Data transmission method and device for multi-path parallel transmission and electronic equipment
Through multi-path parallel transmission and intelligent packet segmentation, combined with network exception monitoring and repair mechanisms, the inefficiency and insufficient robustness of single-path transmission under network congestion and failures are solved, and efficient and intelligent data transmission is achieved.
Patent Information
- Application Number
- CN202411055633.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing single-path data transmission protocol leads to low data transmission rates and packet loss in the event of network congestion or link failure, and the existing TCP/IP protocol packet loss and retransmission mechanism cannot effectively solve the problem of high packet loss rate.
By using multi-path parallel transmission method, by obtaining the status information of the transmission path, predicting future status, evaluating the quality of the transmission path, and intelligently splitting data packets for multi-path transmission, monitoring and repairing network abnormalities in real time, improving data transmission efficiency and robustness .
It significantly improves data transmission efficiency and resource scheduling intelligence, improves data transmission performance when network resources are limited, and reduces the impact of network anomalies by automatically switching redundant paths, enhancing the robustness of data transmission.
Smart Images

Figure CN119071247A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of network communications, and in particular to a data transmission method, device and electronic equipment for multi-path parallel transmission. Background Art
[0002] In the field of network communications, the traditional TCP / IP protocol, HTTP protocol, and FTP protocol are all single-path data transmission protocols. These existing single-path data transmission methods will have problems such as low data transmission rate and packet loss in the case of network congestion and link failure. Although the existing TCP / IP protocol has a packet loss retransmission mechanism, a high packet loss rate will still cause a significant drop in data transmission rate. Summary of the invention
[0003] Based on this, the purpose of the present invention is to provide a data transmission method, device and electronic device for multi-path parallel transmission, which can not only realize multi-path parallel transmission, but also predict the future performance of multiple transmission paths, evaluate the quality of the transmission paths based on the predicted future performance, and intelligently divide data packets according to the quality of multiple transmission paths, thereby greatly improving data transmission efficiency.
[0004] The present invention provides a data transmission method for multi-path parallel transmission, comprising the steps of: S1, acquiring state information of a data packet to be transmitted and multiple transmission paths; S2, predicting future state information of each transmission path based on the state information; S3, evaluating the quality of each transmission path based on the future state information; S4, dividing the data packet into a plurality of data blocks based on the quality of the transmission path, and distributing the plurality of data blocks to a plurality of transmission paths; and S5, transmitting the plurality of data blocks in parallel through the plurality of transmission paths.
[0005] The present invention has the following technical effects: 1. The present invention realizes multi-path parallel transmission, and compared with the existing single-path data transmission method, the data transmission efficiency is greatly improved. 2. The present invention can predict the future performance of the transmission path according to the state information of the transmission path, and evaluate the quality of the transmission path based on the predicted future performance, and finally intelligently allocate data blocks to multiple transmission paths according to the quality of the transmission path, which greatly improves the intelligence of resource scheduling under the condition of limited network resources and further improves the data transmission efficiency.
[0006] Furthermore, the method also includes the following steps for network anomaly monitoring and repair: SA1, real-time monitoring of the status information of the transmission path; SA2, judging whether the transmission path is abnormal based on the status information of each transmission path through a baseline model; SA3, if it is judged that the transmission path is abnormal, obtaining the abnormal transmission path; SA4, stopping using the abnormal transmission path, enabling the pre-planned redundant transmission path, and reallocating the data blocks to multiple transmission paths; SA5, continuously monitoring the status information of the transmission path, and judging whether the abnormal transmission path has returned to normal based on the status information through a baseline model; SA6, if it is judged that the abnormal transmission path has returned to normal, stopping using the redundant transmission path, re-enabling the abnormal transmission path, and reallocating the data blocks to multiple transmission paths.
[0007] Furthermore, the baseline model is an isolation forest model; step SA2 specifically includes: vectorizing a large amount of state information and current state information of the transmission path under normal conditions through the isolation forest model to obtain a set of points in the data space; then recursively dividing the data space through random hyperplanes to generate multiple subspaces; repeating the step of dividing the data space until each subspace of the last level contains only one data point; judging whether the point corresponding to the current state information is an abnormal point according to the level of the subspace where the point corresponding to the current state information is located, thereby judging whether the transmission path is abnormal.
[0008] Furthermore, step S2 is specifically as follows: using the LSTM model, based on the state information, predicting the future state information of each transmission path; the LSTM model is trained through the following steps: ST1, obtaining the historical state information of each transmission path; ST2, dividing the historical state information into several segments, and inputting them into the LSTM model as training samples; ST3, the LSTM model extracts and predicts features for each segment to obtain a predicted segment; ST4, comparing the predicted segment with the next segment corresponding to the current segment, and calculating a loss function; ST5, optimizing the parameters of the LSTM model according to the loss function; ST6, repeating steps ST3-ST5, the LSTM model updates the parameters each time a predicted segment is obtained, until the parameters of the LSTM model converge or the number of parameter updates reaches a set threshold, and a trained LSTM model is obtained.
[0009] Furthermore, step S3 specifically includes: S31, normalizing each indicator of the future state information to obtain a number of normalized indicator data; S32, calculating a weighted average of the normalized number of indicator data according to preset weights, and using the weighted average as a quality evaluation result of the transmission path.
[0010] Furthermore, step S4 specifically includes: S41, calculating the optimal size of the data block according to the quality of the transmission path; and dividing the data packet into a number of data blocks according to the optimal size of the data block; S42, converting the number of data blocks into a format that complies with a preset protocol to obtain data blocks that comply with the preset protocol; S43, according to the transmission priority of the data block and the quality of the transmission path, distributing the data blocks that comply with the preset protocol to multiple transmission paths in accordance with the load balancing principle.
[0011] Furthermore, step S41 specifically includes: calculating the average of the qualities of all available transmission paths to obtain a network quality parameter; substituting the network quality parameter into a preset function to calculate the optimal size of the data block; the preset function is an increasing function.
[0012] Furthermore, in step S43, data blocks are allocated to multiple transmission paths according to the load balancing principle, specifically: data blocks with higher transmission priority are preferentially allocated to transmission paths with higher quality.
[0013] Based on the same inventive concept, the present invention also provides a data transmission device for multi-path parallel transmission, including: a data acquisition module, used to obtain the status information of the data packet to be transmitted and multiple transmission paths; a state prediction model, used to predict the future state information of each transmission path according to the state information; a quality assessment model, used to assess the quality of each transmission path according to the future state information; a data segmentation module, used to segment the data packet into a plurality of data blocks according to the quality of the transmission path, and distribute the plurality of data blocks to a plurality of transmission paths; a transmission module, used to transmit a plurality of data blocks in parallel through a plurality of transmission paths.
[0014] Based on the same inventive concept, the present invention also provides an electronic device, comprising: a processor; a memory for storing a computer program executed by the processor; wherein, when the processor executes the computer program, it implements any of the above-mentioned multi-path parallel transmission data transmission methods.
[0015] For better understanding and implementation, the present invention is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 A schematic diagram of a module of a data transmission device for multi-path parallel transmission according to Embodiment 1 of the present invention;
[0017] Figure 2 A schematic flow chart of a data transmission method for multi-path parallel transmission according to Embodiment 1 of the present invention;
[0018] Figure 3 This is a schematic diagram of the structure of the LSTM model;
[0019] Figure 4 A unit schematic diagram of a network anomaly monitoring and repairing module of a data transmission device for path parallel transmission according to Embodiment 2 of the present invention;
[0020] Figure 5 This is a flow chart of the network anomaly monitoring and repairing steps of the data transmission method for path parallel transmission according to Embodiment 2 of the present invention. DETAILED DESCRIPTION
[0021] It should be clear that the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the embodiments of the present application.
[0022] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the embodiments of the present application. The singular forms of "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.
[0023] When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects, and do not have to be used to describe a specific order or sequence, nor can they be understood as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to the specific circumstances.
[0024] In addition, in the description of this application, unless otherwise specified, "plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.
[0025] It should be understood that the embodiments of the present application are not limited to the precise structures described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the embodiments of the present application is limited only by the appended claims.
[0026] Example 1
[0027] See also Figure 1-2 , Figure 1 This is a module diagram of a data transmission device for multi-path parallel transmission according to Embodiment 1 of the present invention. Figure 2 It is a flowchart of a data transmission method for multi-path parallel transmission according to Embodiment 1 of the present invention.
[0028] The data transmission device for multi-path parallel transmission of the present invention comprises: a data acquisition module 1, a state prediction model 2, a quality assessment model 3, a data segmentation module 4 and a transmission module 5.
[0029] The data acquisition module 1 is used to execute step S1: acquiring data packets to be transmitted and status information of multiple transmission paths. The status information includes bandwidth utilization, delay and packet loss rate of the transmission paths.
[0030] The state prediction model 2 is used to execute step S2: predict the future state information of each transmission path according to the state information. The state prediction model predicts the future state information of the transmission path according to the historical state information input in the past and the state information input at present, and includes but is not limited to machine learning models such as Convolutional Neural Networks (CNN), Recurrent Neural Network (RNN), and Long Short-term Memory (LSTM).
[0031] In this embodiment, the state prediction model is an LSTM model (Long Short-Term Memory Network). The LSTM model is a special recurrent neural network that controls the flow of information by introducing the concept of "gate", thereby solving the gradient vanishing and gradient explosion problems that are prone to occur in traditional recurrent neural networks when processing long sequences. The gated unit structure of the LSTM model consists of an input gate, an output gate, and a forget gate. The LSTM model can process the relationship between historical input information and current input information through three types of gate calculations.
[0032] See also Figure 3 , Figure 3 This is a structural diagram of the LSTM model. Figure 3 Middle C t represents the cell state at time t, h t represents the information at time t, x t Represents the input at the current moment, tanh is the activation function, and σ is the sigmoid function. Figure 3The region g1 in the figure represents the forget gate, the region g2 represents the input gate, and the region g3 represents the output gate.
[0033] The forget gate is used to decide which information to keep and which to forget. t-1 and the current information x t-1 Processing is performed and a value between 0 and 1 is output through the sigmoid function to determine the cell state C t-1 How much should be remembered. The forget gate expression is as follows: t =σ(W f ·[h t-1 ,x t ]+b f ), where W f and b f are the weight and bias of the forget gate respectively.
[0034] The input gate determines what information is added to the unit. First, h t-1 and x t-1 The sigmoid function is used to select which information to update, and the tanh function is used to obtain the candidate unit state. The expression of the input gate is as follows: t =σ(W i ·[h t-1 ,x t ]+b i ), The input gate updates the cell state by the following formula: Where i t represents the output of the input gate, W i and b i is the weight and bias of the input gate, W c and b c is the candidate unit state at time t The weights and biases.
[0035] The output gate controls what information the storage unit outputs. First, the output gate uses the sigmoid function to determine part of the output, and then uses the tanh function to process the storage unit state C. t , and multiply it with the output of the sigmoid function to get the output h at the current moment t The output gate expression is as follows: t =σ(W o ·[h t-1 ,x t ]+b o )h t ,h t =o t tanh(C t ), where ot represents the output of the output gate, W o and b o are the weight and bias of the output gate respectively.
[0036] Furthermore, the LSTM model is trained by the following training method:
[0037] ST1, obtain the historical status information of each transmission path.
[0038] ST2, dividing the historical status information into several segments and inputting them into the LSTM model as training samples.
[0039] ST3, the LSTM model extracts features and predicts each segment to obtain a predicted segment.
[0040] ST4, comparing the predicted segment with the next segment corresponding to the current segment, and calculating a loss function.
[0041] ST5, optimize the parameters of the LSTM model according to the loss function.
[0042] ST6, repeat steps ST3-ST5, and the LSTM model updates parameters every time a prediction fragment is obtained, until the parameters of the LSTM model converge or the number of parameter updates reaches a set threshold, and a trained LSTM model is obtained.
[0043] The quality assessment model 3 is used to perform step S3: evaluate the quality of each transmission path according to the future state information. The quality assessment model evaluates the quality of the transmission path according to the future state information predicted by the state prediction model, which includes but is not limited to mathematical models such as linear weighted model, rank-sum ratio model (RSR), K-Nearest Neighbor model (KNN), support vector machine model (SVM), decision tree model (DT), etc.
[0044] In this embodiment, the quality assessment model is a linear weighted model. The linear weighted model is used to perform the following steps: S31, obtaining the future state information of the transmission path predicted by the state prediction model, normalizing each indicator of the future state information, and obtaining normalized multiple indicator data. S32, calculating the weighted average of the normalized multiple indicator data according to the preset weights, and using the weighted average as the quality assessment result of the transmission path.
[0045] The data segmentation module 4 is used to execute step S4: according to the quality of the transmission path, the data packet is segmented into a plurality of data blocks, and the plurality of data blocks are allocated to a plurality of transmission paths. The data segmentation module 4 specifically includes: an optimal size calculation unit 41, a format conversion unit 42 and the path allocation unit 43.
[0046] The optimal size calculation unit 41 is used to perform step S41: calculate the optimal size of the data block according to the quality of the transmission path; and divide the data packet into a plurality of data blocks according to the optimal size of the data block. The optimal size of the data block will be adjusted in real time according to the load information and the predicted future state information, so as to adapt to the changing network conditions.
[0047] In one embodiment, the optimal size of the data block is calculated according to the quality of the transmission path, specifically: the average of the quality of all available transmission paths is calculated to obtain a network quality parameter; the network quality parameter is substituted into a preset function to calculate the optimal size of the data block. The preset function is an increasing function, and its specific shape can be set as needed. The value of the network quality parameter can reflect the quality of the network. When the network quality is high, the probability of data block loss and retransmission is low. At this time, a larger data block size can maximize the transmission efficiency. When the network quality is low, the probability of data block loss and retransmission is high. At this time, a smaller data block size can reduce the cost of retransmission. Therefore, the preset function of this embodiment is an increasing function, that is, the optimal size of the data block is positively correlated with the network quality parameter.
[0048] The format conversion unit 42 is used to execute step S42: convert the plurality of data blocks into a format that complies with a preset protocol to obtain data blocks that comply with the preset protocol. In this embodiment, step S42 includes: adding a unique identifier that complies with the preset protocol to each data block.
[0049] The path allocation unit 43 is used to execute step S43: according to the transmission priority of the data block and the quality of the transmission path, the data block that complies with the preset protocol is allocated to multiple transmission paths according to the load balancing principle. The data block is allocated to multiple transmission paths according to the load balancing principle, which specifically means: the data block with higher transmission priority is preferentially matched to the transmission path with higher quality. In one embodiment, a load balancing algorithm such as a weighted polling method and a weighted random method can be used to realize the allocation of data blocks to multiple transmission paths according to the load balancing principle.
[0050] The transmission module 5 is used to execute S5: transmitting a plurality of data blocks in parallel through a plurality of transmission paths. After receiving all the data blocks, the receiving end restores the original data packets according to a preset protocol, thereby completing the data transmission.
[0051] Example 2
[0052] See also Figure 4-5 , Figure 4 This is a unit diagram of a network anomaly monitoring and repairing module of a data transmission device for path parallel transmission according to Embodiment 2 of the present invention, Figure 5 The flowchart of the network anomaly monitoring and repairing steps of the data transmission method of path parallel transmission of embodiment 2 of the present invention is shown. The data transmission device of multi-path parallel transmission of embodiment 2 of the present invention adds a network anomaly monitoring and repairing module A on the basis of the data transmission device of path parallel transmission of embodiment 1, which includes: a network status monitoring unit A1, a baseline model A2, an abnormal path acquisition unit A3, a first reallocation unit A4, a recovery monitoring unit A5 and a second reallocation unit A6.
[0053] The network status monitoring unit A1 is used to execute step SA1: monitor the status information of the transmission path in real time; the status information includes the bandwidth utilization, delay and packet loss rate of the transmission path.
[0054] The baseline model A2 is used to execute step SA2: judging whether an abnormality occurs in the transmission path according to the status information of each transmission path.
[0055] The abnormal path acquisition unit A3 is used to execute step SA3: if an abnormality occurs in the transmission path, the abnormal transmission path is acquired.
[0056] The first reallocation unit A4 is used to execute step SA4: stop using the abnormal transmission path, enable the pre-planned redundant transmission path, and reallocate the data blocks to multiple transmission paths. In one embodiment, step SA4 is: SA41, replace the abnormal transmission path with the redundant transmission path, and reallocate the data blocks to multiple transmission paths based on the principle of load balancing. In one embodiment, step SA4 is: SA42, convert the data blocks originally transmitted by the abnormal transmission path to be transmitted by the redundant transmission path.
[0057] The recovery monitoring unit A5 is used to execute step SA5: continuously monitor the status information of the transmission path, and determine whether the abnormal transmission path has returned to normal according to the status information through a baseline model.
[0058] The second reallocation unit A6 is used to execute step SA6: if it is determined that the abnormal transmission path has returned to normal, then stop using the redundant transmission path, resume using the abnormal transmission path, and distribute the data blocks to multiple transmission paths. In one embodiment, step SA6 is: SA61, allowing the abnormal transmission path to replace the redundant transmission path, and redistribute the data blocks to multiple transmission paths based on the principle of load balancing. In one embodiment, step SA6 is: SA62, converting the data blocks originally transmitted by the redundant transmission path to be transmitted by the abnormal transmission path.
[0059] The baseline model is constructed by collecting state information of the transmission path under a number of normal / abnormal conditions. The baseline model can be any statistical model and / or machine learning model with classification function, which determines whether the transmission path is abnormal by comparing the difference between the state information under normal / abnormal conditions and the current state information.
[0060] In one embodiment, the baseline model is a support vector machine model, which is constructed by the following steps: collecting state information of the transmission path under normal and abnormal conditions as samples; vectorizing the state information of the transmission path under normal and abnormal conditions to obtain normal vectors and abnormal vectors respectively; solving a hyperplane equation that can divide the normal vectors and abnormal vectors into two categories based on the normal vectors and abnormal vectors; the hyperplane equation is the constructed support vector machine model. When judging whether the transmission path is abnormal, the current state information is vectorized to obtain the current vector; and the current vector is substituted into the hyperplane equation to obtain the calculation result; finally, according to the sign of the calculation result, it is judged whether the current state information belongs to the normal situation or the abnormal situation, so as to judge whether the transmission path is abnormal.
[0061] In one embodiment, the baseline model is an isolation forest model, which is constructed by collecting a large amount of state information of the transmission path under normal conditions as samples. When judging whether the transmission path is abnormal, both the state information of the transmission path under normal conditions and the current state information are vectorized to obtain a set of points in the data space; then the data space is recursively divided by a random hyperplane to generate multiple subspaces; the steps of dividing the data space are repeated until each subspace of the last level contains only one data point; according to the level of the subspace where the point corresponding to the current state information is located, it is judged whether the point corresponding to the current state information is an abnormal point, thereby judging whether the transmission path is abnormal.
[0062] The present invention has the following technical effects: 1. The present invention realizes multi-path parallel transmission, which greatly improves the data transmission efficiency compared with the existing single-path data transmission method. 2. The present invention can predict the future performance of the transmission path based on the state information of the transmission path, and evaluate the quality of the transmission path based on the predicted future performance. Finally, the data blocks are intelligently allocated to multiple transmission paths according to the quality of the transmission path, which greatly improves the intelligence of resource scheduling under the condition of limited network resources and further improves the data transmission efficiency. 3. The present invention also adds a network anomaly monitoring and repair mechanism, which can automatically start a redundant transmission path when an abnormality occurs in an individual transmission path, thereby reducing the impact of network anomalies and improving the robustness of data transmission.
[0063] Based on the same inventive concept, the present application also provides an electronic device, which may be a terminal device such as a server, a desktop computing device or a mobile computing device (e.g., a laptop computing device, a handheld computing device, a tablet computer, a netbook, etc.). The device includes one or more processors and a memory, wherein the processor is used to execute a program to implement the data transmission method of multi-path parallel transmission of an embodiment of the present invention; and the memory is used to store a computer program executable by the processor.
[0064] Based on the same inventive concept, the present application also provides a computer-readable storage medium, corresponding to the embodiment of the aforementioned multi-path parallel transmission data transmission method, and the computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the multi-path parallel transmission data transmission method recorded in any of the above-mentioned embodiments.
[0065] The present application may take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. Computer-usable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be achieved by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include but are not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.
[0066] The above-mentioned embodiments only express several implementation methods of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, and the present invention is also intended to include these modifications and modifications.
Claims
1. A data transmission method for multi-path parallel transmission, characterized in that: Includes steps: S1, obtaining data packets to be transmitted and status information of multiple transmission paths; S2, predicting future state information of each transmission path according to the state information; S3, evaluating the quality of each transmission path according to the future state information; S4, dividing the data packet into a plurality of data blocks according to the quality of the transmission path, and distributing the plurality of data blocks to a plurality of transmission paths; S5, transmitting several data blocks in parallel through multiple transmission paths.
2. The data transmission method of multi-path parallel transmission according to claim 1, characterized in that: It also includes the following steps for network anomaly monitoring and repair: SA1, real-time monitoring of the status information of the transmission path; SA2, through the baseline model, determines whether the transmission path is abnormal according to the status information of each transmission path; SA3, if it is determined that the transmission path is abnormal, obtain the abnormal transmission path; SA4, stop using the abnormal transmission path, enable the pre-planned redundant transmission path, and reallocate the data blocks to multiple transmission paths; SA5, continuously monitoring the status information of the transmission path, and judging whether the abnormal transmission path has returned to normal according to the status information through a baseline model; SA6: If it is determined that the abnormal transmission path has returned to normal, stop using the redundant transmission path, resume enabling the abnormal transmission path, and reallocate the data blocks to multiple transmission paths.
3. The data transmission method of multi-path parallel transmission according to claim 2, characterized in that: The baseline model is the isolation forest model; Step SA2 specifically includes: vectorizing a large amount of state information and current state information of transmission paths under normal conditions through the isolation forest model to obtain a set of points in the data space; then recursively dividing the data space through random hyperplanes to generate multiple subspaces; repeating the steps of dividing the data space until each subspace of the last level contains only one data point; judging whether the point corresponding to the current state information is an abnormal point according to the level of the subspace where the point corresponding to the current state information is located, thereby judging whether the transmission path is abnormal.
4. The data transmission method of multi-path parallel transmission according to claim 1, characterized in that: Step S2 specifically includes: predicting the future state information of each transmission path according to the state information through the LSTM model; The LSTM model is trained through the following steps: ST1, obtain the historical status information of each transmission path; ST2, dividing the historical state information into several segments and inputting them into the LSTM model as training samples; ST3, the LSTM model extracts features and predicts each segment to obtain a predicted segment; ST4, comparing the predicted segment with the next segment corresponding to the current segment, and calculating a loss function; ST5, optimizing the parameters of the LSTM model according to the loss function; ST6, repeat steps ST3-ST5, and the LSTM model updates parameters every time a prediction fragment is obtained, until the parameters of the LSTM model converge or the number of parameter updates reaches a set threshold, and a trained LSTM model is obtained.
5. The data transmission method of multi-path parallel transmission according to claim 1, characterized in that: Step S3 specifically includes: S31, normalizing each indicator of the future state information to obtain normalized indicator data; S32, calculating a weighted average of a plurality of normalized indicator data according to preset weights, and taking the weighted average as a quality evaluation result of the transmission path.
6. The data transmission method of multi-path parallel transmission according to claim 1, characterized in that: Step S4 specifically includes: S41, calculating an optimal size of a data block according to the quality of the transmission path; and dividing the data packet into a plurality of data blocks according to the optimal size of the data block; S42, converting the plurality of data blocks into a format conforming to a preset protocol to obtain data blocks conforming to the preset protocol; S43, allocating the data blocks conforming to the preset protocol to multiple transmission paths according to the transmission priority of the data blocks and the quality of the transmission path and in accordance with the load balancing principle.
7. The data transmission method of multi-path parallel transmission according to claim 6, characterized in that: Step S41 specifically includes: calculating the average of the qualities of all available transmission paths to obtain a network quality parameter; substituting the network quality parameter into a preset function to calculate the optimal size of the data block; the preset function is an increasing function.
8. The data transmission method of multi-path parallel transmission according to claim 7, characterized in that: In step S43, data blocks are allocated to multiple transmission paths according to the load balancing principle, specifically: data blocks with higher transmission priority are preferentially allocated to transmission paths with higher quality.
9. A data transmission device for multi-path parallel transmission, characterized in that: include: A data acquisition module, used to acquire data packets to be transmitted and status information of multiple transmission paths; A state prediction model, used to predict future state information of each transmission path according to the state information; a quality assessment model, used for assessing the quality of each transmission path according to the future state information; A data segmentation module, used for segmenting the data packet into a plurality of data blocks according to the quality of the transmission path, and distributing the plurality of data blocks to a plurality of transmission paths; The transmission module is used to transmit a plurality of data blocks in parallel through a plurality of transmission paths.
10. An electronic device, characterized in that: include: processor; a memory for storing a computer program executed by the processor; Wherein, when the processor executes the computer program, it implements any one of the multi-path parallel transmission data transmission methods described in claims 1-8.
Citation Information
Patent Citations
Link congestion detection method and related equipment
CN113923164A
Abnormal link detection method and device, equipment and storage medium
CN114793205A
5G communication abnormity early warning method and system based on state detection
CN115022916A
Garment workshop intelligent management method and system
CN115576284A
Multi-path transmission congestion control method and device
CN115834489A
Cited By
Content distribution network data transmission method and device
CN119814649A
Network data transmission method, device and equipment based on multi-path parallel transmission and medium
CN120528858A
Large map data transmission method
CN120687537A
A method for large map data transmission
CN120687537B
