Data transmission method and system based on data analysis

Abstract

The present invention relates to the field of computer technology, and in particular to a data analysis-based data transmission method and system. The method obtains a reference data volume sequence by predicting an uploaded data volume sequence, analyzes the reference data volume sequence of each sending end, and determines a reference time point for priority adjustment, thereby improving the timeliness of priority adjustment. The method determines the reference priority of each sending end based on the statistical data of each sending end, and then obtains the target data ratio of each sending end through mapping function, and then performs bandwidth allocation. In this way, bandwidth allocation can be performed based on the statistical data. Compared with the existing technology, the method can timely perceive the change of data volume and then accurately allocate bandwidth to the sending end that actually has demand, thereby improving data transmission efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of data transmission, and in particular to a data transmission method and system based on data analysis. Background Art

[0002] In the data transmission sector, with the rapid development of cloud computing, the Internet of Things, high-definition video streaming, and other services, scenarios where multiple senders simultaneously upload data to the same transmission path are becoming increasingly common. Traditional transmission bandwidth allocation methods, which often use fixed allocations or simple priority allocation strategies, are unable to adapt to the needs of dynamically changing data volumes.

[0003] On the one hand, the data volume of some senders will fluctuate greatly at different time points. If a fixed allocation method is adopted, it is easy to lead to waste of bandwidth resources or excessive transmission delays at some senders. On the other hand, the allocation strategy based on static priority has difficulty in sensing changes in data volume in real time, and cannot accurately allocate bandwidth to senders that really have demand, resulting in low data transmission efficiency and poor user experience.

[0004] Although existing technologies have proposed methods for setting dynamic priorities, the main purpose of dynamic priority setting is to ensure that each sender has the opportunity to obtain high priority, so as to avoid the high priority continuously occupying the bandwidth of the transmission path under the static priority strategy. However, dynamic priority setting still cannot accurately allocate bandwidth to senders that actually have demand, resulting in poor data transmission efficiency.

[0005] Therefore, how to improve data transmission efficiency has become an urgent problem to be solved. Summary of the Invention

[0006] In order to solve the above technical problems, the present invention adopts a data transmission method based on data analysis, which includes the following steps:

[0007] S101, obtaining upload data amount sequences corresponding to M sending ends, wherein the upload data amount sequence includes upload data amounts corresponding to a start time point to a target time point, and M is a positive integer.

[0008] S102: For any transmitting end, the uploaded data volume sequence corresponding to the transmitting end is input into a trained data volume prediction model to predict a reference data volume sequence corresponding to the transmitting end.

[0009] S103: Determine a data volume change time point corresponding to the sending end according to a reference data volume sequence corresponding to the sending end.

[0010] S104: Determine a reference time point according to the data volume change time points corresponding to the respective sending ends.

[0011] S105 , determining the statistical data amount corresponding to each transmitting end according to the reference time point and the reference data amount sequence corresponding to each transmitting end.

[0012] S106: Determine the reference priority corresponding to each transmitting end according to the statistical data corresponding to each transmitting end.

[0013] S107 , mapping the reference priorities corresponding to the respective transmitting ends to target data proportions corresponding to the respective transmitting ends according to a preset mapping function.

[0014] S108, obtaining the total bandwidth of the transmission path, and determining the target bandwidth corresponding to each transmitting end according to the target data ratio corresponding to each transmitting end and the total bandwidth, wherein the target bandwidth is used to determine the amount of data that can be transmitted by the corresponding transmitting end through the transmission path per unit time.

[0015] The present invention also provides a data analysis-based data transmission system, the data analysis-based data transmission system comprising:

[0016] The sequence acquisition module is used to obtain the upload data volume sequences corresponding to M sending ends, wherein the upload data volume sequence includes the upload data volumes corresponding to the start time point to the target time point, and M is a positive integer.

[0017] The sequence prediction module is used to input the uploaded data volume sequence corresponding to any sending end into the trained data volume prediction model to predict the reference data volume sequence corresponding to the sending end.

[0018] The sequence analysis module is used to determine the data volume change time point corresponding to the sending end according to the reference data volume sequence corresponding to the sending end.

[0019] The time point determination module is used to change the time point according to the data volume corresponding to each sending end and determine the reference time point.

[0020] The data volume statistics module is used to determine the statistical data volume corresponding to each sending end according to the reference time point and the reference data volume sequence corresponding to each sending end.

[0021] The priority determination module is used to determine the reference priority corresponding to each sending end according to the statistical data corresponding to each sending end.

[0022] The ratio mapping module is used to map the reference priority corresponding to each sending end to the target data ratio corresponding to each sending end according to a preset mapping function.

[0023] The transmission control module is used to obtain the total bandwidth of the transmission path and determine the target bandwidth corresponding to each transmitting end based on the target data ratio corresponding to each transmitting end and the total bandwidth. The target bandwidth is used to determine the amount of data that the corresponding transmitting end can transmit through the transmission path per unit time.

[0024] The present invention has at least the following beneficial effects: a reference data volume sequence is obtained by predicting the uploaded data volume sequence, the reference data volume sequence of each sending end is analyzed, and a reference time point for priority adjustment is determined, thereby improving the timeliness of priority adjustment; based on the statistical data of each sending end, a reference priority of each sending end is determined, and then a target data ratio of each sending end is obtained by mapping through a mapping function, and then bandwidth allocation is performed, so that bandwidth allocation can be performed based on statistical data. Compared with the existing technology, the change in data volume can be perceived in a timely manner, and the bandwidth can be accurately allocated to the sending end that actually has demand, thereby improving data transmission efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 A flowchart of a data transmission method based on data analysis provided in Example 1 of the present invention;

[0027] Figure 2 A structural diagram of a data transmission system based on data analysis provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0028] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It is understood that, where appropriate, the above-mentioned terms used to distinguish similar objects can be interchanged so that the present invention can also implement other embodiments other than the above-mentioned illustrated embodiments or described embodiments. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] Example 1

[0031] This embodiment provides a data transmission method based on data analysis, such as Figure 1 FIG. 1 is a flow chart of a data transmission method based on data analysis provided in Embodiment 1 of the present invention. The data transmission method based on data analysis includes the following steps:

[0032] S101, obtaining a sequence of uploaded data volumes corresponding to M sending ends, wherein the sequence of uploaded data volumes includes the uploaded data volumes corresponding to the start time point to the target time point, and M is a positive integer;

[0033] S102: For any transmitting end, input the uploaded data volume sequence corresponding to the transmitting end into the trained data volume prediction model to predict a reference data volume sequence corresponding to the transmitting end;

[0034] S103, determining a data volume change time point corresponding to the sending end based on a reference data volume sequence corresponding to the sending end;

[0035] S104, determining a reference time point based on the data volume change time points corresponding to each sending end;

[0036] S105, determining the statistical data amount corresponding to each transmitting end according to the reference time point and the reference data amount sequence corresponding to each transmitting end;

[0037] S106, determining a reference priority corresponding to each transmitting end based on the statistical data corresponding to each transmitting end;

[0038] S107, mapping the reference priorities corresponding to the respective transmitting ends to target data ratios corresponding to the respective transmitting ends according to a preset mapping function;

[0039] S108, obtaining the total bandwidth of the transmission path, and determining the target bandwidth corresponding to each transmitting end according to the target data ratio corresponding to each transmitting end and the total bandwidth, wherein the target bandwidth is used to determine the amount of data that can be transmitted by the corresponding transmitting end through the transmission path per unit time.

[0040] Among them, the sending end uploads data to the transmission path, the transmission path corresponds to M sending ends, the starting time point and the target time point can determine the collection time period, the determined collection time period corresponds to several collection time points, accordingly, the starting time point corresponds to the first collection time point in the collection time period, the target time point corresponds to the last collection time point in the collection time period, and the amount of uploaded data corresponding to the starting time point to the target time point can refer to the amount of uploaded data corresponding to each collection time point in the collection time period.

[0041] The data volume prediction model can adopt a time series prediction model, and the time series prediction model can use a recurrent neural network model, a long short-term memory network model, a time domain convolution model, etc.

[0042] The reference data volume sequence can be formed by the predicted reference data volume. The data volume change time point can refer to the time point when the reference data volume of the corresponding sending end changes significantly. The reference time point can refer to the time point for setting the reference priority determined by comprehensively considering the data volume change time points of multiple sending ends. The statistical data volume can represent the sum of the reference data volumes of the corresponding sending end from the target time point to the reference time point. The reference priority can be used to characterize the priority of the corresponding sending end in using the transmission path.

[0043] Specifically, in this embodiment, the target data ratio corresponding to each transmitting end is determined according to the reference priority corresponding to each transmitting end, that is, the total bandwidth of the transmission path is used to determine the target bandwidth corresponding to each transmitting end according to the reference priority, so that each transmitting end can upload data at any point in time, and the data upload rate of each transmitting end is different, thereby avoiding the situation under the existing priority allocation strategy that the higher priority transmitting end preempts the transmission path and the lower priority transmitting end is unable to upload data, thereby improving the data transmission efficiency of the lower priority transmitting end.

[0044] In a specific embodiment, inputting the uploaded data amount sequence corresponding to the sending end into a trained data amount prediction model to predict a reference data amount sequence corresponding to the sending end includes:

[0045] Inputting the uploaded data volume sequence corresponding to the sending end into the trained data volume prediction model to predict the reference data volumes corresponding to K preset time points after the target time point, where K is a positive integer;

[0046] The K reference data amounts corresponding to the transmitting end form a reference data amount sequence corresponding to the transmitting end.

[0047] The uploaded data volume sequence includes uploaded data volumes corresponding to S collection time points, where S can be a positive integer.

[0048] Specifically, the uploaded data volumes corresponding to the 1st collection time point to the Sth collection time point are respectively input into the trained data volume prediction model to predict the reference data volume corresponding to the 1st preset time point; the uploaded data volumes corresponding to the 2nd collection time point to the Sth collection time point and the reference data volume corresponding to the 1st preset time point are respectively input into the trained data volume prediction model to predict the reference data volume corresponding to the 2nd preset time point; the uploaded data volumes corresponding to the 3rd collection time point to the Sth collection time point and the reference data volume corresponding to the 1st preset time point to the 2nd collection time point are respectively input into the trained data volume prediction model to predict the reference data volume corresponding to the 3rd preset time point; and so on, until the reference data volume corresponding to the Kth preset time point is obtained, and then a reference data volume sequence is obtained.

[0049] It should be noted that the time interval between adjacent collection time points, the time interval between adjacent collection time points and preset time points, and the time interval between adjacent preset time points are all the same.

[0050] In a specific embodiment, determining the data volume change time point corresponding to the transmitting end according to the reference data volume sequence corresponding to the transmitting end includes:

[0051] The first preset time point after the target time point is used as the sliding window starting point;

[0052] Based on the sliding window starting point, a sliding window of length P is used to slide the reference data sequence corresponding to the transmitting end according to a preset step size, and an average calculation is performed on each sliding to obtain K-P+1 average calculation results, where P is a positive integer less than K;

[0053] When the ratio of the absolute value of the difference between the Qth mean calculation result and the Q-1th mean calculation result to the Q-1th mean calculation result is greater than a preset ratio threshold, determining the preset time point corresponding to the Qth mean calculation result as a temporary time point, where Q is an integer in the range of [2, K-P+1];

[0054] If there is a temporary time point, the first determined temporary time point is used as the data volume change time point corresponding to the sending end;

[0055] If there is no temporary time point, the Kth preset time point after the target time point is determined as the data volume change time point corresponding to the sending end.

[0056] The preset step size may be 1, and the preset ratio threshold may be 0.2.

[0057] Specifically, this embodiment determines the temporary time point by the degree of change of the mean calculation result, and adopts a sliding window method. During the sliding window movement, the amount of reference data contained in the sliding window may overlap, so that the preset time point at which the reference data amount undergoes a sudden change can be determined as the preset time point.

[0058] In a specific embodiment, determining the statistical data amount corresponding to each transmitting end according to the reference time point and the reference data amount sequence corresponding to each transmitting end includes:

[0059] For any transmitting end, summing the reference data amounts corresponding to the first preset time point after the target time point to the reference time point in the reference data amount sequence corresponding to the transmitting end, and using the summed result as the statistical data amount corresponding to the transmitting end;

[0060] Traverse each sending end and obtain the statistical data corresponding to each sending end.

[0061] In a specific implementation, after step S104, the method further includes:

[0062] For any preset time point after the reference time point, summing the reference data amounts corresponding to the respective transmitting ends at the preset time point to obtain a reference statistic corresponding to the preset time point;

[0063] Updating the target time point to a preset time point corresponding to the minimum reference statistic;

[0064] Update the starting time point according to the number of acquisition time points and the target time point;

[0065] When the current time point is the target time point, the process returns to step S101 .

[0066] Among them, the target time point is updated to the preset time point corresponding to the minimum reference statistic, so as to ensure that the change of reference priority will not have a significant impact on the data upload process of each sending end, and ensure that the data upload process of each sending end will not exceed the total allocated bandwidth due to the change of reference priority.

[0067] Specifically, the number of collection time points is set to S, and the collection time point corresponding to the target time point is moved forward by S collection time points in chronological order. The collection time point corresponding to the moving result is used as the starting time point, and then the process returns to step S101, so that the reference priority is updated at a higher frequency, thereby ensuring the timeliness of data volume perception and improving the efficiency of data transmission.

[0068] In a specific implementation, determining the reference time point according to the data volume change time points corresponding to the respective sending ends includes:

[0069] A preset clustering algorithm is used to cluster the data volume change time points corresponding to each sending end to obtain several data volume change time points corresponding to noise points;

[0070] The reference time point is obtained by performing mean calculation based on the time points of data amount change that do not correspond to the noise points.

[0071] The preset clustering algorithm may be a density-based spatial clustering algorithm (Density-Based Spatial Clustering of Applications with Noise, DBSCAN), which can determine a time point at which the amount of data belonging to a noise point changes.

[0072] In a specific implementation, determining the reference priority corresponding to each transmitting end according to the statistical data corresponding to each transmitting end includes:

[0073] Calculate the absolute value of the difference between each two statistical data quantities to obtain several intermediate differences;

[0074] Determine the smallest middle difference as the benchmark value;

[0075] Setting the reference priority of the transmitting end corresponding to the maximum statistical data amount to a first preset value;

[0076] For any transmitter for which a reference priority is not set, calculating a temporary difference between the maximum statistical data amount and the statistical data amount corresponding to the transmitter;

[0077] determining a priority adjustment value corresponding to the sending end according to a ratio of the temporary difference to the reference value;

[0078] The priority adjustment value corresponding to the sending end is added to the first preset value, and the addition result is used as the reference priority corresponding to the sending end.

[0079] The first preset value may be 1, and the priority adjustment value corresponding to the transmitting end is obtained by rounding up the ratio of the temporary difference to the reference value.

[0080] Specifically, this embodiment determines a reference value and determines the priority adjustment value corresponding to the sending end based on the ratio of the temporary difference to the reference value. This can characterize the difference in statistical data quantity through the level difference of the reference priority, so that the target data ratio obtained according to the reference priority mapping is associated with the statistical data quantity. Compared with the prior art method of allocating priorities according to the sorting results, the setting of the target data ratio is more flexible and more closely related to the data volume. When allocating bandwidth, it can better allocate target bandwidth to the sending end according to actual needs, thereby improving data transmission efficiency.

[0081] In a specific embodiment, mapping the reference priorities corresponding to the respective transmitting ends to the target data ratios corresponding to the respective transmitting ends according to a preset mapping function includes:

[0082] According to the preset mapping function, mapping the reference priorities corresponding to the respective transmitting ends to target mapping values ​​corresponding to the respective transmitting ends;

[0083] The target mapping values ​​corresponding to each sending end are normalized to obtain the target data ratio corresponding to each sending end.

[0084] Among them, the preset mapping function is set to: m =-log(r×x m +1)-1, where x m is the reference priority corresponding to the mth transmitter, y m is the target mapping value corresponding to the mth transmitting end, and r is the adjustment coefficient. In this embodiment, r can be set to 0.5. The implementer can update the adjustment coefficient according to actual conditions.

[0085] Specifically, the preset mapping function maps a larger reference priority to a larger value and a smaller reference priority to a smaller value, and as the reference priority decreases, the difference in mapping values ​​between adjacent reference priorities also decreases, so that the target bandwidth allocated to the transmitter corresponding to the larger reference priority can be significantly better than the target bandwidth allocated to the transmitter corresponding to the smaller reference priority, thereby ensuring the upload efficiency of key data, and the target bandwidth allocated to the transmitter corresponding to the smaller reference priority is similar, to ensure that each transmitter can be allocated an approximate target bandwidth to support normal data transmission of each transmitter.

[0086] In a specific embodiment, obtaining the total bandwidth of the transmission path and determining the target bandwidth corresponding to each transmitting end according to the target data ratio corresponding to each transmitting end and the total bandwidth includes:

[0087] For any transmitting end, multiply the target data ratio corresponding to the transmitting end by the total bandwidth to obtain the target bandwidth corresponding to the transmitting end;

[0088] Traverse all sending ends and obtain the target bandwidth corresponding to each sending end.

[0089] The sum of the target data ratios corresponding to each sending end is 1.

[0090] In the first embodiment of the present invention, a reference data volume sequence is obtained by predicting the uploaded data volume sequence, and the reference data volume sequence of each sending end is analyzed to determine the reference time point for priority adjustment, thereby improving the timeliness of priority adjustment. The reference priority of each sending end is determined based on the statistical data of each sending end, and then the target data ratio of each sending end is obtained through mapping function, and then bandwidth allocation is performed. In this way, bandwidth allocation can be performed based on statistical data. Compared with the existing technology, changes in data volume can be perceived in a timely manner, and bandwidth can be accurately allocated to sending ends that actually have demand, thereby improving data transmission efficiency.

[0091] Example 2

[0092] This embodiment 2 provides a data transmission system based on data analysis, such as Figure 2 FIG. 1 is a schematic diagram of a data transmission system based on data analysis according to a second embodiment of the present invention. The data transmission system based on data analysis includes:

[0093] The sequence acquisition module 201 is configured to acquire a sequence of uploaded data volumes corresponding to M transmitting ends, wherein the sequence of uploaded data volumes includes the uploaded data volumes corresponding to the start time point to the target time point, and M is a positive integer;

[0094] The sequence prediction module 202 is used to input the uploaded data volume sequence corresponding to any sending end into the trained data volume prediction model to predict the reference data volume sequence corresponding to the sending end;

[0095] A sequence analysis module 203 is configured to determine a data volume change time point corresponding to the sending end based on a reference data volume sequence corresponding to the sending end;

[0096] A time point determination module 204 is configured to determine a reference time point based on the data volume change time points corresponding to each sending end;

[0097] The data volume statistics module 205 is configured to determine the statistical data volume corresponding to each transmitting end according to the reference time point and the reference data volume sequence corresponding to each transmitting end;

[0098] The priority determination module 206 is configured to determine a reference priority corresponding to each transmitting end according to the statistical data corresponding to each transmitting end;

[0099] A ratio mapping module 207 is configured to map the reference priorities corresponding to the respective transmitting ends to target data ratios corresponding to the respective transmitting ends according to a preset mapping function;

[0100] The transmission control module 208 is used to obtain the total bandwidth of the transmission path and determine the target bandwidth corresponding to each transmitting end based on the target data ratio corresponding to each transmitting end and the total bandwidth. The target bandwidth is used to determine the amount of data that the corresponding transmitting end can transmit through the transmission path per unit time.

[0101] It should be noted that the specific limitations of the data analysis-based data transmission system can be found in the limitations of the data analysis-based data transmission method described above and will not be repeated here. The information interaction and execution process between the above modules, etc., are based on the same concept as the method embodiments of the present invention. Their specific functions and technical effects can be found in the method embodiments and will not be repeated here.

[0102] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any form. Although the present invention has been disclosed as above in terms of preferred embodiments, they are not intended to limit the present invention. Any technician familiar with this profession can make some changes or modifications to equivalent embodiments of equivalent changes using the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A data transmission method based on data analysis, characterized in that: The data transmission method based on data analysis comprises the following steps: S101, obtaining a sequence of uploaded data volumes corresponding to M sending ends, wherein the sequence of uploaded data volumes includes the uploaded data volumes corresponding to the start time point to the target time point, and M is a positive integer; S102: For any transmitting end, input the uploaded data volume sequence corresponding to the transmitting end into the trained data volume prediction model to predict a reference data volume sequence corresponding to the transmitting end; S103, determining a data volume change time point corresponding to the sending end based on a reference data volume sequence corresponding to the sending end; S104, determining a reference time point based on the data volume change time points corresponding to each sending end; S105, determining the statistical data amount corresponding to each transmitting end according to the reference time point and the reference data amount sequence corresponding to each transmitting end; S106, determining a reference priority corresponding to each transmitting end based on the statistical data corresponding to each transmitting end; S107, mapping the reference priorities corresponding to the respective transmitting ends to target data ratios corresponding to the respective transmitting ends according to a preset mapping function; S108, obtaining the total bandwidth of the transmission path, and determining the target bandwidth corresponding to each transmitting end according to the target data ratio corresponding to each transmitting end and the total bandwidth, wherein the target bandwidth is used to determine the amount of data that can be transmitted by the corresponding transmitting end through the transmission path per unit time.

2. The data transmission method based on data analysis according to claim 1, characterized in that: The step of inputting the uploaded data amount sequence corresponding to the sending end into the trained data amount prediction model to predict a reference data amount sequence corresponding to the sending end includes: Inputting the uploaded data volume sequence corresponding to the sending end into the trained data volume prediction model to predict the reference data volumes corresponding to K preset time points after the target time point, where K is a positive integer; The K reference data amounts corresponding to the transmitting end form a reference data amount sequence corresponding to the transmitting end.

3. The data transmission method based on data analysis according to claim 2, characterized in that: The determining, based on the reference data amount sequence corresponding to the transmitting end, the data amount change time point corresponding to the transmitting end includes: The first preset time point after the target time point is used as the sliding window starting point; Based on the sliding window starting point, a sliding window of length P is used to slide the reference data sequence corresponding to the transmitting end according to a preset step size, and an average calculation is performed on each sliding to obtain K-P+1 average calculation results, where P is a positive integer less than K; When the ratio of the absolute value of the difference between the Qth mean calculation result and the Q-1th mean calculation result to the Q-1th mean calculation result is greater than a preset ratio threshold, determining the preset time point corresponding to the Qth mean calculation result as a temporary time point, where Q is an integer in the range of [2, K-P+1]; If there is a temporary time point, the first determined temporary time point is used as the data volume change time point corresponding to the sending end; If there is no temporary time point, the Kth preset time point after the target time point is determined as the data volume change time point corresponding to the sending end.

4. The data transmission method based on data analysis according to claim 2, characterized in that: The determining, based on the reference time point and the reference data amount sequence corresponding to each transmitting end, the statistical data amount corresponding to each transmitting end includes: For any transmitting end, summing the reference data amounts corresponding to the first preset time point after the target time point to the reference time point in the reference data amount sequence corresponding to the transmitting end, and using the summed result as the statistical data amount corresponding to the transmitting end; Traverse each sending end and obtain the statistical data corresponding to each sending end.

5. The data transmission method based on data analysis according to claim 2, characterized in that: After step S104, the method further includes: For any preset time point after the reference time point, summing the reference data amounts corresponding to the respective transmitting ends at the preset time point to obtain a reference statistic corresponding to the preset time point; Updating the target time point to a preset time point corresponding to the minimum reference statistic; Update the starting time point according to the number of acquisition time points and the target time point; When the current time point is the target time point, the process returns to step S101 .

6. The data transmission method based on data analysis according to claim 1, characterized in that: The step of determining the reference time point according to the data volume change time points corresponding to the respective transmitting ends includes: A preset clustering algorithm is used to cluster the data volume change time points corresponding to each sending end to obtain several data volume change time points corresponding to noise points; The reference time point is obtained by performing mean calculation based on the time points of data amount change that do not correspond to the noise points.

7. The data transmission method based on data analysis according to claim 1, characterized in that: The determining, based on the statistical data corresponding to each transmitting end, the reference priority corresponding to each transmitting end, includes: Calculate the absolute value of the difference between each two statistical data quantities to obtain several intermediate differences; Determine the smallest middle difference as the benchmark value; Setting the reference priority of the transmitting end corresponding to the maximum statistical data amount to a first preset value; For any transmitter for which a reference priority is not set, calculating a temporary difference between the maximum statistical data amount and the statistical data amount corresponding to the transmitter; determining a priority adjustment value corresponding to the sending end according to a ratio of the temporary difference to the reference value; The priority adjustment value corresponding to the sending end is added to the first preset value, and the addition result is used as the reference priority corresponding to the sending end.

8. The data transmission method based on data analysis according to claim 1, characterized in that: Mapping the reference priorities corresponding to the respective transmitting ends to the target data ratios corresponding to the respective transmitting ends according to the preset mapping function includes: According to the preset mapping function, the reference priority corresponding to each transmitting end is mapped to the target mapping value corresponding to each transmitting end, wherein the preset mapping function is: m =-log(r×x m +1)-1, where x m is the reference priority corresponding to the mth transmitter, y m is the target mapping value corresponding to the mth sender, and r is the adjustment coefficient; The target mapping values ​​corresponding to each sending end are normalized to obtain the target data ratio corresponding to each sending end.

9. The data transmission method based on data analysis according to claim 1, characterized in that: Obtaining the total bandwidth of the transmission path, and determining the target bandwidth corresponding to each sending end according to the target data ratio corresponding to each sending end and the total bandwidth, including: For any transmitting end, multiply the target data ratio corresponding to the transmitting end by the total bandwidth to obtain the target bandwidth corresponding to the transmitting end; Traverse all sending ends and obtain the target bandwidth corresponding to each sending end.

10. A data transmission system based on data analysis, characterized in that: The data analysis-based data transmission system includes: A sequence acquisition module is used to obtain an upload data volume sequence corresponding to M sending ends, wherein the upload data volume sequence includes the upload data volume corresponding to the start time point to the target time point, and M is a positive integer; The sequence prediction module is used to input the uploaded data volume sequence corresponding to any sending end into the trained data volume prediction model to predict the reference data volume sequence corresponding to the sending end; A sequence analysis module, configured to determine a data volume change time point corresponding to the sending end based on a reference data volume sequence corresponding to the sending end; A time point determination module is used to change the time point according to the data volume corresponding to each sending end and determine the reference time point; A data volume statistics module, configured to determine the statistical data volume corresponding to each sending end according to the reference time point and the reference data volume sequence corresponding to each sending end; A priority determination module, configured to determine a reference priority corresponding to each transmitting end based on statistical data corresponding to each transmitting end; A ratio mapping module, configured to map the reference priorities corresponding to the respective transmitting ends to target data ratios corresponding to the respective transmitting ends according to a preset mapping function; The transmission control module is used to obtain the total bandwidth of the transmission path and determine the target bandwidth corresponding to each transmitting end based on the target data ratio corresponding to each transmitting end and the total bandwidth. The target bandwidth is used to determine the amount of data that the corresponding transmitting end can transmit through the transmission path per unit time.

Images