Data transmission control method, device, storage medium and program product

Abstract

The application provides a data transmission control method, device, storage medium and program product, and the method comprises the following steps: obtaining a source decision state and a channel decision state of to-be-transmitted data; obtaining a source decision result according to the source decision state, wherein the source decision result is used for controlling a source device to transmit the to-be-transmitted data, and the source device is used for generating the to-be-transmitted data; and obtaining a channel decision result according to the channel decision state, wherein the channel decision result is used for controlling a channel device to transmit the to-be-transmitted data. The present scheme can achieve timely response to user operation and improve user experience.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field, and particularly relates to a data transmission control method, device, storage medium and program product. BACKGROUND

[0002] Web industry collaboration is a new type of collaborative work mode, which can improve the flexibility and efficiency of business, and optimize the configuration and use of resources, so that the whole system is more intelligent and efficient.

[0003] In the prior art, the service experience quality difference of the target service is predicted according to the correlation between the service experience indicators and the network performance indicators of various services, and the existing service experience indicators are difficult to directly measure the response timeliness of user operation, and there is a certain deviation in the measurement of user experience. SUMMARY

[0004] The purpose of the present application is to provide a data transmission control method, device, storage medium and program product, which is used to solve the problem that the existing web industry collaboration method is difficult to measure the response timeliness of user operation due to the use of service experience indicators, resulting in a certain deviation in the measurement of user experience.

[0005] In order to solve the above technical problems, in a first aspect, the embodiments of the present application provide a data transmission control method, which comprises:

[0006] Obtaining the source decision state and the channel decision state of the to-be-transmitted data;

[0007] According to the source decision state, a source decision result is obtained, the source decision result is used to control the transmission of the to-be-transmitted data by a source device, and the source device is used to generate the to-be-transmitted data;

[0008] According to the channel decision state, a channel decision result is obtained, and the channel decision result is used to control the transmission of the to-be-transmitted data by a channel device.

[0009] In some embodiments, the obtaining of the source decision state and the channel decision state of the to-be-transmitted data comprises:

[0010] Obtaining a communication network state;

[0011] According to the communication network state and the historical channel decision result, the source decision state is obtained;

[0012] According to the communication network state and the historical source decision result, the channel decision state is obtained.

[0013] In some embodiments, the obtaining of the source decision result according to the source decision state comprises:

[0014] inputting the source decision state into an intelligent decision network to obtain the source decision result;

[0015] The source decision result comprises a resolution and a sending frame rate.

[0016] inputting the channel decision state into an intelligent decision network to obtain the channel decision result;

[0017] The intelligent decision network comprises an information value function of a user, the information value function being related to an information age of the to-be-transmitted data and a quality of service corresponding to the to-be-transmitted data, and the information value function being used to adjust network parameters of the intelligent decision network.

[0018] In some embodiments, the method further comprises:

[0019] sending the source decision result to the source device and obtaining a first execution result of the to-be-transmitted data transmission performed by the source device based on the source decision result;

[0020] sending the channel decision result to the channel device and obtaining a second execution result of the to-be-transmitted data transmission performed by the channel device based on the channel decision result;

[0021] calculating a first information value by using the information value function according to the first execution result and the second execution result;

[0022] adjusting the network parameters of the intelligent decision network by using the first information value to obtain an adjusted intelligent decision network.

[0023] In some embodiments, the method further comprises:

[0024] obtaining the information age of the to-be-transmitted data based on a probability density function of a data packet arrival in the to-be-transmitted data and a probability density function of a data transmission rate;

[0025] obtaining the information value function according to the information age of the to-be-transmitted data and a quality of service modeling function.

[0026] In some embodiments, the obtaining the communication network state comprises:

[0027] obtaining a parameter related to data transmission;

[0028] calculating the communication network state according to the parameter related to data transmission.

[0029] In some embodiments, the source decision result comprises a resolution and a sending frame rate, and the channel decision result comprises a bandwidth allocation and a power allocation of data transmission.

[0030] In a second aspect, the embodiments of the present application further provide a data transmission control device, comprising:

[0031] an acquisition module, configured to acquire a source decision state and a channel decision state of to-be-transmitted data;

[0032] a first processing module, configured to obtain a source decision result according to the source decision state, the source decision result being used to control a source device to transmit the to-be-transmitted data, the source device being used to generate the to-be-transmitted data;

[0033] a second processing module, configured to obtain a channel decision result according to the channel decision state, the channel decision result being used to control a channel device to transmit the to-be-transmitted data.

[0034] In a third aspect, the embodiments of the present application further provide a data transmission control device, comprising a memory, a processor and a program stored in the memory and executable on the processor; the processor implements the steps in the data transmission control method according to the first aspect when executing the program.

[0035] In a fourth aspect, the embodiments of the present application further provide a computer readable storage medium, which stores a program, the program being executable on a processor to implement the steps in the data transmission control method according to the first aspect.

[0036] In a fifth aspect, the embodiments of the present application further provide a computer program product, comprising computer instructions, the computer instructions being executable on a processor to implement the steps in the data transmission control method according to the first aspect.

[0037] The above technical solutions of the present application have the following beneficial effects:

[0038] In the above solution, the source decision state and the channel decision state of to-be-transmitted data are acquired; then, a source decision result is obtained according to the source decision state, the source decision result being used to control a source device to transmit the to-be-transmitted data, the source device being used to generate the to-be-transmitted data; a channel decision result is obtained according to the channel decision state, the channel decision result being used to control a channel device to transmit the to-be-transmitted data; in this way, the above processing can achieve timely response to user operation and improve user experience. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 FIG. 1 is one of flowcharts of the data transmission control method according to the embodiments of the present application;

[0040] Figure 2 FIG. 2 is an example diagram of information freshness change over time according to the embodiments of the present application.

[0041] Figure 3 This is a second schematic flowchart of the data transmission control method according to an embodiment of the present invention;

[0042] Figure 4 This is a schematic diagram of the data transmission control device according to an embodiment of the present invention;

[0043] Figure 5 This is a schematic diagram of the hardware structure of the data transmission control device according to an embodiment of the present invention. Detailed Implementation

[0044] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0045] To facilitate understanding, the relevant knowledge involved in this invention will be introduced first.

[0046] 1. Network-industry collaboration

[0047] Network-business collaboration is a new collaborative work model that emphasizes deep integration and real-time interaction between the network and service sides. In this model, the network can perceive service needs in real time and dynamically adjust and optimize network resources accordingly to achieve a precise match between network capabilities and service requirements. Through cross-layer and cross-domain collaborative processing, network-business collaboration not only improves the flexibility and efficiency of services but also optimizes resource allocation and utilization, making the entire system more intelligent and efficient. In the rapidly developing internet era, network-business collaboration is becoming an important means for various industries to achieve digital transformation and intelligent upgrading.

[0048] 2. Information Age

[0049] Information age (AoI) is a concept used to measure the freshness of information, especially in real-time communication and data processing systems. Below is a clear introduction to the concept of information age (AoI):

[0050] Information age (AoI) refers to the time interval from when information is generated by a sensor or data source to when that information is successfully received and used by the target receiver. This differs significantly from latency, which typically refers to the time interval between data transmission and reception. AoI focuses on the time interval between data generation and actual usage, thus better reflecting the freshness of the information. In some classic system models, such as the M / M / 1 queue system, system latency monotonically increases as customer arrival rate (i.e., data generation rate) increases. However, AoI behaves differently; it may initially drop to a minimum and then gradually increase with increasing system load.

[0051] AoI can be expressed mathematically as: A(t) = tg(t), where t represents the current time and g(t) represents the generation time (also known as the timestamp of the generation time) of the latest data packet received by the receiver.

[0052] Information age (AoI) is particularly important in applications with high real-time requirements, such as real-time data transmission for cloud gaming, remote monitoring, and autonomous driving. In these scenarios, keeping information fresh is crucial for making accurate decisions.

[0053] In summary, Information Age (AoI) is a concept used to measure the freshness of information, focusing on the time interval between data generation and use. Unlike latency, AoI better reflects the real-time nature and freshness of information within a system. In applications with high real-time requirements, optimizing AoI is crucial for improving system performance and decision accuracy. Applying Information Age to cloud gaming can effectively measure the timeliness between user actions and system feedback, providing a more direct reflection of user experience.

[0054] 3. Strengthen learning

[0055] Reinforcement learning (RL) is an important branch of machine learning. Its core is to solve the problem of how an agent can learn strategies to maximize rewards or achieve specific goals in the process of interacting with the environment.

[0056] Reinforcement learning is a method that allows an agent to learn and optimize its behavioral policies through interactions with its environment. It does not require pre-provided data; instead, it updates model parameters by receiving rewards (feedback) from the environment for actions.

[0057] Reinforcement learning is based on a "trial and error" approach. An agent evaluates the quality of its actions by performing them, observing changes in the environment, and receiving reward signals, and adjusts its subsequent behavioral strategies accordingly. The goal of reinforcement learning is to maximize the agent's cumulative reward in the long run.

[0058] Algorithms used to solve reinforcement learning problems can be divided into two categories: policy search algorithms and value function algorithms. Deep learning models can be used in reinforcement learning, forming deep reinforcement learning.

[0059] Reinforcement learning focuses on online learning and attempts to maintain a balance between exploration and exploitation. It does not require any pre-given data; instead, it learns by receiving rewards (feedback) from the environment for actions and updates its model parameters.

[0060] In summary, reinforcement learning is a method of learning and optimizing behavior strategies through the interaction of agents and the environment, which has shown great potential in solving complex decision-making problems.

[0061] In the prior art, network industry cooperation is usually based on the correlation between various business experience indicators and network performance indicators to predict the quality of the target business experience, but the existing business experience indicators are difficult to directly measure the response and timeliness of user operation, and there is a certain deviation in measuring user experience.

[0062] To solve the above technical problems, the present application provides a data transmission control method, device, storage medium and program product, wherein the method and device are based on the same application concept, and the implementation of the device and the method can be referred to each other because the principles of the device and the method for solving problems are similar, and the repeated parts will not be described again.

[0063] As shown in the flowchart of the data transmission control method provided by the embodiment of the present application. Figure 1 As shown in the flowchart of the data transmission control method provided by the embodiment of the present application.

[0064] Step 101, obtaining the source decision state and channel decision state of the to-be-transmitted data;

[0065] The to-be-transmitted data is generated by the source device. It can be that the user equipment side sends a data request corresponding to the user operation to the source device through user operation, and the source device generates to-be-transmitted data corresponding to the user operation based on the data request. For example, in the cloud game scenario, the user equipment (such as a terminal) sends a data request corresponding to the user operation to the corresponding source device (such as a server) through user operation on the game interface, and the source device generates to-be-transmitted data corresponding to the user operation based on the data request.

[0066] Here, the source decision state of the to-be-transmitted data provides a basis for obtaining the source decision result to control the transmission of the to-be-transmitted data by the source device. The source decision state includes the current communication network state and the historical channel decision result. Here, the historical channel decision result includes the bandwidth allocation and power allocation during the historical data transmission.

[0067] The channel decision state of the to-be-transmitted data provides a basis for obtaining the channel decision result to control the transmission of the to-be-transmitted data by the channel device. The channel decision state includes the current communication network state and the historical source decision result. The historical source decision result includes the corresponding resolution and transmission frame rate during the historical data transmission.

[0068] At step 102, a source decision result is obtained according to the source decision state, the source decision result being used to control transmission of the to-be-transmitted data by a source device, the source device being used to generate the to-be-transmitted data.

[0069] Optionally, the source decision result includes a resolution and a sending frame rate. That is, the data is finally presented to the user device side in the resolution and the sending frame rate in the source decision result. Before that, the source device adjusts the resolution and the sending frame rate of the to-be-transmitted data according to the source decision result.

[0070] At step 103, a channel decision result is obtained according to the channel decision state, the channel decision result being used to control transmission of the to-be-transmitted data by a channel device.

[0071] Optionally, the channel decision result includes a bandwidth allocation and a power allocation of data transmission. That is, the channel device transmits the to-be-transmitted data to the user device side according to the bandwidth allocation and the power allocation of data transmission in the channel decision result.

[0072] In an optional embodiment, the step 102 of obtaining the source decision result according to the source decision state includes:

[0073] inputting the source decision state into an intelligent decision network to obtain the source decision result;

[0074] It should be understood that the intelligent decision network is a reinforcement learning network. After the source decision state is input into the intelligent decision network, the intelligent decision network performs source decision according to the source decision state to obtain the source decision result corresponding to the source decision state.

[0075] The step 103 of obtaining the channel decision result according to the channel decision state includes:

[0076] inputting the channel decision state into an intelligent decision network to obtain the channel decision result;

[0077] Here, after the channel decision state is input into the intelligent decision network, the intelligent decision network performs channel decision according to the channel decision state to obtain the channel decision result corresponding to the channel decision state.

[0078] Here, the source decision state and the channel decision state can be understood as an environment state related to data transmission. As known from the foregoing, the source decision state includes a current communication network state and a historical channel decision result. The channel decision state includes a current communication network state and a historical source decision result. The current communication network state reflects a network environment state of current data transmission, and the historical channel decision result and the historical source decision result reflect historical data transmission conditions, which can also be used as a reference for the transmission environment of current data transmission to some extent.

[0079] The intelligent decision network comprises an information value function of a user, the information value function is related to an information age of the data to be transmitted and a quality of service corresponding to the data to be transmitted, and the information value function is used to adjust a network parameter of the intelligent decision network.

[0080] Here, the information value function is a function used to measure the value of the data to be transmitted. Here, the information value function being related to the information age of the data to be transmitted and the quality of service corresponding to the data to be transmitted specifically means that the information value function is composed of a function representing the information age of the data to be transmitted and a function constructed based on the quality of service corresponding to the data to be transmitted.

[0081] For example, the data to be transmitted is video frame data in a cloud game scenario, and the information value function is related to the information age of the data to be transmitted and the video quality of the data to be transmitted, so as to measure the value of the video frame data in the cloud game scenario by capturing the information age (freshness) of the data to be transmitted and the video quality of the data to be transmitted.

[0082] Since the intelligent decision network has information value orientation, the information value includes real-time and data effectiveness, the information age (freshness) is considered in real-time, and the quality of service corresponding to the data is considered in data effectiveness, and the two points are taken as the measurement indexes of the data transmission value, the data to be transmitted is processed through the intelligent decision network with the above characteristics, and timely response to user operation can be achieved, and user experience is improved.

[0083] In an optional embodiment, the step 101 comprises:

[0084] Step 1011, obtaining a communication network state;

[0085] Optionally, the step 1011 specifically comprises:

[0086] obtaining a parameter related to data transmission;

[0087] According to the parameter related to data transmission, the communication network state is calculated.

[0088] Optionally, the parameter related to data transmission comprises a parameter related to a data transmission rate. The communication network state is calculated according to the parameter related to the data transmission rate.

[0089] Here, the parameter related to the data transmission rate can be a data transmission rate mean R i , wherein:

[0090]

[0091] wherein, B i represents the bandwidth allocated to the i-th user, h i is the time-varying channel gain, P i represents the transmission power allocated to the i-th user, and N0 represents the power spectral density of the Gaussian white noise.

[0092] Here, the communication network state can be embodied by the transmission delay of each data packet, wherein the transmission delay of each data packet is is represented as:

[0093]

[0094] wherein, L T represents the data packet size, R i represents the average data transmission rate.

[0095] Step 1012, obtaining the source decision state according to the communication network state and the historical channel decision result;

[0096] Specifically, the historical channel decision result can be the channel decision result obtained in the last data transmission, and the current source decision state is obtained by combining the communication network state and the channel decision result obtained in the last data transmission (the bandwidth allocation and power allocation of the last data transmission). The channel decision result obtained in the last data transmission is stored in a memory, and the channel decision result obtained in the last data transmission is obtained by reading the memory.

[0097] Step 1013, obtaining the channel decision state according to the communication network state and the historical source decision result.

[0098] Specifically, the historical source decision result can be the source decision result obtained in the last data transmission, and the current channel decision state is obtained by combining the communication network state and the source decision result obtained in the last data transmission (the resolution and transmission frame rate of the last transmitted data). The source decision result obtained in the last data transmission is stored in a memory, and the source decision result obtained in the last data transmission is obtained by reading the memory.

[0099] It should be noted that the intelligent decision network described above is a reinforcement learning network, that is, the present application is a source and channel joint optimization scheme based on reinforcement learning, which considers the correlation between source decision and channel decision. In order to make full use of network resources as much as possible and guarantee the service demand of data transmission, the intelligent decision network needs to be continuously optimized, and the foregoing part also mentioned that reinforcement learning does not need to be given any data in advance, but obtains learning information and updates model parameters by receiving the reward (feedback) of the environment to the action, so the optimization of the intelligent decision network of the present application is also based on this principle. How to optimize, see the following examples.

[0100] In an optional embodiment, the method of the present application further comprises:

[0101] 1) sending the source decision result to the source device; and obtaining a first execution result of the source device performing the data transmission to be transmitted based on the source decision result;

[0102] Wherein, the data transmission control device issues the source decision result to the source device, and the source device performs the data transmission to be transmitted based on the source decision result, and returns the first execution result to the data transmission control device. That is, the source device transmits the data to be transmitted after adjusting the resolution and transmission frame rate according to the source decision result, obtains the actual transmission situation, i.e. the first execution result, and then returns the first execution result to the data transmission control device.

[0103] 2) sending the channel decision result to the channel device; and obtaining a second execution result of the channel device performing the data transmission to be transmitted based on the channel decision result.

[0104] Wherein, the data transmission control device issues the channel decision result to the channel device, and the channel device performs the data transmission to be transmitted based on the channel decision result, and returns the second execution result to the data transmission control device. That is, the channel device transmits the data to be transmitted output from the source device according to the bandwidth allocation and power allocation in the channel decision result (at this time, the resolution and transmission frame rate of the data to be transmitted have been adjusted), obtains the actual transmission situation, i.e. the second execution result, and then returns the second execution result to the data transmission control device.

[0105] It should be understood that the first execution result and the second execution result are the rewards (feedback) of the environment to the action in reinforcement learning.

[0106] 3) calculating a first information value using the information value function according to the first execution result and the second execution result.

[0107] 4) adjusting the network parameters of the intelligent decision network using the first information value to obtain an adjusted intelligent decision network.

[0108] Here, the first execution result and the second execution result are the rewards of the environment to the action in reinforcement learning, and the reward value (i.e. the first information value) is calculated by the information value function. Finally, the first information value (i.e. the reward value) is used to adjust the network parameters of the intelligent decision network to obtain an adjusted intelligent decision network, which is used for the next data transmission. The above entire process is a process of optimizing the intelligent decision network.

[0109] It should be noted that before the first data transmission, the initialization of the network parameters of the intelligent decision network, the source decision result and the channel decision result is completed.

[0110] The optimization of the intelligent decision network cannot be separated from the information value function (i.e. reward function), which is related to the information age of the data to be transmitted and the quality of service corresponding to the data to be transmitted. How to obtain the information value function, see the following examples.

[0111] In an optional embodiment, the method of the present application further comprises:

[0112] Based on the probability density function of the data packet arrival in the data to be transmitted and the probability density function of the data transmission rate, the information age of the data to be transmitted is obtained;

[0113] According to the information age of the data to be transmitted and the service quality modeling function, the information value function is obtained.

[0114] Here, the information age of the data to be transmitted is determined by the delay of preprocessing and transmission and the queuing delay. The purpose of the service quality modeling function is to ensure the intelligibility of the data. The intelligibility of the data is determined by the compression rate of the data. Accordingly, the information value (VoI) metric is designed and the minimization problem is proposed to optimize the data transmission efficiency of task-oriented communication.

[0115] 1) Expression about information age

[0116] According to the factors affecting the information age (information freshness) of data in task-oriented communication, the mathematical model about information age in VoI is involved, and the closed-form expression is deduced. See Figure 2 , an example diagram of the change of information freshness with time, for illustration.

[0117] For ( representing a positive integer) set t j representing the time when the jth transmission request is generated, representing the time when the jth data packet extraction and compression related calculation is completed (in the data to be transmitted), representing the time when the jth data packet transmission is completed, representing the total duration of data packet calculation, representing the total duration of data packet transmission, X j = t j -t j-1 (j>1) representing the time interval of data packet arrival, representing the total duration of data packet processing, representing the waiting duration of calculation, the waiting time of the representative transmission, the computation time of the representative data packet, the transmission time of the representative data packet, Δ(0) represents the initial value of the system AoI (age of information).

[0118] In the link-level model of communication, information processing and data transmission are two independent processes. The source device (node) responsible for generating data to be transmitted (such as cloud game video generation) first performs local computation tasks such as video compression, and then transmits data packets to the channel for end users to use. That is, the i-th data packet does not generate waiting time due to the incomplete transmission of the i-1-th data packet.

[0119] Here, the AoI in the communication system is analyzed using geometric analysis. The average information age is Figure 2 the area under the sawtooth function, denoted as:

[0120]

[0121] where, denotes the average information age, denotes the area marked as q1 in the figure, X j denotes the time interval of the arrival of data packet j, T j denotes the total processing time of data packet j, T n denotes the total processing time of data packet n, denotes the total transmission time of data packet n.

[0122] Therefore, the AoI of the data to be transmitted (such as video data packets) of user i can be represented as:

[0123]

[0124] where, E[XT] represents the mean of the product of the time interval of the arrival of data packets and the processing time of data packets, E[X 2 ] represents the mean of the square of the time interval of the arrival of data packets.

[0125] 2) Quality of service modeling (if it is for cloud game scenario, it corresponds to video quality modeling in cloud game)

[0126] terminal video resolution F task is mainly affected by the video compression rate F com and the influence of the bit error rate:

[0127] F task (θ, SE) = F com (θ) · F trans (SE)

[0128] where, Ftask F represents the task performance in task-oriented communication. com F represents the effect of compression ratio. trans This represents the impact of bit error rate. Where θ represents compression ratio and SE represents bit error rate.

[0129] Based on the information age and service quality modeling function of the data to be transmitted obtained above, the process of obtaining the information value function can be described as follows:

[0130] The goal of intelligent decision-making network optimization is to allocate channel resources under constraints to transmit preprocessed compressed video data, achieving optimal task performance by transmitting the most valuable information. This can be represented as a joint source-channel optimization problem. The system allocates channel resources based on channel conditions; each source device makes source decisions based on the allocated channel resources, deciding on different source compression rates to maximize the current data transmission value. Task-oriented communication transmission is achieved through joint source-channel optimization. The joint source-channel optimization problem (i.e., the information value function) can be modeled as:

[0131]

[0132] Where ω1 represents the first weight value and ω2 represents the second weight value; F represents the average information age of terminal (user) i. i task P0 represents the video resolution of terminal (user) i. The result indicates that B represents the bandwidth and D represents the transmission time interval.

[0133] The constraints include:

[0134] C1: Bandwidth meets resource constraints;

[0135] C2: The impact of compression ratio is fixed to ensure that the requirements of terminal cloud gaming users are met.

[0136] C3: The impact of the bit error rate is fixed to ensure that the requirements of terminal cloud gaming users are met.

[0137] C4: The total processing latency of data packets in the system should be less than a fixed value to ensure that cloud gaming latency requirements are met.

[0138] C5:S C ≤D, the calculation time does not exceed the data packet generation interval, to ensure ρ C ≤1;

[0139] C6: The average transmission time of each data packet does not exceed the data packet generation interval, so as to ensure that T ≤1;

[0140] C7: The transmission power satisfies the resource restriction.

[0141] The application also provides a data transmission control system, i.e., a source-channel joint control system of network cooperation, which is shown in Figure 3 The system includes three modules: a source module, a channel module, and an intelligent decision module. Optionally, the system includes n source modules, wherein n is the number of users, one channel module, and one intelligent decision module. The flow of the data transmission control method based on the system is as follows:

[0142] 1. Initialize the network parameters of the intelligent decision module, the source decision result, and the channel decision result.

[0143] 2. Obtain the current communication network state and the current source decision state, wherein the current source decision state is obtained by the intelligent decision module according to the current communication network state and the channel decision result obtained in the last data transmission (the channel decision result of the first round of decision is obtained by initialization in 1).

[0144] The current communication network state is calculated according to the data transmission rate related parameters.

[0145] 3. The information value function obtained by deducing the information freshness expression and the video quality modeling function is used for the decision of the intelligent decision module, and the current source decision result (resolution and transmission frame rate) is obtained by the intelligent decision module according to the current source decision state and the network parameters.

[0146] 4. The current source decision is executed by the source module according to the current source decision result, the resolution and the transmission frame rate of each user are adjusted, and the cloud game data to be transmitted is input into the cache queue.

[0147] After the source module executes the source decision, the execution result is returned to the intelligent decision module for step 8.

[0148] 5. Obtain the current communication network state and the current channel decision state, wherein the current channel decision state is obtained by the intelligent decision module according to the current communication network state and the source decision result obtained in the last data transmission (the source decision result of the first round of decision is obtained by initialization in 1).

[0149] 6. An information value function obtained by deriving an information freshness expression and a video quality modeling function, used for decision of the intelligent decision module, wherein the intelligent decision module obtains a current channel decision result according to a current channel decision state and network parameters of the intelligent decision module, and the current channel decision result includes bandwidth allocation and power allocation of data transmission.

[0150] 7. The channel module executes the current channel decision according to the current channel decision result, adjusts power allocation and bandwidth allocation of each user, and transmits the to-be-transmitted data content.

[0151] 8. The current source decision result and the channel decision result are stored and the network parameters of the intelligent decision module are updated, and after the source decision result is stored, the step 5 is performed, and after the channel decision result is stored, the step 2 is performed.

[0152] The optimization target of the intelligent decision module is described in the above source channel joint optimization problem modeling part.

[0153] The network parameters of the intelligent decision module include but are not limited to weights and biases in the intelligent decision network (i.e. artificial intelligence network).

[0154] Through optimization of the intelligent decision network, coding efficiency can be improved, channel resources can be fully utilized, the most valuable information for the receiving end (user equipment side) can be transmitted, and user experience can be improved.

[0155] As shown in Figure 4 The embodiment of the present application also provides a data transmission control device 400, wherein the device comprises:

[0156] An acquisition module 401 is configured to acquire a source decision state and a channel decision state of to-be-transmitted data.

[0157] A first processing module 402 is configured to obtain a source decision result according to the source decision state, wherein the source decision result is used to control transmission of the to-be-transmitted data by a source device, and the source device is used to generate the to-be-transmitted data.

[0158] A second processing module 403 is configured to obtain a channel decision result according to the channel decision state, wherein the channel decision result is used to control transmission of the to-be-transmitted data by a channel device.

[0159] In some embodiments, the acquisition module 401 comprises:

[0160] A first acquisition unit is configured to acquire a communication network state.

[0161] A second acquisition unit is configured to obtain the source decision state according to the communication network state and a historical channel decision result.

[0162] The third obtaining unit is configured to obtain the channel decision state according to the communication network state and the historical source decision result.

[0163] In some embodiments, the first processing module 402 includes:

[0164] The first processing unit is configured to input the source decision state into an intelligent decision network to obtain the source decision result.

[0165] The second processing module 403 includes:

[0166] The second processing unit is configured to input the channel decision state into an intelligent decision network to obtain the channel decision result.

[0167] The intelligent decision network includes an information value function of a user, the information value function is related to an information age of the to-be-transmitted data and a quality of service corresponding to the to-be-transmitted data, and the information value function is used to adjust network parameters of the intelligent decision network.

[0168] In some embodiments, the apparatus further includes:

[0169] The first transmission module is configured to send the source decision result to the source device and obtain a first execution result of the to-be-transmitted data transmission performed by the source device based on the source decision result.

[0170] The second transmission module is configured to send the channel decision result to the channel device and obtain a second execution result of the to-be-transmitted data transmission performed by the channel device based on the channel decision result.

[0171] The computing module is configured to calculate a first information value by using the information value function according to the first execution result and the second execution result.

[0172] The network optimization module is configured to adjust the network parameters of the intelligent decision network by using the first information value to obtain an adjusted intelligent decision network.

[0173] In some embodiments, the apparatus further includes:

[0174] The third processing module is configured to obtain the information age of the to-be-transmitted data based on a probability density function of a data packet arrival in the to-be-transmitted data and a probability density function of a data transmission rate.

[0175] The fourth processing module is configured to obtain the information value function according to the information age of the to-be-transmitted data and a quality of service modeling function.

[0176] In some embodiments, the first obtaining unit is specifically configured to:

[0177] obtaining a parameter related to data transmission;

[0178] calculating the communication network state according to the parameter related to data transmission.

[0179] In some embodiments, the source decision result includes resolution and transmission frame rate; and the channel decision result includes bandwidth allocation and power allocation of data transmission.

[0180] The implementation embodiments of the data transmission control method described above are applicable to the embodiments of the data transmission control device, and can achieve the same technical effects.

[0181] As shown in Figure 5 The embodiments of the present application also provide a data transmission control device 500, which comprises a memory 502, a processor 501 and a program stored in the memory 502 and executable on the processor 501; the processor 501 implements the steps of the data transmission control method described above when executing the program.

[0182] The implementation embodiments of the data transmission control method described above are applicable to the embodiments of the data transmission control device, and can achieve the same technical effects.

[0183] The embodiments of the present application also provide a computer readable storage medium, which stores a program; the program is executed by a processor to implement the steps of the data transmission control method described above.

[0184] The implementation embodiments of the data transmission control method described above are applicable to the embodiments of the computer readable storage medium, and can achieve the same technical effects.

[0185] The embodiments of the present application also provide a computer program product, which comprises computer instructions; the computer instructions are executed by a processor to implement the processes of the method embodiments described above, and can achieve the same technical effects; to avoid repetition, details are not described here.

[0186] It should be noted that many functional components described in this specification are referred to as modules, in order to emphasize the independence of their implementation.

[0187] In the embodiments of the present application, the modules can be implemented in software, and executed by various types of processors. For example, an identified module of executable code can consist of one or more physical or logical blocks of computer instructions. For example, a module can be implemented in object-oriented programming, such as a Java® class, or in procedural programming, such as an OpenCL® kernel. Generally, a module of executable code can be implemented in a variety of ways, including procedureally, in an object-oriented language, and / or in a visually-oriented programming language that transmits or receives data or control signals. Although the modules of executable code are illustrated as single blocks in the flowcharts, the modules of executable code can be stored on or transmitted across various types of signals, and can include one or more instructions for implementing data storage and transmission of signals. In some embodiments, the modules of executable code can be stored on a computer-readable medium, which can be a non-transitory computer-readable medium, such as a floppy disk, a hard disk, a CD-ROM, a DVD, a Blu-ray Disc™, a RAM, a ROM, a PROM, an EPROM, a EEPROM, a SRAM, as well as a Flash memory, a register, or any other suitable memory component. The computer-readable medium can be a non-transitory computer-readable medium. In some embodiments, the modules of executable code can be transmitted across a variety of signal-bearing media, including wires, cables, communication networks, wireless networks, and / or computer program products.

[0188] In fact, the modules of executable code can be a single instruction, or many instructions, and can even be distributed over several different code segments, among different programs, and across several memory devices. Also, operational data can be identified within the modules, and can be expressed in a data structure. The operational data can be collected as a single data set, or distributed in different locations including over different storage devices, and can exist even only as manifestative as electronic signals on a system or network.

[0189] When the modules can be implemented in software, the modules of executable code, in view of the level of the existing hardware technology, can be built into corresponding hardware circuitry by those skilled in the art without considering the cost, and the hardware circuitry includes conventional very large scale integration (VLSI) circuitry or gate array, and existing semiconductors such as logic chips, transistors, and other discrete elements. The modules can also be implemented by programmable hardware devices, such as field programmable gate array, programmable array logic, programmable logic device, and the like.

[0190] The above is the preferred embodiment of the present application, and it should be pointed out that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A data transmission control method, characterized in that, include: Obtain the source decision state and channel decision state of the data to be transmitted; Based on the source decision state, a source decision result is obtained. The source decision result is used to control the source device to transmit the data to be transmitted. The source device is used to generate the data to be transmitted. Based on the channel decision state, a channel decision result is obtained. The channel decision result is used to control the channel device to transmit the data to be transmitted. The source decision result includes resolution and transmission frame rate. The channel decision results include bandwidth allocation and power allocation for data transmission; The acquisition of the source decision state and channel decision state of the data to be transmitted includes: Obtain the status of the communication network; The source decision state is obtained based on the communication network status and historical channel decision results; The channel decision state is obtained based on the communication network status and historical source decision results; The step of obtaining the source decision result based on the source decision state includes: The source decision state is input into the intelligent decision network to obtain the source decision result; The step of obtaining the channel decision result based on the channel decision state includes: The channel decision state is input into the intelligent decision network to obtain the channel decision result; The intelligent decision-making network includes a user's information value function, which is related to the information age of the data to be transmitted and the service quality corresponding to the data to be transmitted. The information value function is used to adjust the network parameters of the intelligent decision-making network.

2. The method according to claim 1, characterized in that, The method further includes: Send the source decision result to the source device; and obtain the first execution result of the source device executing the data transmission to be transmitted based on the source decision result; Send the channel decision result to the channel device; and obtain the second execution result of the channel device executing the data transmission to be transmitted based on the channel decision result; Based on the first execution result and the second execution result, the first information value is calculated using the information value function; The network parameters of the intelligent decision-making network are adjusted using the first information value to obtain the adjusted intelligent decision-making network.

3. The method according to claim 1, characterized in that, The method further includes: Based on the probability density function of the arrival of data packets and the probability density function of the data transmission rate in the data to be transmitted, the information age of the data to be transmitted is obtained. The information value function is obtained based on the information age and service quality modeling function of the data to be transmitted.

4. The method according to claim 1, characterized in that, The step of obtaining the communication network status includes: Obtain parameters related to data transmission; The communication network status is calculated based on the parameters related to data transmission.

5. A data transmission control device, comprising a memory, a processor, and a program stored in the memory and executable on the processor; characterized in that, When the processor executes the program, it implements the steps in the data transmission control method as described in any one of claims 1 to 4.

6. A computer-readable storage medium having a program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the data transmission control method as described in any one of claims 1 to 4.

7. A computer program product, characterized in that, It includes computer instructions that, when executed by a processor, implement the steps in the data transmission control method as described in any one of claims 1 to 4.

Images