base station terminal link. In some communication scenarios with high-speed movement, such as the communication scenario on a high-speed train, a relay auxiliary terminal is added to the communication system. The relay auxiliary terminal is connected to the relay terminal and the base station terminal. At this time, the communication link is the user terminal-relay terminal-relay auxiliary terminal-base station terminal. The user terminal sends part of the service data to the relay terminal through the user terminal-relay terminal link, and sends part of the service data to the base station terminal through the user terminal-relay terminal-relay auxiliary terminal-base station terminal link. Among them, the relay terminal can specifically include various types, for example, it can be a mobile relay (MR); the relay auxiliary terminal can include various types, for example, it can be an unmanned aerial vehicle (UAV), which is not limited here. For the convenience of description, hereinafter, the user terminal is abbreviated as "UE", the base station terminal is abbreviated as "BS", the relay auxiliary terminal is abbreviated as "UAV", and the relay terminal is abbreviated as "MR".

[0078] The following further describes the present disclosure in conjunction with the accompanying drawings and specific embodiments, but it is not a limitation of the present disclosure.

[0079] In the first embodiment of the disclosure, refer to Figure 1 , Figure 1 which shows a flowchart of a method for transmitting service data provided by the first embodiment of the present disclosure. This method is applied to a user terminal and includes the following steps:

[0080] S101. Based on the currently to-be-transmitted service data, determine the offloading location of the service data, and determine the first service data offloading ratio and the first bandwidth resource allocation ratio for offloading the service data to the relay terminal, and, determine the second service data offloading ratio and the second bandwidth resource allocation ratio for offloading to the base station terminal.

[0081] Among them, the offloading location includes at least one of the base station terminal and the relay terminal.

[0082] S102. Based on the determined offloading location, send the service data corresponding to the first service data offloading ratio through the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio to the relay terminal, and, send the service data corresponding to the second service data offloading ratio through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio to the base station terminal.

[0083] Among them, the relay terminal is configured to receive and process the service data corresponding to the first service data offloading ratio, and, the base station terminal is configured to receive and process the service data corresponding to the second service data offloading ratio.

[0084] It should be noted that the user terminal usually offloads service data to one of the relay end or the base station end. In some scenarios, it can be offloaded to both the relay end and the base station end. If it is offloaded to one of the relay end or the base station end, the corresponding service data at the other end is empty; correspondingly, the service data offloading ratio and the bandwidth resource allocation ratio are empty.

[0085] In the method provided by the present disclosure, the user terminal can determine the offloading location to be offloaded, that is, determine to offload service data to at least one of the base station end or the relay end. Then, based on the determined offloading location, determine how much service data to offload to the base station end and / or the relay end, and determine the terminal bandwidth resources occupied by the service data sent to the base station end and the terminal bandwidth resources occupied by the service data sent to the relay end according to the terminal bandwidth resources it has. Among them, if the offloading location only includes one, the corresponding service data offloading ratio or bandwidth resource allocation ratio at the other end is empty; in this way, by receiving service data at both ends, the service data can be processed quickly to reduce the service data processing time, and the service data and terminal bandwidth resources can be accurately allocated to reduce congestion.

[0086] In some examples, S101 can be processed by a deep learning model (such as a reinforcement learning model). Based on this, S101 specifically includes:

[0087] According to the current communication state of the user terminal, based on the reinforcement learning model, obtain the predicted transmission action of the user terminal.

[0088] Among them, the user terminal sends service data according to the transmission action of the user terminal; the transmission action of the user terminal at least includes: the offloading location of the service data, the first service data offloading ratio, the first bandwidth resource allocation ratio, the second service data offloading ratio, and the second bandwidth resource allocation ratio.

[0089] In some examples, the reinforcement learning model can include DQN (Deep Q-Network), SAC (Soft Actor-Critic), PPO (Proximal Policy Optimization), etc.

[0090] In some examples, the communication state of the user terminal at least includes: the data volume of the service data, the currently remaining terminal bandwidth resources, and the number of successful transmissions of the service data.

[0091] In the reinforcement learning model, the interaction process between the agent (user terminal) and the environment is as follows: the agent obtains a state S from the environment tAfter (the communication state of the user terminal), the best action (the transmission action of the user terminal) is selected according to the policy, and this action is executed in the environment to generate a new state S t+1 and a reward r t . The above is a cycle. The process of reinforcement learning is to repeat this cycle, enabling the agent to continuously optimize its policy, and ultimately learn the optimal policy to maximize the cumulative reward. Based on this, S101 adopts a reinforcement learning model to decide the transmission action of the user terminal that matches the communication state of the user terminal.

[0092] According to the above, the state space of the reinforcement model (i.e., the communication state of the user terminal) can include: the data volume of service data, the remaining terminal bandwidth resources currently, the number of successfully transmitted service data, etc., and can also include the computing resources required to calculate a data bit, latency constraints, etc., which are not limited here.

[0093] The action space (i.e., the transmission action of the user terminal) can include: the offloading location of service data, the first service data offloading ratio, the first bandwidth resource allocation ratio, the second service data offloading ratio, and the second bandwidth resource allocation ratio, etc., which are not limited here.

[0094] The reward parameter can be: the sum of the negative value of energy consumption and a preset fixed parameter.

[0095] In some examples, the offloading location is determined in the following way: the offloading location is determined based on the ε-greedy strategy.

[0096] Among them, determining the offloading location based on the ε-greedy strategy specifically includes:

[0097] Select a random action with probability ε, and select the action that maximizes the Q value (the value of the Q value function) with probability 1 - ε, that is, k t = argmax k∈ K] Q(s t , k, x k ; ω t ).

[0098] Among them, k represents the discrete action of the offloading location (i.e., the offloading action), the subscript t represents the current round of reinforcement learning, S t represents the communication state of the user terminal, X k represents the total user terminal transmission action parameter; ω t represents the network weight of the Q value function.

[0099] In some examples, the reinforcement learning model is trained in the following way:

[0100] ​Step 1: Based on the current communication status of the user terminal and the reinforcement learning model to be trained, obtain the predicted transmission action of the user terminal.

[0101] Before Step 1, it is necessary to initialize the network weights of the user terminal transmission action and the Q-value function.

[0102] Based on the current communication status of the user terminal (the data volume of service data, the remaining terminal bandwidth resources currently, the number of successful transmissions of service data), and based on the reinforcement learning model to be trained, obtain the predicted transmission action of the user terminal, specifically the total user terminal transmission action parameter x t , where the unloading action a corresponding to the unloading location included in the user terminal transmission action t , is determined by the ε-greedy policy, and the specific process is the same as above.

[0103] Step 2: Execute the user terminal transmission action to obtain a new user terminal communication status and reward parameters.

[0104] Execute the user terminal transmission action including the unloading action a t to obtain a new user terminal communication status s t+1 and reward parameter r t .

[0105] Actually, here, the unloading location a in the user terminal transmission action t , the current user terminal communication status S t , the new user terminal communication status s t+1 and reward parameter r t (collectively referred to as training data) are stored in the buffer D. The buffer D contains training data at multiple times in the country area. Sample part of the training data from these training data to form a subset B. The subset B is expressed as: {s b , a b , r b , s b+1} b∈B .

[0106] Step 3: Define a target network based on the unloading location, the current user terminal communication status, the new user terminal communication status, and the reward parameter in the user terminal transmission action to obtain a target vector.

[0107] Specifically, the target vector is y b , when s b+1 is the termination state, y b = r b ; when s b+1 is not the termination state, y b = r b + max k∈K γQ(s b+1, k, x k (s b+1 , θ t ); ω t ), where γ is a preset learning rate.

[0108] Step 4: Calculate the stochastic gradient based on the target vector, the current communication state of the user terminal, and the offloading location in the user terminal transmission action, and update the reinforcement learning model according to the calculated stochastic gradient to obtain a trained reinforcement learning model.

[0109] Specifically, use the difference between the predicted Q value and the target Q value (the value of the Q value function) as the loss function to update the Q network. Specifically, use {y b , s b , a b} to calculate the stochastic gradient, and adopt stochastic gradient descent to update the reinforcement learning model to obtain a trained reinforcement learning model.

[0110] In this way, through the deep reinforcement learning method of the hybrid action space, the discrete policy of the offloading location and the continuous variables of the offloading ratio / bandwidth allocation ratio can be directly jointly learned during the learning process, obtaining a better convergence result, thereby determining a more accurate offloading location decision, the offloading ratio of service data, and the bandwidth allocation ratio to achieve higher-quality service data transmission.

[0111] In some examples, the relay end is connected to the first Mobile Edge Computing (MEC) server; the base station end is connected to the second MEC server.

[0112] In S102, the relay end is configured to receive the service data corresponding to the first service data offloading ratio, and send the service data corresponding to the first service data offloading ratio to the first MEC server for processing, and the base station end is configured to receive the service data corresponding to the second service data offloading ratio, and send the service data corresponding to the second service data offloading ratio to the second MEC server for processing.

[0113] In some communication scenarios, the service data grows exponentially, resulting in task queue congestion and extended waiting times. Among them, mobile edge computing, as a key enabling technology for 5G and 6G, can offload the computing tasks of resource-constrained devices to the resource-rich proximal cloud environment for processing, thereby solving the problem of insufficient computing resource capabilities of mobile devices and reducing the pressure on the core network. In this disclosure, edge computing is implemented by setting up MEC at both ends. The service data can be offloaded to both ends (the relay end and the base station end) for edge computing, thereby minimizing the service waiting time and reducing queue congestion.

[0114] Based on the above, for the high-speed rail communication scenario, a system model is established by adding a semi-following UAV as a relay auxiliary end-to-end, as follows:

[0115] Assume that the UAV moves in the same direction as the high-speed rail, and the horizontal flight speed of the UAV is V UAV , and the horizontal speed of the high-speed rail is V HSR , and the relative speed ΔV between the two is V HSR - V UAV . The user terminal is associated with the MR of this carriage. Each MR serves the user terminals corresponding to M service data, and the corresponding user terminal set is M = {1, 2,..., i,..., M}. Assume that there are N user terminals selected to be offloaded to the second edge computing server at the BS end, and the corresponding user terminal set is N = {1, 2,..., j,..., N}. The offloading path of the service data of the user terminal is UE - MR or UE - MR - UAV - BS. At the same time, the dual-band design adopted by the UAV includes two parts of bandwidth. The millimeter wave method in the high-frequency band is used for communication in the UE - MR - UAV sub-link, and the relay-assisted bandwidth resource is expressed as W UE , and the sub-6GHz band is used for communication in the UAV - BS sub-link, and the relay-assisted bandwidth resource is W BS . The relay-assisted bandwidth resources of these two parts are allocated proportionally. The proportion of the relay-assisted bandwidth resource allocated to user terminal m i is α i , and the proportion of the relay-assisted bandwidth resource allocated to user terminal n j is n j .

[0116] In addition, for user terminal m corresponding to the service data i , design parameters (d i , c i , γ i , τ i ), d i represents the data volume of the service data of user terminal m i , c i is the computing resource required to process one data bit (in units of core processor cycles). γ i ∈ [0, 1] represents the proportion of the service data of user terminal m i that needs to be offloaded, and τi represents the delay constraint. Then, the part of (1 - γ i )·d i represents the part of the service data locally processed by user terminal m i . Further, introduce a set of binary variables representing the selection of the remote offloading location When When it is, it means that part of the service data is offloaded to the first edge computing server at the MR side. When it is, it means that part of the service data is offloaded to the second edge computing server at the BS side. For the system energy consumption, the weighted sum of the communication and computing energy consumption of the user terminal, the MR side, the UAV side, and the BS side is used to represent the system energy consumption.

[0117] In the disclosed second embodiment, refer to Figure 2 , Figure 2 which shows a flowchart of a method for transmitting service data provided by the second embodiment of the present disclosure. This method is applied at the relay side and includes the following steps:

[0118] S201. Receive and process the service data corresponding to the first service data offloading ratio sent by the user terminal through the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio.

[0119] Among them, the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio and the service data corresponding to the first service data offloading ratio can be determined by the method of S101 - S102.

[0120] Among them, based on S101 and S102, the user terminal sends the service data corresponding to the second service data offloading ratio to the base station side through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio.

[0121] In some examples, the relay side is connected to the first edge computing server; the base station side is connected to the second edge computing server. Based on this, S201 specifically includes:

[0122] Receive the service data corresponding to the first service data offloading ratio, and send the service data corresponding to the first service data offloading ratio to the first edge computing server for processing.

[0123] And, based on this, the base station side is configured to receive the service data corresponding to the second service data offloading ratio, and send the service data corresponding to the second service data offloading ratio to the second edge computing server for processing.

[0124] In some communication scenarios, the service data grows exponentially, resulting in congestion in the task queue and an extended waiting time. Among them, mobile edge computing, as a key enabling technology for 5G and 6G, can offload the computing tasks of resource - constrained devices to the resource - rich proximal cloud environment for processing, thereby solving the problem of insufficient computing resource capabilities of mobile devices and reducing the pressure on the core network. In the present disclosure, the edge computing is implemented by adopting the method of setting MEC at both ends. The service data can be offloaded to both ends (the relay side and the base station side) for edge computing, so as to minimize the service waiting time and reduce queue congestion.

[0125] In some examples, the relay end receives the service data corresponding to the first service data offloading ratio sent by the user terminal by adopting the full-duplex (FD) communication mode within any sub-band. That is to say, the relay end and the user terminal communicate by adopting the FD full-duplex communication mode, so that the communication efficiency is increased.

[0126] In some examples, a UAV is introduced as a relay auxiliary end and is connected to the relay end and the base station end. For example, in the high-speed rail communication scenario, the UAV semi-follows the high-speed rail at two-thirds of the high-speed rail running speed and joins the communication link as a relay auxiliary end, thereby increasing the communication coverage rate of the relay end.

[0127] In this case, the transmission rate from the user terminal to the relay end (i.e., the UE-MR link) is equal to the transmission rate from the relay end to the relay auxiliary end (i.e., the MR-UAV link). The transmission rate is the communication rate for transmitting service data. Since the UE-MR and MR-UAV links share one channel, and the transmission rate of the MR takes the minimum value of multiple transmission rates, by obtaining the monotonicity of the energy consumption expression of the MR end, it is known that the energy consumption of the MR end is monotonically increasing with respect to the transmit power. Therefore, the transmission rate of the MR-UAV is set equal to the transmission rate of the UE-MR, thereby determining the transmit power of the MR end.

[0128] In this way, the energy consumption of the entire communication system is reduced, and from the perspective of reinforcement learning, the successful results of reinforcement learning are increased, which is more convenient for reinforcement learning to learn reasonable offloading strategies and resource allocation strategies.

[0129] Further, in some examples, the relay auxiliary end is connected to the base station end. In S102, the user terminal sends the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio, including:

[0130] The user terminal sends the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio via the link of user terminal-relay end-relay auxiliary end-base station end (i.e., the UE-MR-UAV-BS link).

[0131] Based on the above, in some examples, in the link of user terminal-relay end-relay auxiliary end-base station end (i.e., the UE-MR-UAV-BS link), the sub-link of user terminal-relay end-relay auxiliary end (i.e., the UE-MR-UAV sub-link) transmits service data in the form of millimeter wave (mmWave); the sub-link of relay auxiliary end-base station end (i.e., the UAV-BS sub-link) transmits service data in the form of radio frequency band (i.e., SUB-6Ghz).

[0132] In this way, the relay assistant uses the dual-band transmission method. For short-distance transmission to the UE side and large-information-volume transmission, mmWave technology is used for communication. The core of mmWave technology is to use high-frequency bands from dozens of GHz to hundreds of GHz as the carrier of wireless communication, providing a larger spectral bandwidth compared to traditional low-frequency communication to support the transmission of a large amount of data. Moreover, for long-distance transmission to the BS side and small-information-volume transmission, SUB-6Ghz technology is used for communication. The dual-band transmission method expands the mmWave communication range and improves the spectral efficiency of the system. Furthermore, the UAV semi-following method correspondingly reduces the impact of Doppler frequency offset on the transmission rate.

[0133] In some examples, in the sub-link between the relay assistant and the base station, the proportions of the relay assistant bandwidth resources allocated to each user terminal are the same and the transmission powers are the same. Specifically, in terms of the allocation method of the relay assistant bandwidth resources and transmission power resources, for the same user terminal, the proportion of the relay assistant bandwidth resources and the proportion of the transmission power are the same. Similarly to the above-mentioned MR-UAV, according to monotonicity, it can be obtained that when the power of the relay assistant is smaller, its energy consumption is smaller. Therefore, the proportions of the relay assistant bandwidth resources and transmission power allocated to each user can be obtained by equating the transmission rates.

[0134] In this way, the energy consumption of the entire communication system is reduced, and from the perspective of reinforcement learning, the successful results of reinforcement learning increase, making it easier to learn reasonable offloading strategies and resource allocation strategies.

[0135] Based on the above method simulation, some simulation results are obtained.

[0136] See Figure 3 , Figure 3 , which shows the system throughput comparison diagram in the scenario of introducing semi-following UAVs. It shows that the greater the following speed of the UAV, the greater the system throughput, that is, it proves that the UAV semi-following mobile carrier (such as a high-speed train) can effectively reduce the impact of Doppler frequency offset brought by mmWave technology on the transmission rate and improve the spectral efficiency of the system. In addition, the comparison between the dual-band with relay and the low-frequency band with relay shows that introducing mmWave technology at the UE-UAV side can effectively improve the system throughput.

[0137] See Figure 4 , Figure 4 , which shows the change diagram of the throughput of dual-band and high-frequency band transmissions with distance in the scenario of introducing UAVs. It shows the significance of introducing the mmWave method at the UE-UAV side in the dual-band instead of using the mmWave method in the entire frequency band. Figure 4It is obvious that in mmWave band communication, as the communication distance increases, the attenuation increases much more than that of dual-band transmission. Therefore, in long-distance communication of the UAV-BS link, using the sub-6Ghz method instead of the mmWave band for communication can reduce unnecessary waste of bandwidth resources.

[0138] See Figure 5 、 Figure 6 , Figure 5 FIG. shows a schematic diagram of the convergence process of minimizing the energy consumption of the reinforcement learning model training system; Figure 6 FIG. shows a schematic diagram of the user completion rate under the reinforcement learning model training. Figure 5 It shows that the reinforcement learning model based on P-DQN gradually optimizes and reduces the system energy consumption, indicating that the learning process is in the direction of optimizing and reducing energy consumption, and after 200 times of convergence, the learning changes from task completion learning to further optimizing the energy consumption based on task completion. Figure 6 What is shown is the learning trend of the completion degree of each round of tasks for every 15 users. It can be seen that after 200 rounds, there are a large number of rounds in which all users complete the tasks, which also indicates that the learning process starts to develop in the direction of reducing energy consumption within the constraints from this point on. The situation where the task completion degree is not high that appears during this period is often due to the exploration rate and the update of the target network not screening tasks with a high completion degree.

[0139] In the disclosure of the third embodiment, see Figure 7 , for Figure 1 , Figure 7 FIG. shows a structural diagram of a service data transmission device 70 provided in the third embodiment of the present disclosure. The device includes:

[0140] A determination module 701, configured to determine the offloading location of the service data based on the currently to-be-transmitted service data, and determine the first service data offloading ratio and the first bandwidth resource allocation ratio for offloading the service data to the relay end, and, determine the second service data offloading ratio and the second bandwidth resource allocation ratio for offloading to the base station end, where the offloading location includes at least one of the base station end and the relay end;

[0141] A sending module 702, configured to send the service data corresponding to the first service data offloading ratio to the relay end through the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio based on the determined offloading location, and, send the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio;

[0142] Wherein, the relay end is configured to receive and process the service data corresponding to the first service data offloading ratio, and the base station end is configured to receive and process the service data corresponding to the second service data offloading ratio.

[0143] In some examples, the determining module is specifically configured to:

[0144] Based on the current communication state of the user terminal and a reinforcement learning model, obtain a predicted transmission action of the user terminal;

[0145] Wherein, the user terminal sends service data according to the transmission action of the user terminal; the transmission action of the user terminal at least includes: the offloading location of the service data, the first service data offloading ratio, the first bandwidth resource allocation ratio, the second service data offloading ratio, and the second bandwidth resource allocation ratio.

[0146] In some examples, the communication state of the user terminal at least includes: the data volume of the service data, the current remaining terminal bandwidth resources, and the number of successful transmissions of the service data.

[0147] In some examples, the offloading location in the determining module is determined according to the following method: determining the offloading location based on the ε-greedy policy.

[0148] In some examples, the reinforcement learning model in the determining module is trained by the following method:

[0149] Based on the current communication state of the user terminal and a reinforcement learning model to be trained, obtain a predicted transmission action of the user terminal;

[0150] Execute the transmission action of the user terminal to obtain a new communication state of the user terminal and a reward parameter;

[0151] Define a target network based on the offloading location in the transmission action of the user terminal, the current communication state of the user terminal, the new communication state of the user terminal, and the reward parameter to obtain a target vector;

[0152] Calculate a stochastic gradient based on the target vector, the current communication state of the user terminal, and the offloading location in the transmission action of the user terminal, and update the reinforcement learning model according to the calculated stochastic gradient.

[0153] In some examples, the relay end is connected to a first edge computing server; the base station end is connected to a second edge computing server;

[0154] The relay end is configured to receive the service data corresponding to the first service data offloading ratio and send the service data corresponding to the first service data offloading ratio to the first edge computing server for processing, and the base station end is configured to receive the service data corresponding to the second service data offloading ratio and send the service data corresponding to the second service data offloading ratio to the second edge computing server for processing.

[0155] In the disclosure of the fourth embodiment, refer to Figure 8 , for Figure 2 ,Figure 8 The structural diagram of a transmission device 80 for service data provided by the fourth embodiment of the present disclosure is shown. The device includes:

[0156] A receiving module 801, configured to receive and process service data corresponding to a first service data offloading ratio sent by a user terminal through terminal bandwidth resources corresponding to a first bandwidth resource allocation ratio, wherein the terminal bandwidth resources corresponding to the first bandwidth resource allocation ratio and the service data corresponding to the first service data offloading ratio are determined by the method of S101 - S102;

[0157] Wherein, the user terminal sends service data corresponding to a second service data offloading ratio to the base station through terminal bandwidth resources corresponding to a second bandwidth resource allocation ratio.

[0158] In some examples, the receiving module receives the service data corresponding to the first service data offloading ratio sent by the user terminal by using any in - sub - band full - duplex communication method.

[0159] In some examples, the receiving module is specifically configured to: receive the service data corresponding to the first service data offloading ratio, and send the service data corresponding to the first service data offloading ratio to the first edge computing server for processing.

[0160] In some examples, the relay end is connected to the relay assisting end, and the transmission rate from the user terminal to the relay end is equal to the transmission rate from the relay end to the relay assisting end.

[0161] In some examples, the relay assisting end is connected to the base station end;

[0162] The user terminal sends service data corresponding to a second service data offloading ratio to the base station through terminal bandwidth resources corresponding to a second bandwidth resource allocation ratio, including:

[0163] The user terminal sends service data corresponding to a second service data offloading ratio to the base station through terminal bandwidth resources corresponding to a second bandwidth resource allocation ratio via the link of user terminal - relay end - relay assisting end - base station end.

[0164] In some examples, in the link of user terminal - relay end - relay assisting end - base station end, the sub - link of user terminal - relay end - relay assisting end transmits service data in the millimeter - wave manner; the sub - link of relay assisting end - base station end transmits service data in the radio frequency band manner.

[0165] In some examples, in the sub - link of relay assisting end - base station end, the proportion of relay assisting bandwidth resources allocated to each user terminal is the same and the transmission power is the same.

[0166] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium, and a computer program product.

[0167] Figure 9 FIG. shows a schematic block diagram of an exemplary electronic device 700 that can be used to implement embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as, for example, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, for example, a personal digital processor, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0168] As Figure 9 shown, the device 900 includes a computing unit 901 that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 902 or a computer program loaded from a storage unit 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the device 900 can also be stored. The computing unit 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0169] A plurality of components in the device 900 are connected to the I / O interface 905, including: an input unit 906, such as a keyboard, a mouse, etc.; an output unit 907, such as various types of displays, speakers, etc.; a storage unit 908, such as a magnetic disk, an optical disk, etc.; and a communication unit 909, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 909 allows the device 900 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0170] The computing unit 901 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 901 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 901 executes the various methods and processes described above, such as the method for transmitting service data. For example, in some embodiments, the method for transmitting service data can be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as the storage unit 908. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 900 via the ROM 902 and / or the communication unit 909. When the computer program is loaded into the RAM 903 and executed by the computing unit 901, one or more steps of the method for transmitting service data described above can be executed. Alternatively, in other embodiments, the computing unit 901 can be configured to execute the method for transmitting service data in any other suitable manner (e.g., by means of firmware).

[0171] Various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), systems-on-a-chip (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs, which can be executed and / or interpreted on a programmable system including at least one programmable processor, the programmable processor can be a special or general programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0172] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program codes can be executed entirely on the machine, partially on the machine, executed partially on the machine as an independent software package and partially on a remote machine, or executed entirely on a remote machine or server.

[0173] In the context of this disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0174] In order to provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0175] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0176] A computer system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The relationship of the client and the server is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, can also be a server of a distributed system, or a server incorporating a blockchain.

[0177] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution disclosed in this disclosure can be achieved, and no limitation is imposed herein.

[0178] The above specific embodiments do not constitute a limitation on the protection scope of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub - combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the protection scope of this disclosure.

Claims

1. A method for transmitting service data, applied to a user terminal, includes: Based on the currently to-be-transmitted service data, determining the offloading location of the service data, and determining a first service data offloading ratio and a first bandwidth resource allocation ratio for offloading the service data to a relay end, and determining a second service data offloading ratio and a second bandwidth resource allocation ratio for offloading to a base station end, where the offloading location includes at least one of the base station end and the relay end; Based on the determined offloading location, sending the service data corresponding to the first service data offloading ratio through the terminal bandwidth resources corresponding to the first bandwidth resource allocation ratio to the relay end, and sending the service data corresponding to the second service data offloading ratio through the terminal bandwidth resources corresponding to the second bandwidth resource allocation ratio to the base station end; Wherein, the relay end is configured to receive and process the service data corresponding to the first service data offloading ratio, and the base station end is configured to receive and process the service data corresponding to the second service data offloading ratio.

2. The method according to claim 1, wherein The determining the offloading location of the service data, and determining the first service data offloading ratio and the first bandwidth resource allocation ratio for offloading the service data to the relay end, and determining the second service data offloading ratio and the second bandwidth resource allocation ratio for offloading to the base station end based on the currently to-be-transmitted service data includes: According to the current communication state of the user terminal, obtaining a predicted user terminal transmission action based on a reinforcement learning model; Wherein, the user terminal sends the service data according to the user terminal transmission action; the user terminal transmission action at least includes: the offloading location of the service data, the first service data offloading ratio, the first bandwidth resource allocation ratio, the second service data offloading ratio, and the second bandwidth resource allocation ratio.

3. The method according to claim 2, wherein, The communication state of the user terminal at least includes: the data volume of the service data, the currently remaining terminal bandwidth resources, and the number of successful transmissions of the service data.

4. The method according to claim 2, wherein The offloading location is determined in the following manner: determining the offloading location based on an ε-greedy policy.

5. The method according to claim 2, wherein, The reinforcement learning model is trained in the following manner: According to the current communication state of the user terminal, obtaining a predicted user terminal transmission action based on the to-be-trained reinforcement learning model; Executing the user terminal transmission action to obtain a new user terminal communication state and a reward parameter; Defining a target network based on the offloading location in the user terminal transmission action, the current user terminal communication state, the new user terminal communication state, and the reward parameter to obtain a target vector; Calculating a stochastic gradient based on the target vector, the current user terminal communication state, and the offloading location in the user terminal transmission action, and updating the reinforcement learning model according to the calculated stochastic gradient.

6. The method according to any one of claims 1-5, wherein, The relay end is connected to a first edge computing server; the base station end is connected to a second edge computing server; The relay end is configured to receive the service data corresponding to the first service data offloading ratio, and send the service data corresponding to the first service data offloading ratio to the first edge computing server for processing. Also, the base station end is configured to receive the service data corresponding to the second service data offloading ratio, and send the service data corresponding to the second service data offloading ratio to the second edge computing server for processing.

7. A method for transmitting service data, applied to a relay end, includes: Receiving and processing the service data corresponding to the first service data offloading ratio sent by the user terminal through the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio, where the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio and the service data corresponding to the first service data offloading ratio are determined by the method described in claim 1; Wherein, the user terminal sends the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio.

8. The method according to claim 7, wherein, The relay end receives the service data corresponding to the first service data offloading ratio sent by the user terminal by using the full-duplex communication method within any sub-band.

9. The method according to claim 7, wherein The relay end is connected to a relay auxiliary end, and the transmission rate from the user terminal to the relay end is equal to the transmission rate from the relay end to the relay auxiliary end.

10. The method according to claim 9, wherein, The relay auxiliary end is connected to the base station end; The user terminal sending the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio includes: The user terminal sends the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio via the link of the user terminal - the relay end - the relay auxiliary end - the base station end.

11. The method according to claim 10, wherein, In the link of the user terminal - the relay end - the relay auxiliary end - the base station end, the sub-link of the user terminal - the relay end - the relay auxiliary end transmits service data in the millimeter wave manner; the sub-link of the relay auxiliary end - the base station end transmits service data in the radio frequency band manner.

12. The method according to claim 11, wherein, In the sub-link of the relay auxiliary end - the base station end, the ratio of the relay auxiliary bandwidth resources allocated to each user terminal is the same and the transmission power is the same.

13. The method according to any one of claims 7 - 12, wherein, The relay end is connected to the first edge computing server; The receiving and processing the service data corresponding to the first service data offloading ratio sent by the user terminal through the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio includes: Receiving the service data corresponding to the first service data offloading ratio, and sending the service data corresponding to the first service data offloading ratio to the first edge computing server for processing.

14. A device for transmitting service data includes: A determination module, configured to determine an offloading location of the service data based on the service data to be transmitted currently, and determine a first service data offloading ratio and a first bandwidth resource allocation ratio for offloading the service data to a relay end, and a second service data offloading ratio and a second bandwidth resource allocation ratio for offloading to a base station end, where the offloading location includes at least one of a base station end and a relay end; A sending module, configured to send, based on the determined offloading location, the service data corresponding to the first service data offloading ratio to the relay end through the terminal bandwidth resource corresponding to the first bandwidth resource allocation ratio, and send the service data corresponding to the second service data offloading ratio to the base station end through the terminal bandwidth resource corresponding to the second bandwidth resource allocation ratio; Wherein, the relay end is configured to receive and process the service data corresponding to the first service data offloading ratio, and the base station end is configured to receive and process the service data corresponding to the second service data offloading ratio.

15. The apparatus according to claim 14, wherein, The determination module is specifically configured to: Based on the current communication state of the user terminal and a reinforcement learning model, obtain a predicted transmission action of the user terminal; Wherein, the user terminal sends the service data according to the transmission action of the user terminal; the transmission action of the user terminal at least includes: the offloading location of the service data, the first service data offloading ratio, the first bandwidth resource allocation ratio, the second service data offloading ratio, and the second bandwidth resource allocation ratio.

16. The apparatus according to claim 15, wherein, The communication state of the user terminal at least includes: the data volume of the service data, the remaining terminal bandwidth resource currently, and the number of successful transmissions of the service data.

17. The apparatus according to claim 15, wherein, The offloading location in the determination module is determined in the following manner: determining the offloading location based on an ε-greedy strategy.

18. The apparatus according to claim 15, wherein, The reinforcement learning model in the determination module is trained in the following manner: Based on the current communication state of the user terminal and a reinforcement learning model to be trained, obtain a predicted transmission action of the user terminal; Execute the transmission action of the user terminal to obtain a new communication state of the user terminal and a reward parameter; Define a target network based on the offloading location in the transmission action of the user terminal, the current communication state of the user terminal, the new communication state of the user terminal, and the reward parameter to obtain a target vector; Calculate a stochastic gradient based on the target vector, the current communication state of the user terminal, and the offloading location in the transmission action of the user terminal, and update the reinforcement learning model according to the calculated stochastic gradient.

19. The device according to any one of claims 15-18, wherein, The relay end is connected to a first edge computing server; the base station end is connected to a second edge computing server; The relay end is configured to receive the service data corresponding to the first service data offloading ratio and send the service data corresponding to the first service data offloading ratio to the first edge computing server for processing, and the base station end is configured to receive the service data corresponding to the second service data offloading ratio and send the service data corresponding to the second service data offloading ratio to the second edge computing server for processing.

20. A transmission device for service data, comprising: A receiving module, configured to receive and process service data corresponding to a first service data offloading ratio sent by a user terminal through terminal bandwidth resources corresponding to a first bandwidth resource allocation ratio, wherein the terminal bandwidth resources corresponding to the first bandwidth resource allocation ratio and the service data corresponding to the first service data offloading ratio are determined by the method described in claim 1; Wherein, the user terminal sends the service data corresponding to the second service data offloading ratio to the base station through the terminal bandwidth resources corresponding to the second bandwidth resource allocation ratio.

21. The apparatus according to claim 20, wherein, The receiving module receives the service data corresponding to the first service data offloading ratio sent by the user terminal by using a full-duplex communication method within any sub-band.

22. The apparatus according to claim 20, wherein Specifically, the receiving module is configured to: receive the service data corresponding to the first service data offloading ratio, and send the service data corresponding to the first service data offloading ratio to a first edge computing server for processing.

23. An electronic device, comprising: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in any one of claims 1-6, or execute the method described in any one of claims 7-13.

24. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to execute the method described in any one of claims 1-6, or execute the method described in any one of claims 7-13.

25. A computer program product, comprising a computer program which, when executed by a processor, implements the method described in any one of claims 1-6, or the method described in any one of claims 7-13.

Images