Data transmission optimization method for dynamic spectrum access, terminal device and storage medium

Abstract

The application discloses a data transmission optimization method for dynamic spectrum access, a terminal device and a storage medium. The method comprises the following steps: acquiring data transmission state information of a to-be-processed service and a preset to-be-executed action; inputting the data transmission state information and the to-be-executed action into a preset learning network model to obtain neural network weights; when a weight acquisition request sent by a management node is received, the neural network weights are sent to the management node, so that the management node allocates a time-frequency resource technical scheme according to the neural network weights. Through the technical scheme, the data transmission delay is shortened, the spectrum use efficiency is improved, and the consensus algorithm cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a data transmission optimization method, terminal equipment, and storage medium for dynamic spectrum access. Background Technology

[0002] For scenarios such as integrated air-space-ground applications under 6G networks, since the coverage area provided by low-orbit satellites cannot maintain continuity, and the latency and channel reliability are not as stable as those of terrestrial networks, the current goal is to automatically optimize data transmission latency, spectrum utilization, and consensus algorithm overhead for terminals deployed in remote areas.

[0003] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] This invention provides a data transmission optimization method, terminal device, and storage medium for dynamic spectrum access, aiming to solve the technical problem of automatically optimizing data transmission latency, spectrum utilization, and consensus algorithm overhead for terminals deployed in remote areas.

[0005] This invention provides a data transmission optimization method for dynamic spectrum access, applied to a first terminal device, which is a bidding node for time and frequency resources. The data transmission optimization method for dynamic spectrum access includes:

[0006] Obtain the data transmission status information of the pending business and the preset actions to be executed;

[0007] The data transmission status information and the action to be executed are input into a preset learning network model to obtain neural network weights;

[0008] Upon receiving a weight acquisition request from the management node, the neural network weights are sent to the management node so that the management node can allocate time-frequency resources based on the neural network weights.

[0009] Optionally, the step of inputting the data transmission status information and the action to be executed into a preset learning network model to obtain neural network weights includes:

[0010] Based on the data buffer occupancy information of the pending service corresponding to each time slot and the pending action, the transaction cost-efficiency ratio corresponding to the application for time-frequency resources in each time slot is determined. The data transmission status information includes the data buffer occupancy information of the pending service corresponding to multiple time slots.

[0011] The neural network weights are obtained by using the maximum transaction cost-efficiency ratio among the transaction cost-efficiency ratios corresponding to the time-frequency resources requested in each time slot and the preset learning network model.

[0012] Optionally, the step of determining the transaction cost-effectiveness ratio corresponding to the application for time-frequency resources in each time slot, based on the data buffer occupancy information of the pending service and the pending action for each time slot, includes:

[0013] Determine the data transmission efficiency value of the pending service based on the data buffer occupancy information after the execution of the action to be executed;

[0014] Obtain payment information for the bidding application time slot;

[0015] The transaction cost-effectiveness ratio corresponding to the application for time-frequency resources in each time slot is determined based on the data transmission efficiency value and the payment information.

[0016] Optionally, the step of determining the data transmission benefit value of the pending service based on the data buffer occupancy information after the execution of the action to be executed includes:

[0017] Obtain the data transmission latency weighted value, throughput, and time-frequency resource utilization rate of the service to be processed after the execution of the action to be executed;

[0018] The data transmission benefit value is determined based on the data transmission delay weighting, throughput, and utilization rate of time-frequency resources.

[0019] Furthermore, to achieve the above objectives, the present invention also provides a data transmission optimization system for dynamic spectrum access, applied to a second terminal device, which is a management node for allocating time and frequency resources. The data transmission optimization method for dynamic spectrum access includes:

[0020] Send a weight acquisition request to the bidding nodes that applied for time-frequency resources;

[0021] When receiving the neural network weights fed back by the bidding node based on the weights request, the neural network weights corresponding to the bidding node are updated according to the neural network weights.

[0022] Time-frequency resources are allocated to the bidding nodes based on the updated neural network weights.

[0023] Optionally, the number of bidding nodes applying for time-frequency resources is at least two, and the step of updating the neural network weights corresponding to the bidding nodes according to the neural network weights includes:

[0024] The neural network weights between at least two of the bidding nodes are compared to obtain the comparison results;

[0025] The target neural network weights to be updated are determined based on the comparison results.

[0026] Based on the target neural network weights and the preset consensus algorithm, the target neural network weights are consensus-updated to each of the bidding nodes in the blockchain network.

[0027] Optionally, the step of allocating time-frequency resources to the bidding nodes based on the updated neural network weights includes:

[0028] Determine the execution action corresponding to the updated neural network weights;

[0029] When the executed action matches the preset executed action, time and frequency resources are allocated to the bidding node according to the time and frequency resource application information of the bidding node.

[0030] Furthermore, to achieve the above objectives, the present invention also provides a terminal device, characterized in that the terminal device comprises:

[0031] The acquisition module is used to acquire the data transmission status information of the business to be processed and the preset actions to be executed;

[0032] The input module is used to input the data transmission status information and the action to be executed into a preset learning network model to obtain neural network weights;

[0033] The first sending module, upon receiving a weight acquisition request from the management node, sends the neural network weights to the management node, so that the management node allocates time-frequency resources according to the neural network weights; or,

[0034] The terminal device includes:

[0035] The second sending module is used to send weight acquisition requests to the bidding nodes that applied for time and frequency resources;

[0036] The update module is used to update the neural network weights corresponding to the bidding node according to the neural network weights when the bidding node receives the neural network weights fed back by the request based on the weights.

[0037] The allocation module is used to allocate time-frequency resources to the bidding nodes according to the updated neural network weights.

[0038] Furthermore, to achieve the above objectives, the present invention also provides a terminal device comprising: a memory, a processor, and a dynamic spectrum access data transmission optimization program stored in the memory and executable on the processor, wherein the dynamic spectrum access data transmission optimization program, when executed by the processor, implements the steps of the dynamic spectrum access data transmission optimization method described above.

[0039] Furthermore, to achieve the above objectives, the present invention also provides a storage medium storing a data transmission optimization program for dynamic spectrum access, wherein the data transmission optimization program for dynamic spectrum access, when executed by a processor, implements the steps of the data transmission optimization method for dynamic spectrum access described above.

[0040] The data transmission optimization method, terminal equipment, and storage medium for dynamic spectrum access provided in this embodiment of the invention have at least the following technical effects or advantages:

[0041] This technical solution employs a method that acquires the data transmission status information and preset actions to be executed for pending services; inputs the data transmission status information and the actions to be executed into a preset learning network model to obtain neural network weights; and upon receiving a weight acquisition request from a management node, sends the neural network weights to the management node, enabling the management node to allocate time-frequency resources based on the neural network weights. This solution addresses the issues of data transmission latency, spectrum utilization, and consensus algorithm overhead in terminals deployed in remote areas, achieving reduced data transmission latency, thereby improving spectrum utilization efficiency and reducing consensus algorithm costs. Attached Figure Description

[0042] Figure 1 The diagram shows the structure of the terminal devices involved in various embodiments of the dynamic spectrum access data transmission optimization method of the present invention.

[0043] Figure 2 This is a schematic diagram illustrating the process of applying the dynamic spectrum access data transmission optimization method of the present invention to a first terminal device;

[0044] Figure 3 This is a schematic diagram of the process for obtaining neural network weights in the dynamic spectrum access data transmission optimization method of the present invention;

[0045] Figure 4 This is a flowchart illustrating the process of determining the transaction cost-effectiveness ratio in the data transmission optimization method for dynamic spectrum access of the present invention.

[0046] Figure 5 A diagram illustrating how bidding nodes use blockchain to apply for and use time-frequency resources;

[0047] Figure 6 This is a schematic diagram illustrating the process of applying the dynamic spectrum access data transmission optimization method of the present invention to a second terminal device;

[0048] Figure 7 A schematic diagram of the module composition of the first terminal device provided by the present invention;

[0049] Figure 8 A schematic diagram of the module composition of the second terminal device provided by the present invention. Detailed Implementation

[0050] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the invention to those skilled in the art.

[0051] In the following description, the use of suffixes such as "module," "part," or "unit" to denote elements is solely for the purpose of illustrative purposes and has no specific meaning in itself. Therefore, "module," "part," or "unit" may be used interchangeably.

[0052] Please refer to Figure 1 , Figure 1 This is a schematic diagram of the terminal devices involved in various embodiments of the dynamic spectrum access data transmission optimization method of the present invention. The terminal devices involved in the dynamic spectrum access data transmission optimization method of the present invention may include terminal devices such as mobile phones, tablet computers, laptops, handheld computers, and personal digital assistants (PDAs).

[0053] like Figure 1 As shown, the terminal device may include a memory 101 and a processor 102. Those skilled in the art will understand that... Figure 1 The structural block diagram of the terminal shown does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown, or combine certain components, or have different component arrangements. The memory 101 stores the operating system and a data transmission optimization program for dynamic spectrum access. The processor 102 is the control center of the terminal device. The processor 102 executes the data transmission optimization program for dynamic spectrum access stored in the memory 101 to implement the steps of the various embodiments of the data transmission optimization method for dynamic spectrum access of the present invention.

[0054] Optionally, the terminal device may also include a communication unit 103, which establishes data communication with other terminal devices such as computers via a network protocol (the data communication may be IP communication or Bluetooth channel) to realize data transmission with other terminal devices.

[0055] It should be noted that when the terminal device is the first terminal device, the first terminal device is an auction node in the blockchain network used to apply for time and frequency resources. When the data transmission optimization program for dynamic spectrum access in the memory 101 of the first terminal is executed by the processor 102, it implements the following steps:

[0056] Obtain the data transmission status information of the pending business and the preset actions to be executed;

[0057] The data transmission status information and the action to be executed are input into a preset learning network model to obtain neural network weights;

[0058] Upon receiving a weight acquisition request from the management node, the neural network weights are sent to the management node so that the management node can allocate time-frequency resources based on the neural network weights.

[0059] Furthermore, the processor 102 can be used to call the data transmission optimization program for dynamic spectrum access stored in the memory 101, and perform the following operations: inputting the data transmission status information and the action to be executed into a preset learning network model to obtain neural network weights includes the following steps:

[0060] Based on the data buffer occupancy information of the pending service corresponding to each time slot and the pending action, the transaction cost-efficiency ratio corresponding to the application for time-frequency resources in each time slot is determined. The data transmission status information includes the data buffer occupancy information of the pending service corresponding to multiple time slots.

[0061] The neural network weights are obtained by using the maximum transaction cost-efficiency ratio among the transaction cost-efficiency ratios corresponding to the time-frequency resources requested in each time slot and the preset learning network model.

[0062] Furthermore, the processor 102 can be used to call the data transmission optimization program for dynamic spectrum access stored in the memory 101, and perform the following steps: determining the transaction cost-effectiveness ratio corresponding to the application for time-frequency resources in each time slot based on the data buffer occupancy information of the service to be processed corresponding to each time slot and the action to be executed, including:

[0063] Determine the data transmission efficiency value of the pending service based on the data buffer occupancy information after the execution of the action to be executed;

[0064] Obtain payment information for the bidding application time slot;

[0065] The transaction cost-effectiveness ratio corresponding to the application for time-frequency resources in each time slot is determined based on the data transmission efficiency value and the payment information.

[0066] Furthermore, the processor 102 can be used to call the data transmission optimization program for dynamic spectrum access stored in the memory 101, and perform the following operations: the step of determining the data transmission benefit value of the service to be processed based on the data buffer occupancy information after the execution of the action to be executed includes:

[0067] Obtain the data transmission latency weighted value, throughput, and time-frequency resource utilization rate of the service to be processed after the execution of the action to be executed;

[0068] The data transmission benefit value is determined based on the data transmission delay weighting, throughput, and utilization rate of time-frequency resources.

[0069] Correspondingly, when the terminal device is a second terminal device, the second terminal device is a management node in the blockchain network used to allocate time and frequency resources. When the data transmission optimization program for dynamic spectrum access in the memory 101 of the second terminal is executed by the processor 102, it implements the following steps:

[0070] Send a weight acquisition request to the bidding nodes that applied for time-frequency resources;

[0071] When receiving the neural network weights fed back by the bidding node based on the weights request, the neural network weights corresponding to the bidding node are updated according to the neural network weights.

[0072] Time-frequency resources are allocated to the bidding nodes based on the updated neural network weights.

[0073] Furthermore, the processor 102 can be used to call the data transmission optimization program for dynamic spectrum access stored in the memory 101 and perform the following operations: the number of bidding nodes applying for time-frequency resources is at least two, and the step of updating the neural network weights corresponding to the bidding nodes according to the neural network weights includes:

[0074] The neural network weights between at least two of the bidding nodes are compared to obtain the comparison results;

[0075] The target neural network weights to be updated are determined based on the comparison results.

[0076] Based on the target neural network weights and the preset consensus algorithm, the target neural network weights are consensus-updated to each of the bidding nodes in the blockchain network.

[0077] Furthermore, the processor 102 can be used to call the data transmission optimization program for dynamic spectrum access stored in the memory 101, and perform the following operations: the step of allocating time-frequency resources to the bidding node according to the updated neural network weights includes:

[0078] Determine the execution action corresponding to the updated neural network weights;

[0079] When the executed action matches the preset executed action, time and frequency resources are allocated to the bidding node according to the time and frequency resource application information of the bidding node.

[0080] This invention provides an embodiment of a data transmission optimization method for dynamic spectrum access. It should be noted that although the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown here.

[0081] It should be noted that the data transmission optimization method for dynamic spectrum access in this invention employs a convolutional neural network model, and the deep Q-learning of each node constituting the blockchain network adopts the Dueling DQN reinforcement learning method. The Dueling DQN model structure consists of two parts: a state-value function and an advantage function. The state-value function is independent of actions and returns only a state value, while the advantage function is related to both the executed action and the state. The state-value function and the advantage function share neural network weights but each has its own unique parameters. The principle of the Dueling DQN network is to utilize the state-value function to obtain the stability of the neural network weights within a certain learning period. During the learning period, when updating the neural network weights through the advantage function, normalization and the difference average of the changes in the neural network weights of the advantage function are usually performed. After a certain number of training iterations, the neural network weights determined by the state function are updated again to achieve better estimation results.

[0082] Optionally, a node can be an auction node or a management node.

[0083] Optionally, the convolutional neural network model may employ the VGG11 neural network model.

[0084] Based on the above structural block diagram of the terminal device, various embodiments of the data transmission optimization method for dynamic spectrum access of the present invention are proposed. In one embodiment, the present invention provides a data transmission optimization method for dynamic spectrum access, which is applied to a first terminal device. Please refer to [reference needed]. Figure 2 , Figure 2 This is a schematic diagram illustrating the application of the dynamic spectrum access data transmission optimization method of the present invention to a first terminal device. In this embodiment, the dynamic spectrum access data transmission optimization method includes the following steps:

[0085] Step S10: Obtain the data transmission status information of the service to be processed and the preset actions to be executed;

[0086] Step S20: Input the data transmission status information and the action to be executed into a preset learning network model to obtain neural network weights;

[0087] Step S30: Upon receiving a weight acquisition request from the management node, the neural network weights are sent to the management node so that the management node allocates time-frequency resources based on the neural network weights.

[0088] The data transmission status information includes data buffer occupancy information for multiple time slots corresponding to pending services. It should be noted that data buffer occupancy information can be represented by a data buffer occupancy ratio. Optionally, the denominator of the data buffer occupancy ratio is "size of buffer occupied by pending services" + "size of free buffer," and the denominator is not the overall size of the physical buffer. Preset actions to be executed can be pre-set as needed. For example, the action to be executed could be either selecting to initiate a blockchain request but not executing it, or selecting to initiate a blockchain request and executing it.

[0089] Optionally, the business to be processed can be a single business or multiple business processes.

[0090] Alternatively, please refer to Figure 3 , Figure 3 This is a flowchart illustrating the process of obtaining neural network weights in the dynamic spectrum access data transmission optimization method of the present invention. Step S10 includes:

[0091] Step S11: Determine the transaction cost-effectiveness ratio corresponding to the application for time-frequency resources in each time slot based on the data buffer occupancy information of the service to be processed corresponding to each time slot and the action to be executed.

[0092] Step S12: Based on the maximum transaction cost efficiency ratio among the transaction cost efficiency ratios corresponding to the time-frequency resources applied for in each time slot and the preset learning network model, the weights of the neural network are obtained.

[0093] The data transmission status information includes data buffer occupancy information for the pending services corresponding to multiple time slots.

[0094] By determining the transaction cost-efficiency ratio corresponding to the application for time-frequency resources in each time slot, and then determining the maximum transaction cost-efficiency ratio among the transaction cost-efficiency ratios corresponding to the application for time-frequency resources in each time slot, the maximum transaction cost-efficiency ratio is used to maximize the output efficiency with less investment. This optimizes various transmission parameters during data transmission, such as reducing the transmission latency of pending business data transmission, increasing the actual utilization rate of wireless time-frequency resources obtained through bidding, and reducing the transmission cost-efficiency ratio of data transmission achieved by successfully applying for time-frequency resources. Finally, based on the maximum transaction cost-efficiency ratio and the preset learning network model, neural network weights are obtained. The decision-making effect and rationality of the bidding nodes applying for time-frequency resources are evaluated through the neural network weights.

[0095] Alternatively, please refer to Figure 4 , Figure 4 This is a flowchart illustrating the process of determining the transaction cost-effectiveness ratio in the dynamic spectrum access data transmission optimization method of the present invention. Step S20 includes:

[0096] Step S21: Determine the data transmission benefit value of the pending service based on the data buffer occupancy information after the execution of the action to be executed;

[0097] Step S22: Obtain payment information for the bidding application time resources;

[0098] Step S23: Determine the transaction cost-efficiency ratio corresponding to the application for time-frequency resources in each time slot based on the data transmission benefit value and the payment information.

[0099] Optionally, step S21 includes:

[0100] Obtain the data transmission latency weighted value, throughput, and time-frequency resource utilization rate of the service to be processed after the execution of the action to be executed;

[0101] The data transmission benefit value is determined based on the data transmission delay weighting, throughput, and utilization rate of time-frequency resources.

[0102] The data transmission delay weighting value for a pending service can be determined based on the weighting value corresponding to the pending service and whether the pending service meets the delay requirements. For example, if the pending service meets the delay requirements, the value is 1, and if the pending service does not meet the delay requirements, the value is 0. The data transmission delay weighting value can be directly obtained by multiplying the weighting value corresponding to the pending service and the product between the pending service meeting the delay requirements. If the pending service consists of at least two pending services, the sum of the data transmission delay weighting values ​​corresponding to each of the pending services can be obtained, and this sum can be determined as the data transmission delay weighting value for the pending service.

[0103] Similarly, the throughput of a pending service can be determined based on the number of bits transmitted in the current window when the pending service is a single service; when the pending service comprises at least two services, the sum of the number of bits transmitted for each service can be obtained, and this sum can be determined as the throughput of the pending service. The utilization rate of time-frequency resources is the utilization rate of wireless time-frequency resources as counted on a frame-by-frame basis.

[0104] The data transmission benefit value is determined based on the data transmission delay weighting value, throughput, and time-frequency resource utilization rate. The sum of the data transmission delay weighting value, throughput, and time-frequency resource utilization rate can be obtained, and the data transmission benefit value is determined based on the sum, such as directly determining the sum as the data transmission benefit value.

[0105] Obtain payment information for bidding on time-frequency resources. The payment information includes the amount of virtual currency paid by the bidding node to the management node for the time-frequency resources and the amount of virtual currency required to purchase the required number of time-frequency resources. For example, the amount of virtual currency required to purchase the required number of time-frequency resources can be determined by spending 1 virtual currency for every 100 time-frequency resource blocks.

[0106] The transaction cost efficiency ratio corresponding to the application for time and frequency resources in each time slot is determined based on the data transmission efficiency value and the payment information. The transaction cost efficiency ratio can be directly determined based on the ratio between the data transmission efficiency value and the payment information.

[0107] For example, to facilitate understanding of this embodiment, it is assumed that the data transmission status information of the service to be processed in the bidding node applying for time and frequency resources includes data transmission status and voice status, and the data transmission status is s. t ∈{S0 / S1 / S2 / S3 / ...... / S 015}, voice status is r t ∈: {R0 / R1 / R2 / R3}, let v t ={s t r t}, v t This represents the data buffer occupancy information of the terminal's pending services at time t, and the set of preset actions to be executed is represented by act = {a0, a1, ..., a100}. For the current time slot t, the action to be executed is denoted as act(t).

[0108] Step S11: Determine the transaction cost-effectiveness ratio corresponding to the application for time-frequency resources in each time slot based on the data buffer occupancy information of the service to be processed corresponding to each time slot and the action to be executed.

[0109] Specifically, taking the data buffer occupancy information and pending actions of the service to be processed corresponding to a time slot as an example, the transaction cost-effectiveness ratio corresponding to requesting time-frequency resources in that time slot is determined. Assume that in time slot t, in a certain state v t After executing the action act(t), the transaction cost efficiency ratio rate(v) t The expression (act(t)) can be determined by the following expression: (A+B+C) / D

[0110] in,

[0111] A = ∑ 遍历当前业务 (Service weighting value * each service meets transmission delay requirements),

[0112] B = ∑ 遍历当前企业 Number of bits sent per service in the current window

[0113] C = ∑ Wireless resource utilization rate per frame.

[0114] D = Pay0 + Cost 100 *The number of RB resources requested in the current window. It should be noted that...

[0115] ∑ 遍历当前业务 (Service weighted value * each service meets the transmission delay requirement) is the data transmission delay weighted value of the service to be processed. When the service to be processed transmits data, the transmission delay is weighted. If the transmission delay is met, it is 1; if the delay requirement is exceeded, it is 0.

[0116] ∑ 遍历当前企业 The number of bits sent in the current window for each service represents the throughput of the service to be processed, indicating whether the current terminal, i.e., the bidding node, has efficiently utilized the requested time-frequency resources within a continuously allocated time-frequency resource period. The statistical time can be set to a finite time window, for example, configuring a statistical window length of 10 seconds for IoT services, with the terminal using a 10-second window length to statistically analyze the data throughput of transmission and reception.

[0117] The ∑ frame-by-frame wireless resource utilization rate represents the utilization rate of time-frequency resources for the services to be processed. It is used to assess whether the services transmitted in each time slot within the current statistical window occupy the entire available bandwidth or only use a portion of it. For reference... Figure 5 , Figure 5 A diagram illustrating how bidding nodes utilize blockchain to apply for and use time-frequency resources.

[0118] Furthermore, when setting up spectrum auctions, each bidding node applying for time-frequency resources must pay 20 virtual currency to the management node for each bidding transaction, i.e., Pay0 = 20. Every 100 time-frequency resource blocks requires 1 virtual currency, i.e., Cost. 100 =1.

[0119] Step S12: Based on the maximum transaction cost efficiency ratio among the transaction cost efficiency ratios corresponding to the time-frequency resources applied for in each time slot and the preset learning network model, the weights of the neural network are obtained.

[0120] It should be noted that the input variable in the above formula is: (v t `act(t)` refers to the Q-value. By employing deep Q-learning, an optimal strategy can be obtained, making the transaction cost-effectiveness ratio `rate(v)` in the above formula equal to the Q-value. t The overall optimization objective of act(t) is to maximize the value, i.e., maximize the benefit, and the expression is:

[0121] rate t =arg max act∈A Q(v t ,act(t))

[0122] The technical solution disclosed in this embodiment employs the following approach: acquiring data transmission status information of the service to be processed and preset actions to be executed; inputting the data transmission status information and the actions to be executed into a preset learning network model to obtain neural network weights; and upon receiving a weight acquisition request from a management node, sending the neural network weights to the management node so that the management node allocates time-frequency resources according to the neural network weights. This technical solution solves the problem of automatically optimizing data transmission latency, spectrum utilization, and consensus algorithm overhead for terminals deployed in remote areas, thereby shortening data transmission latency, improving spectrum utilization efficiency, and reducing consensus algorithm costs.

[0123] Correspondingly, the present invention also provides a data transmission optimization method for dynamic spectrum access, which is applied to a second terminal device. Please refer to [reference needed]. Figure 6 , Figure 6 This is a schematic diagram illustrating the application of the dynamic spectrum access data transmission optimization method of the present invention to a second terminal device. In this embodiment, the dynamic spectrum access data transmission optimization method includes the following steps:

[0124] Step S40: Send a weight acquisition request to the bidding node that applied for the time and frequency resources;

[0125] Step S50: When receiving the neural network weights of the bidding node based on the weights to obtain the request feedback, update the neural network weights corresponding to the bidding node according to the neural network weights.

[0126] Step S60: Allocate time-frequency resources to the bidding nodes according to the updated neural network weights.

[0127] As an optional implementation, the number of bidding nodes applying for time-frequency resources is at least two. Step S50, updating the neural network weights corresponding to the bidding nodes based on the neural network weights, includes:

[0128] The neural network weights between at least two of the bidding nodes are compared to obtain the comparison results;

[0129] The target neural network weights to be updated are determined based on the comparison results.

[0130] Based on the target neural network weights and the preset consensus algorithm, the target neural network weights are consensus-updated to each of the bidding nodes in the blockchain network.

[0131] When there are at least two bidding nodes applying for time and frequency resources, in order to maximize the transaction cost-effectiveness ratio of the data transmission of pending services by the bidding nodes that successfully apply for time and frequency resources, and to maximize the benefits of less investment, various transmission parameters are optimized during data transmission. For example, the transmission latency of pending service data transmission is shorter, the actual utilization rate of the wireless time and frequency resources obtained through bidding is larger, and the transmission cost-effectiveness ratio of data transmission achieved by successfully applying for time and frequency resources is smaller.

[0132] The neural network weights between at least two bidding nodes are compared to obtain the comparison results, so as to determine the minimum neural network weight or the maximum neural network weight among the at least two bidding nodes.

[0133] The target neural network weights to be updated are determined based on the comparison results. The largest neural network weight determined in the comparison results can be used as the target neural network weights to be updated. The bidding node corresponding to the largest neural network weight successfully applied for the time-frequency resources, which can make the transmission parameters optimal during data transmission. Then, based on the target neural network weights and the preset consensus algorithm, the target neural network weights are consensus-updated to each bidding node in the blockchain network, so that each bidding node can know the neural network weights of the bidding nodes that successfully bid for the time-frequency resources.

[0134] Optionally, step S60 includes:

[0135] Determine the execution action corresponding to the updated neural network weights;

[0136] When the executed action matches the preset executed action, time and frequency resources are allocated to the bidding node according to the time and frequency resource application information of the bidding node.

[0137] The preset execution action can be set in advance, and the preset execution action can be the action of requesting time-frequency resources. When the execution action corresponding to the updated neural network weights matches the preset execution action, it indicates that the bidding node corresponding to the neural network weights has successfully applied for time-frequency resources, and the management node can allocate time-frequency resources to the bidding node according to the time-frequency resource application information of the bidding node.

[0138] like Figure 7 As shown, Figure 7 This is a schematic diagram of the module composition of the first terminal device provided by the present invention. The first terminal device 100 includes:

[0139] The acquisition module 110 is used to acquire the data transmission status information of the service to be processed and the preset actions to be executed;

[0140] The input module 120 is used to input the data transmission status information and the action to be executed into a preset learning network model to obtain neural network weights;

[0141] When the first sending module 130 receives a weight acquisition request sent by the management node, it sends the neural network weights to the management node so that the management node can allocate time-frequency resources according to the neural network weights.

[0142] Optionally, the input module 120 includes a first determining module and a calculation module.

[0143] The first determining module is used to determine the transaction cost-effectiveness ratio corresponding to the application for time-frequency resources in each time slot based on the data buffer occupancy information of the pending service corresponding to each time slot and the pending action. The data transmission status information includes the data buffer occupancy information of the pending service corresponding to multiple time slots.

[0144] The calculation module is used to obtain the neural network weights based on the maximum transaction cost-efficiency ratio among the transaction cost-efficiency ratios corresponding to the time-frequency resources applied for in each time slot and the preset learning network model.

[0145] Optionally, the first determining module includes a second determining module, a first sub-acquisition module, and a third determining module. The second determining module is used to determine the data transmission benefit value of the pending service based on the data buffer occupancy information after the execution of the action to be executed.

[0146] The first sub-acquisition module is used to obtain payment information for bidding applications for time-frequency resources;

[0147] The third determining module is used to determine the transaction cost-effectiveness ratio corresponding to the application for time-frequency resources in each time slot based on the data transmission benefit value and the payment information.

[0148] Optionally, the second determining module includes a second sub-acquisition module and a fourth determining module. The second sub-acquisition module is used to acquire the data transmission delay weighting value, throughput, and time-frequency resource utilization rate of the service to be processed after the execution of the action to be executed.

[0149] The fourth determining module is used to determine the data transmission benefit value based on the data transmission delay weighting value, throughput, and utilization rate of time and frequency resources.

[0150] Or, such as Figure 8 As shown, Figure 8 This is a schematic diagram of the module composition of the second terminal device provided by the present invention. The second terminal device 200 includes:

[0151] The second sending module 210 is used to send a weight acquisition request to the bidding node that applied for time and frequency resources;

[0152] The update module 220 is used to update the neural network weights corresponding to the bidding node according to the neural network weights when the bidding node receives the neural network weights fed back by the request based on the weights.

[0153] The allocation module 230 is used to allocate time-frequency resources to the bidding nodes according to the updated neural network weights.

[0154] Optionally, the number of bidding nodes applying for time-frequency resources is at least two. The update module 220 includes a comparison module, a fifth determination module, and a consensus module. The comparison module is used to compare the neural network weights between at least two bidding nodes to obtain the comparison result.

[0155] The fifth determining module is used to determine the target neural network weights to be updated based on the comparison results;

[0156] The consensus module is used to update the target neural network weights to each bidding node in the blockchain network according to the target neural network weights and a preset consensus algorithm.

[0157] Optionally, the allocation module 230 includes a sixth determining module and a sub-allocation module. The sixth determining module is used to determine the execution action corresponding to the updated neural network weights.

[0158] The sub-allocation module is used to allocate time-frequency resources to the bidding node according to the time-frequency resource application information of the bidding node when the executed action matches the preset executed action.

[0159] The specific implementation of the terminal devices of the present invention, such as the first terminal device and / or the second terminal device, is basically the same as the embodiments of the data transmission optimization method for dynamic spectrum access described above, and will not be repeated here.

[0160] The present invention also proposes a terminal device, which is either a first terminal device or a second terminal device. The terminal device includes a memory, a processor, and a data transmission optimization program for dynamic spectrum access stored in the memory and executable on the processor. When the terminal device is a first terminal, the data transmission optimization program for dynamic spectrum access, when executed by the processor of the first terminal, implements the steps of the data transmission optimization method for dynamic spectrum access applied to the first terminal in any of the above embodiments. When the terminal device is a second terminal, the data transmission optimization program for dynamic spectrum access, when executed by the processor of the second terminal, implements the steps of the data transmission optimization method for dynamic spectrum access applied to the second terminal in any of the above embodiments.

[0161] The present invention also proposes a storage medium storing a data transmission optimization program for dynamic spectrum access, wherein when the data transmission optimization program for dynamic spectrum access is executed by a processor, it implements the steps of the data transmission optimization method for dynamic spectrum access as described in any of the above embodiments.

[0162] The embodiments of the terminal device and storage medium provided by the present invention include all the technical features of the various embodiments of the above-described dynamic spectrum access data transmission optimization method. The extended and explanatory content of the specification is basically the same as the various embodiments of the above-described dynamic spectrum access data transmission optimization method, and will not be repeated here.

[0163] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0164] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0165] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0166] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0167] It should be noted that any reference signs placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0168] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0169] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for optimizing data transmission in dynamic spectrum access, characterized in that, The data transmission optimization method for dynamic spectrum access, applied to a first terminal device (which is a bidding node for time-frequency resources), includes: Obtain the data transmission status information of the pending business and the preset actions to be executed; Based on the data buffer occupancy information of the pending service corresponding to each time slot and the pending action, the transaction cost-efficiency ratio corresponding to the application for time-frequency resources in each time slot is determined. The data transmission status information includes the data buffer occupancy information of the pending service corresponding to multiple time slots. The neural network weights are obtained by using the maximum transaction cost-efficiency ratio among the transaction cost-efficiency ratios corresponding to the time-frequency resources applied for in each time slot, and the preset learning network model. Upon receiving a weight acquisition request from the management node, the neural network weights are sent to the management node so that the management node can allocate time-frequency resources based on the neural network weights.

2. The method as described in claim 1, characterized in that, The step of determining the transaction cost-effectiveness ratio corresponding to the application for time-frequency resources in each time slot based on the data buffer occupancy information of the pending service and the pending action for each time slot includes: Determine the data transmission efficiency value of the pending service based on the data buffer occupancy information after the execution of the action to be executed; Obtain payment information for the bidding application time slot; The transaction cost-effectiveness ratio corresponding to the application for time-frequency resources in each time slot is determined based on the data transmission efficiency value and the payment information.

3. The method as described in claim 2, characterized in that, The step of determining the data transmission efficiency value of the pending service based on the data buffer occupancy information after the execution of the action to be executed includes: Obtain the data transmission latency weighted value, throughput, and time-frequency resource utilization rate of the service to be processed after the execution of the action to be executed; The data transmission benefit value is determined based on the data transmission delay weighting, throughput, and utilization rate of time-frequency resources.

4. A data transmission optimization method for dynamic spectrum access, characterized in that, This is applied to second terminal devices, which are responsible for allocating time and frequency resources. Management Node The data transmission optimization method for dynamic spectrum access includes: A weight acquisition request is sent to the bidding node that applied for time-frequency resources. The bidding node acquires the data transmission status information of the pending service and the preset action to be executed. Based on the data buffer occupancy information of the pending service corresponding to each time slot and the action to be executed, it determines the transaction cost-effectiveness ratio corresponding to the application for time-frequency resources in each time slot. Based on the maximum transaction cost-effectiveness ratio among the transaction cost-effectiveness ratios corresponding to the application for time-frequency resources in each time slot and the preset learning network model, it obtains the neural network weights. When it receives the weight acquisition request sent by the management node, it sends the neural network weights to the management node. The data transmission status information includes the data buffer occupancy information of the pending service corresponding to multiple time slots. When receiving the neural network weights fed back by the bidding node based on the weights acquisition request, the neural network weights corresponding to the bidding node are updated according to the neural network weights. Time-frequency resources are allocated to the bidding nodes based on the updated neural network weights.

5. The method as described in claim 4, characterized in that, The number of bidding nodes for the time-frequency resources is at least two, and the step of updating the neural network weights corresponding to the bidding nodes based on the neural network weights includes: The neural network weights between at least two of the bidding nodes are compared to obtain the comparison results; The target neural network weights to be updated are determined based on the comparison results. Based on the target neural network weights and the preset consensus algorithm, the target neural network weights are consensus-updated to each of the bidding nodes in the blockchain network.

6. The method as described in claim 4, characterized in that, The step of allocating time-frequency resources to the bidding nodes based on the updated neural network weights includes: Determine the execution action corresponding to the updated neural network weights; When the executed action matches the preset executed action, time and frequency resources are allocated to the bidding node according to the time and frequency resource application information of the bidding node.

7. A terminal device, characterized in that, The terminal device includes: The acquisition module is used to acquire the data transmission status information of the business to be processed and the preset actions to be executed; The input module includes a first determining module and a calculation module. The first determining module is used to determine the transaction cost efficiency ratio corresponding to the application for time-frequency resources in each time slot based on the data buffer occupancy information of the pending service and the action to be executed for each time slot. The data transmission status information includes the data buffer occupancy information of the pending service corresponding to multiple time slots. The calculation module is used to obtain the neural network weights based on the maximum transaction cost efficiency ratio among the transaction cost efficiency ratios corresponding to the application for time-frequency resources in each time slot and a preset learning network model. The first sending module, upon receiving a weight acquisition request from the management node, sends the neural network weights to the management node, so that the management node allocates time-frequency resources according to the neural network weights; or, The terminal device includes: The second sending module is used to send a weight acquisition request to the bidding node that applied for time-frequency resources. The bidding node acquires the data transmission status information of the pending service and the preset action to be executed. Based on the data buffer occupancy information of the pending service corresponding to each time slot and the action to be executed, it determines the transaction cost-effectiveness ratio corresponding to the application for time-frequency resources in each time slot. Based on the maximum transaction cost-effectiveness ratio among the transaction cost-effectiveness ratios corresponding to the application for time-frequency resources in each time slot and the preset learning network model, it obtains the neural network weights. When it receives the weight acquisition request sent by the management node, it sends the neural network weights to the management node. The data transmission status information includes the data buffer occupancy information of the pending service corresponding to multiple time slots. The update module is used to update the neural network weights corresponding to the bidding node according to the neural network weights when the bidding node receives the neural network weights fed back by the request based on the weights. The allocation module is used to allocate time-frequency resources to the bidding nodes according to the updated neural network weights.

8. A terminal device, characterized in that, include: The system includes a memory, a processor, and a data transmission optimization program for dynamic spectrum access stored in the memory and executable on the processor. When executed by the processor, the data transmission optimization program for dynamic spectrum access implements the steps of the data transmission optimization method for dynamic spectrum access as described in any one of claims 1-3 or 4-6.

9. A storage medium, characterized in that, It stores a data transmission optimization program for dynamic spectrum access, which, when executed by a processor, implements the steps of the data transmission optimization method for dynamic spectrum access as described in any one of claims 1-3 or 4-6.

Images