Data transmission method and device, equipment, storage medium and product

Abstract

The invention discloses a data transmission method and device, equipment, a storage medium and a product, and relates to the technical field of information security, and the method comprises the steps: obtaining an original data packet and a current network state; embedding and packaging the original data packet to obtain a to-be-transmitted data packet; predicting according to the current network state and the to-be-transmitted data packet through the data transmission model to obtain a target encapsulation level, a target traffic shaping parameter and a sending window parameter; sending the to-be-transmitted data packet based on the target encapsulation level, the target traffic shaping parameter and the sending window parameter; according to the method, the current data packet is embedded and packaged, and the network state and the application requirement of the packaged data packet are analyzed in real time according to the data transmission model to obtain the target packaging level, the target traffic shaping parameter and the sending window parameter, so that the optimal packaging strategy, the traffic shaping parameter and the congestion control strategy are dynamically selected; flexible adjustment can be performed according to actual network conditions, and communication performance and resource utilization rate are improved.

Description

Technical Field

[0001] The present application relates to the field of information security technology, and in particular to data transmission methods, devices, equipment, storage media and products. Background Art

[0002] With the development of information technology and changes in the network security situation, traditional virtual private network (VPN) solutions can no longer meet the high security and flexibility requirements of modern distributed and remote office environments. Especially in the context of the concept of zero-trust network architecture gradually becoming an industry consensus, SDP zero-trust, as a new type of security access mode, effectively makes up for the many shortcomings of VPN in dealing with internal threats and mobile office scenarios with its identity-based dynamic access control, the principle of least privilege, and stealth network architecture. In recent years, with the growth of cloud computing, the Internet of Things, and remote office trends, network boundaries have become increasingly blurred, and enterprises have significantly increased their security requirements for data protection and remote access. With its zero-trust model, SDP technology can ensure that only authorized and healthy terminals and users can access specific resources, thereby reducing the risk of internal leakage and external attack surface.

[0003] The communication between the client and the controller, and between the client and the gateway generally uses the UDP protocol as the transport layer protocol. In this technical solution, the client sends and receives data through the UDP protocol, taking advantage of its connectionless, low latency, and high transmission efficiency. It is particularly suitable for application scenarios with high real-time requirements and tolerance to a certain degree of data loss. In the specific implementation, the client does not need to establish a connection in advance to send data, and the UDP protocol allows broadcasting and multicasting, which facilitates one-to-many or many-to-many data transmission. The communication between the client and the controller, and between the client and the gateway usually relies solely on the UDP protocol, taking advantage of its connectionless, low latency, and fast transmission speed. However, this solution has significant limitations: the UDP protocol does not provide a sequential delivery and confirmation mechanism for data packets, nor can it automatically handle network congestion and data packet loss, resulting in serious impacts on communication performance and security in scenarios with high requirements for data integrity and reliability. Summary of the invention

[0004] The main purpose of this application is to provide a data transmission method, device, equipment, storage medium and product, aiming to solve the technical problem that in scenarios with high requirements for data integrity and reliability, network congestion causes serious impact on the communication performance and security of data transmission.

[0005] To achieve the above object, the present application proposes a data transmission method, which includes:

[0006] Get the original data packet and current network status;

[0007] Embed and encapsulate the original data packet to obtain the data packet to be transmitted;

[0008] The target encapsulation level, target traffic shaping parameters and sending window parameters are predicted based on the current network status and the data packets to be transmitted by a data transmission model;

[0009] The data packet to be transmitted is sent based on the target encapsulation level, the target traffic shaping parameter, and the sending window parameter.

[0010] In one embodiment, the data transmission model includes a network state estimation model and a congestion window control model;

[0011] The method of predicting a target encapsulation level, a target traffic shaping parameter, and a sending window parameter according to the current network state and the data packet to be transmitted by using a data transmission model includes:

[0012] Obtain a predicted network state according to the network state prediction model and the current network state, and obtain a target encapsulation level according to the predicted network state;

[0013] Acquire a real-time data transmission rate, and obtain a target traffic shaping parameter according to the real-time data transmission rate, a preset rate threshold, and a real-time capacity of the network;

[0014] The network congestion status is acquired, and a sending window parameter is obtained according to the congestion window control model and the network congestion status.

[0015] In one embodiment, the acquiring of the network congestion status and obtaining the sending window parameter according to the congestion window control model and the network congestion status include:

[0016] Get network congestion status;

[0017] Obtaining a reward value of each sending window parameter at the next moment according to the network congestion state, and obtaining a target reward value according to the reward value;

[0018] A sending window parameter is obtained according to the network congestion status and the target reward value.

[0019] In one embodiment, before the data transmission model is used to predict the target encapsulation level, the target traffic shaping parameter and the sending window parameter according to the current network state and the data packet to be transmitted, the method further includes:

[0020] Build the initial transfer model and obtain training data;

[0021] Obtain a state space according to the network state in the training data, and obtain an action space according to the encapsulation level, traffic shaping parameters, and sending window parameters in the training data;

[0022] Construct a reward function based on the preset network communication goals;

[0023] The initial transmission model is trained according to the state space, the action space and the reward function to obtain a data transmission model.

[0024] In one embodiment, the training of the initial transmission model according to the state space, the action space, and the reward function to obtain the data transmission model includes:

[0025] Obtaining the state space and current reward at the next moment according to the state space, the action space and the reward function;

[0026] The state space, the current action space, the state space at the next moment, and the current reward are used as a training data set;

[0027] Back propagation training is performed based on the training data group to optimize the weight parameters of the initial transmission model to obtain a data transmission model.

[0028] In one embodiment, embedding and encapsulating the original data packet to obtain the data packet to be transmitted includes:

[0029] Converting the format of the original data packet to obtain a formatted data packet;

[0030] Obtaining a unique serial number of the original data packet according to a preset serial number mechanism;

[0031] Encapsulating the formatted data packet to obtain an encapsulated data packet, and obtaining a checksum according to the encapsulated data packet;

[0032] A data packet to be transmitted is obtained according to the encapsulated data packet, the unique sequence number and the checksum.

[0033] In addition, to achieve the above purpose, the present application also proposes a data transmission device, the data transmission device comprising:

[0034] The acquisition module is used to obtain the original data packet and the current network status;

[0035] The acquisition module is further used to embed and encapsulate the original data packet to obtain the data packet to be transmitted;

[0036] A parameter prediction module, used to predict the target encapsulation level, target traffic shaping parameters and sending window parameters according to the current network state and the data packet to be transmitted through a data transmission model;

[0037] A data transmission module is used to send the data packet to be transmitted based on the target encapsulation level, the target traffic shaping parameter and the sending window parameter.

[0038] In addition, to achieve the above objectives, the present application also proposes a data transmission device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the data transmission method described above.

[0039] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the data transmission method described above are implemented.

[0040] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps of the data transmission method described above are implemented.

[0041] One or more technical solutions proposed in the present application have at least the following technical effects: by embedding and encapsulating the current data packet, the reliability of the TCP protocol is combined with the high efficiency of the UDP protocol, and the UDP protocol is used to ensure data integrity, sequentiality and effective utilization of network resources, and then the network status, application requirements and other relevant parameters are analyzed in real time according to the data transmission model, and the optimal encapsulation strategy, traffic shaping parameters and congestion control strategy are dynamically selected, which can make flexible adjustments according to actual network conditions to improve communication performance and resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0043] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0044] Figure 1 A flowchart of the first embodiment of the data transmission method of the present application is provided;

[0045] Figure 2 A schematic diagram of the structure of the intelligent encapsulation TCP protocol system provided in Embodiment 1 of the data transmission method of the present application;

[0046] Figure 3A schematic diagram of the structure between various application modules provided in Example 1 of the data transmission method of the present application;

[0047] Figure 4 A schematic diagram of the embedding and encapsulation process of the original data packet provided in Embodiment 1 of the data transmission method of the present application;

[0048] Figure 5 A schematic diagram of the execution steps of the AI ​​decision module provided in Embodiment 1 of the data transmission method of the present application;

[0049] Figure 6 A schematic diagram of the training steps of the initial transmission model provided in Embodiment 1 of the data transmission method of the present application;

[0050] Figure 7 A schematic diagram of the execution steps of the security threat identification and defense module provided in the first embodiment of the data transmission method of the present application;

[0051] Figure 8 A schematic diagram of a data transmission method flow based on a security threat identification and defense module provided in Embodiment 1 of the data transmission method of the present application;

[0052] Fig. 9 A flowchart of the second embodiment of the data transmission method of the present application is provided;

[0053] Fig.10 A schematic diagram of the execution steps of the traffic shaping and optimization module provided in Embodiment 2 of the data transmission method of the present application;

[0054] Fig.11 A schematic diagram of specific execution steps of the congestion control module provided in Embodiment 2 of the data transmission method of the present application;

[0055] Fig.12 This is a schematic diagram of the module structure of the data transmission device according to an embodiment of the present application;

[0056] Fig.13 Schematic diagram of the device structure of the hardware operating environment involved in the data transmission method in the embodiment of the present application.

[0057] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0058] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0059] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0060] The main solution of the embodiment of the present application is: obtain the original data packet and the current network status; embed and encapsulate the original data packet to obtain the data packet to be transmitted; predict the target encapsulation level, target traffic shaping parameters and sending window parameters according to the current network status and the data packet to be transmitted through the data transmission model; send the data packet to be transmitted based on the target encapsulation level, target traffic shaping parameters and sending window parameters.

[0061] In the existing technology, UDP is a connectionless transport layer protocol that provides unreliable, best-effort data transmission services. The UDP message header structure is concise and does not require a session connection to be established, which enables it to exhibit efficient transmission performance in areas such as real-time communication, multimedia transmission, and gaming applications. However, due to the lack of error detection and correction mechanisms, when network conditions are poor, data packets may be lost, delivered out of order, or duplicated, which is an important issue for business applications that require high reliability. In existing network communication technology solutions, dynamic and fine-grained access control is usually achieved through the collaborative work of clients, management and control platforms, and application gateways based on the SDP infrastructure. Among them, the SDP client is the entrance to zero-trust network security access. It is used to communicate with the access controller and the stealth gateway to establish a zero-trust connection, verify the identity of the access subject in real time, and perform terminal environment detection regularly to ensure that the terminal equipment during user authentication login and access is trustworthy; the SDP stealth gateway can hide the protected business resources and system resources, execute the dynamic access control policy issued by the trust assessment center, avoid business resources from network attacks, and realize the zero-trust network's security concept that is not centered on boundaries; the SDP management and control platform is the decision maker of zero-trust network access behavior, responsible for access authentication of access terminals and user identities, access authentication of users and stealth assets, continuous trust evaluation of access connections, generation and issuance of dynamic access control decisions, and alarm and disposal of abnormal access behaviors.

[0062] The present application provides a solution, which includes acquiring an original data packet and a current network status; embedding and encapsulating the original data packet to obtain a data packet to be transmitted; predicting a target encapsulation level, a target traffic shaping parameter, and a sending window parameter according to the current network status and the data packet to be transmitted through a data transmission model; and sending the data packet to be transmitted based on the target encapsulation level, the target traffic shaping parameter, and the sending window parameter; the method embeds and encapsulates the current data packet, and analyzes the network status and application requirements of the encapsulated data packet in real time according to the data transmission model to obtain the target encapsulation level, the target traffic shaping parameter, and the sending window parameter, thereby dynamically selecting the optimal encapsulation strategy, traffic shaping parameter, and congestion control strategy, which can be flexibly adjusted according to actual network conditions to improve communication performance and resource utilization.

[0063] It should be noted that the execution subject of this embodiment may be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a data transmission method, etc. The following takes the data transmission method as an example to illustrate this embodiment and the following embodiments.

[0064] Based on this, the embodiment of the present application provides a data transmission method, referring to Figure 1 , Figure 1 This is a flowchart of the first embodiment of the data transmission method of the present application.

[0065] In this embodiment, the data transmission method includes steps S10 to S40:

[0066] Step S10: Obtain the original data packet and current network status

[0067] It should be noted that the acquisition of the original data packet can be uploaded by the user or obtained from a pre-defined storage space at a fixed time. The original data packet can also be understood as the data generated by the application layer.

[0068] It should be noted that the current network status may include indicators such as bandwidth utilization, packet loss rate, and delay, and the network status may be acquired in real time by monitoring a network interface.

[0069] It should be noted that the AI ​​decision-making module plays a key role in the data transmission method in the process of communication from client to controller and client to gateway, optimizing the performance and security of network communication through intelligent decision-making. At the same time, the various components work together, combining the efficiency of UDP and the reliability of TCP to adapt to diverse network communication needs.

[0070] In the specific implementation, in order to facilitate the understanding of the data transmission method applied to the intelligent encapsulation TCP protocol system, the intelligent encapsulation TCP protocol system structure can refer to Figure 2 , Figure 2 The system mainly includes: user client, SDP subsystem and business resources. Among them, business resources include security devices, hosts and databases, and SDP subsystem includes SDP controller and SDP gateway. The data transmission method is applied between the user client and the SDP subsystem.

[0071] It should be noted that the data transmission method can be understood as including 5 modules, among which the structure diagram between each application module can refer to Figure 3 , Figure 3 It first passes through the data encapsulation module, then the AI ​​decision-making module, then the traffic shaping and optimization module and the congestion control module, and finally the security threat identification and defense module.

[0072] Step S20: embed and encapsulate the original data packet to obtain a data packet to be transmitted.

[0073] It is understandable that embedded encapsulation can be performed based on the original data packet, and the header of the encapsulated data packet is embedded with control information such as a sequence number, confirmation response, and checksum to complete the embedded encapsulation of the original data packet.

[0074] In a feasible implementation, step S20 may include steps A21 to A24:

[0075] Step A21, converting the format of the original data packet to obtain a formatted data packet.

[0076] It is understandable that the original data packet has not been processed yet and needs to be further processed before transmission.

[0077] It should be noted that the format conversion is to convert the original data packets into a format suitable for network transmission. The format of the original data packets after format conversion and preliminary sorting can be more unified, which is convenient for further processing in subsequent steps.

[0078] It should be noted that the format conversion of the original data packet to obtain the formatted data packet may be conversion into a pre-set format, or segmentation, encoding, compression and the like, which is not limited in this embodiment and can be adjusted according to actual conditions.

[0079] Step A22, obtaining the unique serial number of the original data packet according to a preset serial number mechanism.

[0080] It is understandable that the preset sequence number mechanism can be a TCP-like sequence number mechanism, which assigns a unique sequence number to each data packet to be sent to ensure that the receiving end can reassemble the data stream in sequence. Through sequence number allocation, global uniqueness is ensured under multiple concurrent connections, and retransmissions caused by sequence number conflicts are minimized.

[0081] It should be understood that the TCP-like sequence number mechanism refers to a sequence number processing method used when designing other protocols with similar characteristics to the Transmission Control Protocol (TCP).

[0082] Step A23, encapsulating the formatted data packet to obtain an encapsulated data packet, and obtaining a checksum based on the encapsulated data packet.

[0083] It should be noted that for each data packet after encapsulation, a strong checksum algorithm, Cyclic Redundancy Check (CRC), is used to calculate the checksum. CRC is an error detection technology widely used in the communication field. It can effectively detect most types of changes or damages that occur during data transmission to ensure that the received data maintains integrity. Compared with simple checksum calculation methods, CRC provides stronger error detection capabilities.

[0084] It is understandable that the checksum is calculated for the encapsulated data packet to ensure that the data has not been changed during transmission. A strong check algorithm CRC check mechanism is used to ensure data integrity.

[0085] Step A24, obtaining the data packet to be transmitted according to the encapsulated data packet, the unique sequence number and the checksum.

[0086] It should be noted that the data packet to be transmitted may include a header and a body. The encapsulated data packet is used as the body to ensure data security. At the same time, the unique serial number and the checksum are integrated into the UDP data packet header. At the same time, the header structure is optimized, the additional overhead is reduced, and the original low latency advantage of the UDP protocol is maintained.

[0087] It should be noted that in order to maintain the original low-latency characteristics of the UDP protocol while minimizing additional overhead, the process also includes optimizing the header structure. The purpose of this is to add the necessary reliability features without affecting the inherent advantages of UDP, so that a high data transmission quality can be guaranteed even under poor network conditions.

[0088] It is understandable that the data packet to be transmitted may be a data packet that has been processed and is waiting for data transmission.

[0089] In the specific implementation, the embedding and encapsulation process of the original data packet can refer to Figure 4 , Figure 4 First, the application layer data, that is, the original data, is obtained and preprocessed. After preprocessing, a sequence number is generated and inserted into the header. The checksum of the encapsulated data packet is calculated and added to the header, and the header is further encapsulated and optimized.

[0090] In this implementation, the reliability mechanism of the TCP protocol is simulated by deeply processing the original data, and control information such as sequence numbers, confirmation responses, checksums, etc. are embedded in UDP data packets to ensure that the integrity and orderliness of data transmission are guaranteed while maintaining the high efficiency and low latency advantages of UDP.

[0091] The above are only feasible implementations of step S20 provided in this embodiment, and this embodiment does not specifically limit the specific implementation of step S20.

[0092] Step S30: The target encapsulation level, target traffic shaping parameters and sending window parameters are predicted based on the current network status and the data packet to be transmitted by the data transmission model. (No explanation)

[0093] It is understandable that the encapsulation level can be determined based on the predicted network status, and the size of the encapsulated data packet can be determined based on the network status. The better the network status, the higher the encapsulation level and the larger the data packet. The traffic shaping parameters can be the maximum rate limit, burst size, queue length, etc. The sending window parameters include the data sending window size.

[0094] It is understandable that the target shaping traffic parameters may be parameters such as a target transmission rate and a target unit time transmission limit.

[0095] It should be noted that the data transmission model is applied to the AI ​​decision-making module in the intelligent encapsulation TCP protocol system. The AI ​​decision-making module intelligently judges and selects various key decisions in the network communication process through the data transmission model.

[0096] It should be noted that the data transmission model analyzes network conditions, application requirements and historical data in real time, accurately predicts and selects the best encapsulation strategy, traffic shaping parameters and congestion control rules. Through AI intelligent decision-making, the efficiency and security of network communications are improved, and efficient and secure communications that dynamically adapt to network environments and business needs are achieved.

[0097] In the specific implementation, the execution steps of the AI ​​decision module can refer to Figure 5 , Figure 5 First, the real-time network status is collected, and then further data analysis and modeling are performed to select the best packaging strategy based on the trained model.

[0098] In a feasible implementation manner, step S30 may include steps A301 to A304:

[0099] Step A301, construct an initial transmission model and obtain training data.

[0100] It should be noted that the initial transmission model can be a deep Q-network. The deep Q-network (DQN) is an algorithm that combines reinforcement learning (RL) and deep learning.

[0101] It is understandable that the parameters in the deep Q network that constitutes the initial transmission model are random parameters, and the initial transmission model needs to be further trained through training data to optimize the parameters.

[0102] It should be understood that the training data may be a one-to-one corresponding data packet, network state, target encapsulation level, target traffic shaping parameter, and sending window parameter. The network state in the training data may be a real possible network state measured for training, and the target encapsulation level, target traffic shaping parameter, and sending window parameter may be the best encapsulation level, traffic shaping parameter, and window parameter in the network state.

[0103] Step A302, obtaining a state space according to the network state in the training data, and obtaining an action space according to the encapsulation level, traffic shaping parameters and sending window parameters in the training data.

[0104] It should be noted that the state space in the network communication system is defined as the current network status (such as bandwidth utilization, delay, packet loss rate, etc.), data packet queue length, application requirements (such as data volume, real-time, priority) and security threat level.

[0105] It should be noted that a set of possible decision actions, such as selecting different encapsulation strategies, traffic shaping parameters, congestion window sizes, etc., is set as the action space.

[0106] Step A303, constructing a reward function according to a preset network communication target.

[0107] It should be noted that the reward function is designed according to the goals and requirements of network communication. For example, positive rewards are given for good performance such as reducing latency, increasing throughput, improving data transmission success rate, and reducing packet loss rate, and negative penalties are given otherwise.

[0108] Step A304: training the initial transmission model according to the state space, the action space and the reward function to obtain a data transmission model.

[0109] It can be understood that a deep neural network model is constructed based on the deep Q network as an approximator of the Q-value function, where the input layer corresponds to the state space and the output layer corresponds to the Q-value estimation of each action.

[0110] It should be understood that the corresponding state space, action space, reward and state space at the next time are obtained according to each training data.

[0111] Among them, it is understandable that the training steps of the initial transmission model can refer to Figure 6 , Figure 6In the paper, we first define the state space and action space, then design the reward function, build a model based on the deep Q network, and iteratively train the model to obtain the final decision-making process.

[0112] It should be noted that the initial transmission model is trained according to the state space, the action space and the reward function to obtain the data transmission model, including: obtaining the state space and the current reward at the next moment according to the state space, the action space and the reward function; using the state space, the current action space, the state space at the next moment and the current reward as a training data group; performing back propagation training based on the training data group to optimize the weight parameters of the initial transmission model to obtain the data transmission model.

[0113] It is understandable that, in the actual operation process, actions are selected according to the current network state, and immediate state rewards and new network states are collected after the actions are executed.

[0114] Among them, each group of corresponding state space, action space, reward and state space at the next time can be stored in a specific storage space as a four-dimensional data group, which is convenient for subsequent tracing and secondary utilization of the training data.

[0115] It should be noted that, based on the training data set, back propagation training is performed to optimize the weight parameters of the initial transmission model, which can refer to the following formula:

[0116] Q(s,a)=r+γa′maxQ(s′,a′)

[0117] It can be understood that this formula is an update of the Q value, where Q(s′,a′) is the maximum Q value estimate of the target network taking all actions a′ for the new state s′.

[0118] Among them, Q(s,a) refers to the expected total reward that can be obtained after performing action a in state s. s is the current state; a is the action taken in the current state; r is the reward immediately obtained after performing action a; γ is a discount factor, ranging from 0 to 1, which is used to weigh the importance of current rewards and future rewards; s′ is the new state transferred after performing action a; a′ is the action that maximizes the Q value among all possible actions in the new state s'; maxQ(s′,a′): is the maximum Q value of all possible actions in the new state s'.

[0119] It should be noted that back propagation training can use the gradient descent method to optimize the weight parameters of the deep Q network in a timely manner to minimize the gap between the predicted Q value and the target Q value. Every certain number of iterations, the parameters of the main network are copied to the target network to stabilize the learning process.

[0120] Furthermore, at each decision, the current network state is input into the trained deep Q network to calculate the Q values ​​of all actions. The action with the highest Q value is selected for execution, and the new state and immediate reward generated by this decision are recorded and added to the experience playback buffer to provide data for the next round of iterative training.

[0121] In this implementation, through the AI ​​decision-making module with deep Q network as the core, the encapsulation, traffic shaping, and congestion control strategies can be intelligently and dynamically adjusted according to the ever-changing network environment and application requirements to optimize network communication performance and security.

[0122] The above are merely feasible implementations before step S30 provided in this embodiment, and this embodiment does not specifically limit the specific implementations before step S30.

[0123] Step S40: Send the data packet to be transmitted based on the target encapsulation level, the target traffic shaping parameter and the sending window parameter.

[0124] It is understandable that the data packet encapsulation size, data transmission rate, speed limit, and data transmission window size are intelligently selected for the encapsulated data packet and the current network status through the data transmission model.

[0125] It should be noted that after sending the data packet to be transmitted based on the target encapsulation level, the target traffic shaping parameters and the sending window parameters, the security threat identification and defense module can monitor and deeply analyze the data traffic and user behavior patterns in network communications around the clock to keenly identify potential security threats and abnormal behaviors. Once suspicious or malicious activities are detected, corresponding defense measures can be taken, such as blocking malicious traffic, triggering early warnings, etc., and the relevant information can be fed back to the AI ​​decision-making module. The AI ​​decision-making module can further optimize the encapsulation strategy, traffic control strategy and other network communication strategies based on these security-related real-time information to improve the security and stability of the entire system.

[0126] In the specific implementation, the execution steps of the security threat identification and defense module can refer to Figure 7 , Figure 7 Included are: S501: Continuously monitor network communication traffic and establish a normal user behavior model through pattern recognition, anomaly detection and other methods. S502: Use deep learning algorithms to analyze network traffic characteristics in real time and discover potential security threats, such as DDoS attacks and data injection. S503: When a threat is detected, immediately execute corresponding defense measures and feed back security event information to the AI ​​decision-making module to update the overall communication strategy and improve the robustness and security of the system.

[0127] It is worth noting that by improving the security of network communications, intelligently responding to network attacks and abnormal behaviors, ensuring the security and integrity of data transmission, and dynamically adjusting communication strategies based on security threat information to ensure communication security

[0128] It is worth mentioning that the intelligent encapsulation process is based on the processing of the original data by the data encapsulation module, embedding TCP-related information in the UDP packet, and optimizing the traffic flow and adjusting the transmission rate of the network communication process through AI decision-making. According to the changes in the network status, the data transmission is continuously adjusted. Finally, based on the security threat identification module, the data flow in the network process is analyzed, and the information is fed back to the AI ​​decision-making module again. The AI ​​decision-making module collects the corresponding information and further optimizes the encapsulation strategy, traffic control and other strategies to improve the security and stability of the entire system and the security protection capabilities of network communications.

[0129] It should be noted that the data transmission method based on the security threat identification and defense module can be referred to Figure 8 , Figure 8 The data encapsulation module encapsulates the data, and the encapsulated data is sent to the AI ​​decision-making module, which sends it to the traffic shaping optimization module and the congestion control module, and finally to the security threat identification and defense module. The defense module conducts real-time detection and protection and feeds back to the AI ​​decision-making module to adjust the decision, and then the data transmission network communication is carried out.

[0130] This embodiment provides a data transmission method, which embeds and encapsulates the current data packet, analyzes the network status and application requirements of the encapsulated data packet in real time according to the data transmission model to obtain the target encapsulation level, target traffic shaping parameters and sending window parameters, thereby dynamically selecting the optimal encapsulation strategy, traffic shaping parameters and congestion control strategy, and can be flexibly adjusted according to actual network conditions to improve communication performance and resource utilization.

[0131] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Fig. 9 , step S30 also includes steps S31 to S33:

[0132] Step S31: obtaining a predicted network state according to the network state prediction model and the current network state, and obtaining a target encapsulation level according to the predicted network state.

[0133] It should be noted that the data transmission model includes a network state estimation model and a congestion window control model.

[0134] It should be noted that the network status prediction model can be a machine learning model or a deep learning model, which is not required in this embodiment.

[0135] It should be emphasized that before obtaining the predicted network state according to the network state prediction model and the current network state, the network state prediction model can be trained by historical data, and a dynamic network model can be formed according to the predicted network state change trend. The linear regression formula in this embodiment is used as an example for prediction. The following formula can be used for prediction:

[0136] y=w1*x1+w2*x2+…wn*xn+b

[0137] Among them, y represents the target variable to be predicted (for example, the predicted network delay or the optimal transmission rate); x1, x2, ..., xn are input features (for example, current network bandwidth, historical packet loss rate, time point, etc.); w1, w2, ..., wn are the corresponding feature weights, indicating the influence of each feature on the target variable; b is the intercept term, which indicates the value of the target variable even if all features are 0.

[0138] It should be noted that during the training phase, a set of optimal weights w and intercepts b are found through historical data sets to minimize the error between the model prediction value and the actual value. The prediction model may be used to predict the development trend of network conditions, thereby guiding the formulation of transmission strategies. For example, if it is predicted that the network delay may increase in the future, the encapsulation strategy or congestion control parameters can be adjusted to adapt to possible adverse network environments. According to the network model and application requirements, the most appropriate encapsulation level, transmission rate, and congestion control parameters are dynamically determined through AI algorithms to ensure the best performance balance under different network conditions.

[0139] It should be noted that the predicted network status can be used to obtain the predicted bandwidth utilization, packet loss rate, and delay respectively through the above linear regression formula.

[0140] It should be understood that different encapsulation levels are set in advance according to different bandwidth utilization ranges, packet loss rate ranges, and delay ranges, and the corresponding encapsulation level is obtained according to the predicted bandwidth utilization, packet loss rate, and delay range as the target encapsulation level.

[0141] It is understandable that the encapsulation level may include level 1 to level 3, and may also be adjusted according to a preset range, which is not limited in this embodiment.

[0142] It is understandable that different encapsulation levels correspond to encapsulation packages of different sizes.

[0143] Step S32: Acquire the real-time data transmission rate, and obtain the target traffic shaping parameter according to the real-time data transmission rate, a preset rate threshold, and the real-time capacity of the network.

[0144] It should be noted that the target traffic shaping parameters obtained according to the real-time data transmission rate, the preset rate threshold and the real-time capacity of the network can be realized through a traffic shaping and optimization module, which fine-tunes the output rate of network communication traffic, and effectively shapes and optimizes the traffic according to the network status and business load, so as to avoid network congestion or waste of resources caused by excessive or insufficient traffic. By adjusting the rate in a timely manner, it can not only balance the network load and improve resource utilization, but also prevent performance bottlenecks caused by the mismatch between the communication rate and the back-end processing capacity, thereby ensuring smooth and stable communication throughout the process. The execution steps of the traffic shaping and optimization module are as follows: Fig.10 As shown, Fig.10 The traffic shaping and optimization module first performs baseline setting and threshold management, obtains real-time traffic through monitoring and analyzes it, and dynamically controls the traffic rate.

[0145] It should be noted that the target traffic shaping parameters obtained according to the real-time data transmission rate, the preset rate threshold and the real-time capacity of the network can first establish an initial working benchmark to ensure that basic service quality can be maintained without additional intelligent intervention. Further, the speed at which the data packet starts to be transmitted is defined, which is usually set to be lower than the maximum capacity of the network to reserve a certain buffer space to cope with burst traffic. An upper limit value is set to prevent the data transmission rate at any time from exceeding the maximum value allowed by the physical bandwidth or service level agreement (SLA). By continuously monitoring the network status (such as delay, packet loss rate, available bandwidth, etc.), the above parameters are automatically adjusted to ensure that they always match the current network environment, thereby achieving dynamic adaptability.

[0146] It should be emphasized that by accurately measuring the real-time data transmission rate and comparing it with the preset threshold and the real-time capacity of the network, situations that require action can be identified. Use NetFlow, sFlow, IPFIX or other network monitoring tools to collect real-time traffic statistics, including but not limited to key indicators such as throughput, latency, and jitter. Compare the actual observed data transmission rate with the previously set basic traffic shaping parameters, while considering the real-time capacity of the network to evaluate whether the current transmission status is healthy. Based on the comparison results, use specific algorithms (such as PID controllers, adaptive filters, etc.) to calculate appropriate adjustment strategies to ensure that the transmission rate can fully utilize the available bandwidth without causing congestion.

[0147] It should be emphasized that the rate control logic is further optimized to provide a more accurate and flexible response mechanism. By pre-training the rate adjustment model, which may integrate a machine learning model, a deep learning framework, or a reinforcement learning algorithm, it is possible to predict future traffic trends based on historical data and current network conditions and propose the best rate adjustment suggestions. The specific model is not limited in this embodiment and can be selected according to actual needs.

[0148] Step S33: Acquire the network congestion status, and obtain the sending window parameter according to the congestion window control model and the network congestion status.

[0149] It is understandable that the network congestion state may include a congestion state and a normal state, and the judgment of the network congestion state may be determined based on the network throughput and the network transmission efficiency; for example, by using network feedback information, such as increased delay, continuous retransmission events, etc., it is possible to judge in real time whether the network is in a congested state.

[0150] It should be noted that obtaining the network congestion status and obtaining the sending window parameters according to the congestion window control model and the network congestion status can be achieved through a congestion control module, which monitors and dynamically adjusts the data transmission rate to prevent network congestion and performance degradation.

[0151] In the specific implementation, the specific execution steps of the congestion control module can be referred to Fig.11 , Fig.11 The congestion control module first detects the network congestion status, then adjusts the sending window size according to the network congestion status, and further determines the intelligent retransmission strategy.

[0152] In a feasible implementation, step S33 may include steps A331 to A333:

[0153] Step A331, obtaining network congestion status.

[0154] It should be noted that the sending window parameter can be predicted by the network congestion status through the congestion window control model.

[0155] It should be noted that the congestion window control model adopts a congestion window control strategy similar to TCP, dynamically increasing or decreasing the sending window size according to the degree of congestion and reasonably allocating network resources.

[0156] Step A332, obtaining the reward value of each sending window parameter at the next moment according to the network congestion state, and obtaining the target reward value according to the reward value.

[0157] It should be noted that the real-time reward obtained based on the network congestion status and the sending window parameters at the previous moment can also be obtained through analysis based on the deep Q network. The calculation of the real-time reward can refer to the following formula:

[0158] Q'(s_t,a_t)=Q(s_t,a_t)+x*[r_t+n*max(Q(s_{t+1},a')-Q(s_t,a_t))]

[0159] Among them, s_t represents the current network state, such as the current congestion window size, packet loss rate, delay and other network parameters; a_t represents the action taken in state st, that is, the adjustment of the sending window size (increase, decrease or remain unchanged); Q'(s_t, a_t) represents the expected cumulative return of executing action a_t in state st; r_t: immediate reward, based on the network feedback after executing action a_t (such as the number of successfully transmitted data packets, packet loss, etc.); x represents the learning rate, which determines the influence of new information on the existing Q value, between 0 and 1. The larger the value, the easier it is for the model to learn new experience; n represents the discount factor, which represents the importance of future rewards, between 0 and 1. The closer the value is to 1, the more the model values ​​long-term rewards; max(Q(s_{t+1}, a') represents the Q value that is expected to bring the maximum return among all possible actions a' in the next state s_{t+1}.

[0160] In the specific implementation, the congestion window control strategy is trained in advance. The specific steps are as follows: 1. Initialize the Q value; 2. Interact with the network environment. In each cycle: a. Observe the current network state s_t. b. Select an action a_t (it can be a greedy choice of the action with the highest current Q value, or an exploratory choice). c. Execute the action and observe the result to obtain the immediate reward r_t and the new network state s_{t+1}. d. Use the above formula to update the value of Q(s_t, a_t). 3. Repeat step 2 until the model converges or reaches the predetermined number of training rounds. In the end, the model will derive a strategy for how to adjust the sending window size to maximize long-term rewards under different network conditions through continuous learning and updating. When facing different network conditions, it can independently decide whether to increase or decrease the sending window, thereby effectively controlling congestion and improving network transmission efficiency and stability.

[0161] Finally, AI algorithms are used to predict packet loss and schedule retransmission tasks in advance, or optimize the timing of retransmission to reduce unnecessary waiting and network burden.

[0162] It is understandable that the target reward value can be the maximum of the various reward values. For each possible action (i.e., the setting of the sending window parameter), an immediate reward function is defined, which measures the benefit or cost immediately obtained after taking a specific action. For example, if increasing the window size leads to more data being transmitted quickly, a positive reward is given; conversely, if it causes severe congestion, a negative reward is given.

[0163] It should be understood that the reward value may be a reward value calculated based on a reward function based on different window sizes at the next moment corresponding to the network congestion state.

[0164] Step A333, obtaining a sending window parameter according to the network congestion status and the target reward value.

[0165] It is understandable that the sending window parameter may be the state at the next moment, and the sending window parameter may be predicted based on the network congestion state and the target reward value.

[0166] In this implementation, the sending window size is flexibly adjusted and the retransmission strategy is optimized according to changes in network conditions to ensure smooth data transmission in a complex network environment. Effective congestion control can not only prevent network congestion and data packet loss, but also improve network throughput and transmission efficiency, ensuring real-time and accurate data delivery.

[0167] The above are only feasible implementations of step S33 provided in this embodiment, and this embodiment does not specifically limit the specific implementation of step S33.

[0168] This embodiment provides a data transmission method, which analyzes network status and requirements in real time through a Q network model, calculates the encapsulation strategy Q value, traffic shaping Q value and congestion control Q value, and selects the optimal network communication strategy based on this, and dynamically generates the optimal communication strategy according to the network status information, so as to achieve effective utilization of network resources, reduce network congestion and data loss, and generate the optimal communication strategy that adapts to the current network environment.

[0169] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the data transmission method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0170] This application also provides a data transmission device, please refer to Fig.12 , the data transmission device comprises:

[0171] The acquisition module 10 is used to acquire the original data packet and the current network status;

[0172] The acquisition module 10 is also used to embed and encapsulate the original data packet to obtain the data packet to be transmitted;

[0173] A parameter prediction module 20, configured to predict a target encapsulation level, a target traffic shaping parameter, and a sending window parameter according to the current network state and the data packet to be transmitted through a data transmission model;

[0174] The data transmission module 30 is used to send the data packet to be transmitted based on the target encapsulation level, the target traffic shaping parameter and the sending window parameter.

[0175] The data transmission device provided by the present application adopts the data transmission method in the above embodiment, which can solve the technical problem that the communication performance and security of data transmission are seriously affected by network congestion in scenarios with high requirements for data integrity and reliability. Compared with the prior art, the beneficial effects of the data transmission device provided by the present application are the same as the beneficial effects of the data transmission method provided by the above embodiment, and the other technical features in the data transmission device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0176] In one embodiment, the parameter prediction module 20 is further used to obtain a predicted network state according to the network state prediction model and the current network state, and obtain a target encapsulation level according to the predicted network state;

[0177] Acquire a real-time data transmission rate, and obtain a target traffic shaping parameter according to the real-time data transmission rate, a preset rate threshold, and a real-time capacity of the network;

[0178] The network congestion status is acquired, and a sending window parameter is obtained according to the congestion window control model and the network congestion status.

[0179] In one embodiment, the parameter prediction module 20 is also used to obtain the network congestion status;

[0180] Obtaining a reward value of each sending window parameter at the next moment according to the network congestion state, and obtaining a target reward value according to the reward value;

[0181] A sending window parameter is obtained according to the network congestion status and the target reward value.

[0182] In one embodiment, the parameter prediction module 20 is also used to construct an initial transmission model and obtain training data;

[0183] Obtain a state space according to the network state in the training data, and obtain an action space according to the encapsulation level, traffic shaping parameters, and sending window parameters in the training data;

[0184] Construct a reward function based on the preset network communication goals;

[0185] The initial transmission model is trained according to the state space, the action space and the reward function to obtain a data transmission model.

[0186] In one embodiment, the parameter prediction module 20 is further used to obtain the state space and current reward at the next moment according to the state space, the action space and the reward function;

[0187] The state space, the current action space, the state space at the next moment, and the current reward are used as a training data set;

[0188] Back propagation training is performed based on the training data group to optimize the weight parameters of the initial transmission model to obtain a data transmission model.

[0189] In one embodiment, the acquisition module 10 is further used to convert the format of the original data packet to obtain a formatted data packet;

[0190] Obtaining a unique serial number of the original data packet according to a preset serial number mechanism;

[0191] Encapsulating the formatted data packet to obtain an encapsulated data packet, and obtaining a checksum according to the encapsulated data packet;

[0192] A data packet to be transmitted is obtained according to the encapsulated data packet, the unique sequence number and the checksum.

[0193] The present application provides a data transmission device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the data transmission method in the above-mentioned embodiment one.

[0194] Reference below Fig.13 , which shows a schematic diagram of the structure of a data transmission device suitable for implementing the embodiment of the present application. The data transmission device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Fig.13 The data transmission device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0195] like Fig.13As shown, the data transmission device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the data transmission device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the data transmission device to communicate with other devices wirelessly or by wire to exchange data. Although the data transmission device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0196] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0197] The data transmission device provided by the present application adopts the data transmission method in the above embodiment, which can solve the technical problem that the communication performance and security of data transmission are seriously affected by network congestion in scenarios with high requirements for data integrity and reliability. Compared with the prior art, the beneficial effects of the data transmission device provided by the present application are the same as the beneficial effects of the data transmission method provided by the above embodiment, and the other technical features in the data transmission device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0198] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0199] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0200] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, wherein the computer-readable program instructions are used to execute the data transmission method in the above-mentioned embodiment.

[0201] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0202] The computer-readable storage medium may be included in the data transmission device; or may exist independently without being assembled into the data transmission device.

[0203] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by a data transmission device, the data transmission device enables the following: to obtain the original data packet and the current network status; to embed and encapsulate the original data packet to obtain the data packet to be transmitted; to predict the target encapsulation level, target traffic shaping parameters and sending window parameters according to the current network status and the data packet to be transmitted through a data transmission model; and to send the data packet to be transmitted based on the target encapsulation level, target traffic shaping parameters and sending window parameters.

[0204] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0205] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0206] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0207] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned data transmission method, and can solve the technical problem that the communication performance and security of data transmission are seriously affected by network congestion in scenarios with high requirements for data integrity and reliability. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the data transmission method provided by the above-mentioned embodiment, and will not be repeated here.

[0208] The present application also provides a computer program product, including a computer program, which implements the steps of the above-mentioned data transmission method when executed by a processor.

[0209] The computer program product provided by the present application can solve the technical problem that the communication performance and security of data transmission are seriously affected by network congestion in scenarios with high requirements for data integrity and reliability. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the data transmission method provided by the above embodiment, and will not be repeated here.

[0210] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A data transmission method, characterized in that: The data transmission method comprises: Get the original data packet and current network status; Embed and encapsulate the original data packet to obtain the data packet to be transmitted; The target encapsulation level, target traffic shaping parameters and sending window parameters are predicted based on the current network status and the data packets to be transmitted by a data transmission model; The data packet to be transmitted is sent based on the target encapsulation level, the target traffic shaping parameter, and the sending window parameter.

2. The method according to claim 1, characterized in that The data transmission model includes a network state estimation model and a congestion window control model; The method of predicting a target encapsulation level, a target traffic shaping parameter, and a sending window parameter according to the current network state and the data packet to be transmitted by using a data transmission model includes: Obtain a predicted network state according to the network state prediction model and the current network state, and obtain a target encapsulation level according to the predicted network state; Acquire a real-time data transmission rate, and obtain a target traffic shaping parameter according to the real-time data transmission rate, a preset rate threshold, and a real-time capacity of the network; The network congestion status is acquired, and a sending window parameter is obtained according to the congestion window control model and the network congestion status.

3. The method according to claim 2, characterized in that The obtaining of the network congestion status and obtaining the sending window parameter according to the congestion window control model and the network congestion status include: Get network congestion status; Obtaining a reward value of each sending window parameter at the next moment according to the network congestion state, and obtaining a target reward value according to the reward value; A sending window parameter is obtained according to the network congestion status and the target reward value.

4. The method according to claim 1, characterized in that Before the data transmission model is used to predict the target encapsulation level, the target traffic shaping parameter and the sending window parameter according to the current network state and the data packet to be transmitted, the method further includes: Build the initial transfer model and obtain training data; Obtain a state space according to the network state in the training data, and obtain an action space according to the encapsulation level, traffic shaping parameters, and sending window parameters in the training data; Construct a reward function based on the preset network communication goals; The initial transmission model is trained according to the state space, the action space and the reward function to obtain a data transmission model.

5. The method according to claim 4, characterized in that The training of the initial transmission model according to the state space, the action space and the reward function to obtain a data transmission model includes: Obtaining the state space and current reward at the next moment according to the state space, the action space and the reward function; The state space, the current action space, the state space at the next moment, and the current reward are used as a training data set; Back propagation training is performed based on the training data group to optimize the weight parameters of the initial transmission model to obtain a data transmission model.

6. The method according to claim 1, characterized in that The embedding and encapsulating of the original data packet to obtain the data packet to be transmitted includes: Converting the format of the original data packet to obtain a formatted data packet; Obtaining a unique serial number of the original data packet according to a preset serial number mechanism; Encapsulating the formatted data packet to obtain an encapsulated data packet, and obtaining a checksum according to the encapsulated data packet; A data packet to be transmitted is obtained according to the encapsulated data packet, the unique sequence number and the checksum.

7. A data transmission device, characterized in that: The data transmission device comprises: The acquisition module is used to obtain the original data packet and the current network status; The acquisition module is further used to embed and encapsulate the original data packet to obtain the data packet to be transmitted; A parameter prediction module, used to predict the target encapsulation level, target traffic shaping parameters and sending window parameters according to the current network state and the data packet to be transmitted through a data transmission model; A data transmission module is used to send the data packet to be transmitted based on the target encapsulation level, the target traffic shaping parameter and the sending window parameter.

8. A data transmission device, characterized in that: The data transmission device comprises: a memory, a processor, and a data transmission program stored in the memory and executable on the processor, wherein the data transmission program is configured to implement the data transmission method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium stores a data transmission program, and when the data transmission program is executed by the processor, the data transmission method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The computer program product comprises a data transmission program, and when the data transmission program is executed by a processor, the steps of the data transmission method according to any one of claims 1 to 6 are implemented.

Images