Network speed measurement method and device, equipment and storage medium

Abstract

The invention provides a network speed measurement method and device, equipment and a storage medium, and the method comprises the steps: carrying out the bandwidth prediction of a network to be subjected to speed measurement based on a bottleneck bandwidth and round-trip propagation time congestion control model according to a plurality of detection data packets with preset data sizes, so as to obtain a bandwidth prediction result of the network to be subjected to speed measurement; according to the bandwidth prediction result, configuring sampling parameters of the network to be subjected to speed measurement; according to the sampling parameters, carrying out sampling processing on the network with the speed to be measured so as to obtain a plurality of sampling speed values; calculating a real-time statistical confidence coefficient according to the sampling speed value; when the real-time statistical confidence reaches a confidence threshold, stopping sampling, and determining a speed measurement result according to a sampling speed value and a preset speed measurement prediction model; therefore, the convenience of network speed measurement and the data acquisition efficiency are improved.

Description

Technical Field

[0001] This application relates to the field of communication engineering, and in particular to a network speed measurement method, apparatus, device and storage medium. Background Technology

[0002] With the rapid deployment of 5G / 6G networks, users' demand for real-time network quality awareness is growing. For example, when users frequently switch between Wi-Fi and cellular networks in mobile scenarios (such as subway commutes or outdoor travel), they need to quickly assess the current network's bandwidth, latency, and stability to optimize the experience of video streaming, online gaming, or remote conferencing. Furthermore, operators need to dynamically adjust network resource allocation based on large-scale user speed test data; therefore, a network speed testing method is urgently needed.

[0003] In existing technologies, network speed testing methods mainly rely on chunked transmission and dynamic bandwidth detection mechanisms. For example, the speed testing tool (SpeedTest) dynamically creates multiple concurrent requests using Extensible Markup Language Hypertext Transfer Request (XMLHttpRequest) and uses incremental data blocks (downloads gradually increase from 50KB to 20MB, while uploads are in fixed 4MB chunks) to evaluate bandwidth.

[0004] However, existing technologies suffer from excessive bandwidth consumption, long testing times, poor cross-network adaptability, insufficient dynamic adjustment, and a contradiction between prediction accuracy and stability. These shortcomings collectively lead to technical problems such as poor convenience of network speed testing and low data acquisition efficiency in existing technologies. Summary of the Invention

[0005] The network speed testing method, apparatus, equipment, and storage medium provided in this application are intended to solve the technical problems of poor convenience and low data acquisition efficiency in the existing technology.

[0006] Firstly, this application provides a network speed testing method, including:

[0007] Based on multiple probe data packets of preset data size, and using a bottleneck bandwidth and round-trip time congestion control model, bandwidth prediction is performed on the network to be tested to obtain the bandwidth prediction result of the network to be tested.

[0008] Based on the bandwidth prediction results, configure the sampling parameters of the network to be tested; the sampling parameters include a confidence threshold to characterize the termination of sampling, and the sampling parameters also include the sampling interval and / or the sampling data packet size;

[0009] Based on the sampling parameters, the velocity network to be measured is sampled to obtain multiple sampled velocity values;

[0010] Calculate the real-time statistical confidence level based on the sampling rate value;

[0011] Sampling stops when the real-time statistical confidence level reaches the confidence threshold, and the speed measurement result is determined based on the sampling speed value and the preset speed measurement prediction model.

[0012] In one possible implementation, based on multiple probe data packets of preset data sizes and a bottleneck bandwidth and round-trip time congestion control model, bandwidth prediction is performed on the network to be tested to obtain the bandwidth prediction result of the network to be tested, including:

[0013] The probe data packets are transmitted sequentially in ascending order of data size, and the round-trip latency and data transmission rate of each transmission are obtained.

[0014] The bandwidth prediction results are obtained by calculating based on the bottleneck bandwidth and round-trip propagation time congestion control model, round-trip delay, and data transmission rate.

[0015] In one possible implementation, bandwidth prediction results are obtained by performing calculations based on a bottleneck bandwidth and round-trip time congestion control model, round-trip delay, and data transmission rate, including:

[0016] Based on the bottleneck bandwidth and round-trip propagation time congestion control model, round-trip delay and data transmission rate, the theoretical bandwidth limit and the effective bandwidth after delay correction are calculated.

[0017] The smaller of the theoretical bandwidth limit and the effective bandwidth is determined as the bandwidth prediction result.

[0018] In one possible implementation, the probe data packet includes a first-level probe packet, a second-level probe packet, and a third-level probe packet with sequentially increasing data sizes; wherein...

[0019] The first-level probe packet has a data size of 10KB, the second-level probe packet has a data size of 100KB, and the third-level probe packet has a data size of 1MB.

[0020] In one possible implementation, the sampling parameters of the network to be tested are configured based on the bandwidth prediction results, including:

[0021] Based on the bandwidth prediction results, the network level of the network to be tested is determined; the network level includes a first network level, a second network level, and a third network level, where the bandwidth of the first network level is greater than that of the second network level, and the bandwidth of the second network level is greater than that of the third network level.

[0022] Configure the sampling parameters of the network to be tested according to the network level; wherein the sampling parameters of the first network level are less than the sampling parameters of the second network level, and the sampling parameters of the second network level are less than the sampling parameters of the third network level.

[0023] In one possible implementation, the speed measurement result is determined based on the sampling speed value and a preset speed measurement prediction model, including:

[0024] Determine the input parameters based on the network level corresponding to the bandwidth prediction results;

[0025] Input parameters are fed into a preset speed prediction model to output a first predicted speed based on the preset speed prediction model; wherein, the preset speed prediction model is a lightweight neural network model.

[0026] Calculate the second estimated velocity based on the sampling velocity value;

[0027] The first predicted speed and the second estimated speed are weighted and fused to obtain the speed measurement result.

[0028] In one possible implementation, the input parameters include round-trip time, packet loss rate, latency jitter, and historical speed values.

[0029] In one possible implementation, the real-time statistical confidence level is calculated based on the sampling rate value, including:

[0030] Calculate the standard deviation based on the sampling rate value;

[0031] The real-time statistical confidence level is determined by the ratio of the standard deviation to the maximum value of the sampling rate.

[0032] Secondly, this application provides a network speed testing device, comprising:

[0033] The prediction module is used to predict the bandwidth of the network to be tested based on multiple probe data packets of preset data size and a congestion control model of bottleneck bandwidth and round-trip propagation time, so as to obtain the bandwidth prediction result of the network to be tested.

[0034] The configuration module is used to configure the sampling parameters of the network to be tested based on the bandwidth prediction results; wherein, the sampling parameters include a confidence threshold for characterizing the termination of sampling, and the sampling parameters also include the sampling interval and / or the sampling data packet size;

[0035] The sampling module is used to sample the network to be measured based on the sampling parameters to obtain multiple sampled velocity values.

[0036] The calculation module is used to calculate the real-time statistical confidence level based on the sampling rate value;

[0037] The determination module is used to stop sampling when the real-time statistical confidence level reaches the confidence level threshold, and to determine the speed measurement result based on the sampling speed value and the preset speed measurement prediction model.

[0038] In one possible implementation, the prediction module is also used for:

[0039] The probe data packets are transmitted sequentially in ascending order of data size, and the round-trip latency and data transmission rate of each transmission are obtained.

[0040] The bandwidth prediction results are obtained by calculating based on the bottleneck bandwidth and round-trip propagation time congestion control model, round-trip delay, and data transmission rate.

[0041] In one possible implementation, the prediction module is also used for:

[0042] Based on the bottleneck bandwidth and round-trip propagation time congestion control model, round-trip delay and data transmission rate, the theoretical bandwidth limit and the effective bandwidth after delay correction are calculated.

[0043] The smaller of the theoretical bandwidth limit and the effective bandwidth is determined as the bandwidth prediction result.

[0044] In one possible implementation, the prediction module is further configured to probe data packets, which include a first-level probe packet, a second-level probe packet, and a third-level probe packet with sequentially increasing data sizes; wherein...

[0045] The first-level probe packet has a data size of 10KB, the second-level probe packet has a data size of 100KB, and the third-level probe packet has a data size of 1MB.

[0046] In one possible implementation, the configuration module is further configured to:

[0047] Based on the bandwidth prediction results, the network level of the network to be tested is determined; the network level includes a first network level, a second network level, and a third network level, where the bandwidth of the first network level is greater than that of the second network level, and the bandwidth of the second network level is greater than that of the third network level.

[0048] Configure the sampling parameters of the network to be tested according to the network level; wherein the sampling parameters of the first network level are less than the sampling parameters of the second network level, and the sampling parameters of the second network level are less than the sampling parameters of the third network level.

[0049] In one possible implementation, the determining module is further configured to:

[0050] Determine the input parameters based on the network level corresponding to the bandwidth prediction results;

[0051] Input parameters are fed into a preset speed prediction model to output a first predicted speed based on the preset speed prediction model; wherein, the preset speed prediction model is a lightweight neural network model.

[0052] Calculate the second estimated velocity based on the sampling velocity value;

[0053] The first predicted speed and the second estimated speed are weighted and fused to obtain the speed measurement result.

[0054] In one possible implementation, the determining module is also used to input parameters, including round-trip time, packet loss rate, latency jitter, and historical speed values.

[0055] In one possible implementation, the computing module is also used for:

[0056] Calculate the standard deviation based on the sampling rate value;

[0057] The real-time statistical confidence level is determined by the ratio of the standard deviation to the maximum value of the sampling rate.

[0058] Thirdly, this application provides a network speed testing device, including: a memory and a processor;

[0059] The memory stores instructions that the computer executes;

[0060] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect or various possible implementations of the first aspect as described above.

[0061] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect or various possible implementations of the first aspect.

[0062] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the first aspect or various possible implementations of the first aspect.

[0063] This application provides a network speed measurement method, apparatus, device, and storage medium. Based on a specific model and probe data packets, it predicts bandwidth, enabling a rapid understanding of approximate network performance and providing a foundation for subsequent operations, thus improving convenience. Sampling parameters are configured according to the bandwidth prediction results, making sampling more targeted, avoiding blind sampling, and improving data acquisition efficiency. Multiple sampling speed values ​​are obtained by sampling according to the sampling parameters, providing rich data for subsequent analysis. Real-time statistical confidence levels are calculated to dynamically monitor the sampling situation. Sampling stops when the confidence level reaches a threshold, avoiding invalid sampling and saving resources. The speed measurement result is then determined by combining the sampling speed values ​​and the prediction model. The entire process makes network speed measurement more convenient and efficient, thereby achieving the technical effects of improving the convenience of network speed measurement and data acquisition efficiency. Attached Figure Description

[0064] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0065] Figure 1 This application provides a schematic diagram of an application data processing system architecture.

[0066] Figure 2 A flowchart illustrating the network speed measurement method provided in this application embodiment. Figure 1 ;

[0067] Figure 3 A flowchart illustrating the network speed measurement method provided in this application embodiment. Figure 2 ;

[0068] Figure 4 A flowchart illustrating the network speed measurement method provided in this application embodiment. Figure 3 ;

[0069] Figure 5 A flowchart illustrating the network speed measurement method provided in this application embodiment. Figure 4 ;

[0070] Figure 6 A flowchart illustrating the network speed measurement method provided in this application embodiment. Figure 5 ;

[0071] Figure 7 This is a schematic diagram of the network speed measurement device provided in the embodiments of this application;

[0072] Figure 8 This is a schematic diagram of the network speed testing device provided in an embodiment of this application.

[0073] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0074] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0075] Existing technologies suffer from drawbacks such as excessive bandwidth consumption, long testing times, poor cross-network adaptability, insufficient dynamic adjustment, and a conflict between prediction accuracy and stability. These shortcomings collectively lead to technical problems in existing technologies, including poor convenience for network speed testing and low data acquisition efficiency.

[0076] To address the aforementioned issues, this application provides a network speed measurement method, apparatus, device, and storage medium. Based on a specific model and probe data packet bandwidth prediction, it can quickly understand the approximate network performance, providing a foundation for subsequent operations and improving convenience. Sampling parameters are configured according to the bandwidth prediction results, making sampling more targeted, avoiding blind sampling, and improving data acquisition efficiency. Multiple sampling speed values ​​are obtained by sampling according to the sampling parameters, providing rich data for subsequent analysis. Real-time statistical confidence levels are calculated to dynamically monitor the sampling situation. Sampling stops when the confidence level reaches a threshold, avoiding invalid sampling and saving resources. The speed measurement result is then determined by combining the sampling speed values ​​and the prediction model. The entire process makes network speed measurement more convenient and efficient, thereby achieving the technical effects of improving the convenience of network speed measurement and data acquisition efficiency.

[0077] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0078] Figure 1 This is a schematic diagram of an application data processing system architecture provided in an embodiment of this application. The application data processing system is a computer device. Figure 1 As shown, the above architecture includes at least one of a data acquisition device 101, a processing device 102, and a display device 103.

[0079] It is understood that the structures illustrated in the embodiments of this application do not constitute a specific limitation on the architecture of the application data processing system. In other feasible embodiments of this application, the above architecture may include more or fewer components than illustrated, or combine some components, or split some components, or arrange different components, which can be determined according to the actual application scenario and is not limited here. Figure 1 The components shown can be implemented in hardware, software, or a combination of both.

[0080] In the specific implementation process, the data acquisition device 101 may include an input / output interface or a communication interface, and the data acquisition device 101 can be connected to the processing device through the input / output interface or the communication interface.

[0081] The processing device 102 can predict the bandwidth of the network to be measured using a probe data packet of a preset data size and a specific model. Based on the prediction results, it configures sampling parameters including confidence thresholds, then samples according to the parameters to obtain multiple speed values, calculates the real-time statistical confidence, stops sampling when it reaches the threshold, and finally determines the speed measurement result by combining the sampled speed values ​​and the preset model.

[0082] The display device 103 can also be a touch screen or the screen of a terminal device, used to receive user commands while displaying the above-mentioned content, so as to realize interaction with the user.

[0083] It should be understood that the aforementioned processing device can be implemented by a processor reading instructions from memory and executing those instructions, or it can be implemented by a chip circuit.

[0084] Furthermore, the network architecture and business scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0085] Figure 2 A flowchart illustrating the network speed measurement method provided in this application embodiment. Figure 1 ,like Figure 2 As shown, the network speed measurement method provided in this embodiment includes:

[0086] S201. Based on multiple probe data packets of preset data size, and using a bottleneck bandwidth and round-trip time congestion control model, perform bandwidth prediction on the network to be tested to obtain the bandwidth prediction result of the network to be tested.

[0087] In this embodiment, the bottleneck bandwidth and round-trip time congestion control model refers to the bottleneck bandwidth and round-trip time algorithm model (BBR model).

[0088] In one possible implementation, the probe data packet includes a first-level probe packet, a second-level probe packet, and a third-level probe packet with sequentially increasing data sizes; wherein the data size of the first-level probe packet is 10KB, the data size of the second-level probe packet is 100KB, and the data size of the third-level probe packet is 1MB.

[0089] Based on multiple probe packets of preset data sizes, a bottleneck bandwidth and round-trip time (RTT) congestion control model is used to predict the bandwidth of the network under test. The probe packets are divided into three levels, with the data size increasing sequentially: a first-level probe packet of 10KB, a second-level probe packet of 100KB, and a third-level probe packet of 1MB. By transmitting these probe packets, preliminary information on network performance is obtained, and the bandwidth prediction result is obtained.

[0090] S202. Based on the bandwidth prediction results, configure the sampling parameters of the network to be tested.

[0091] In this embodiment, the sampling parameters include a confidence threshold used to characterize the termination of sampling, and the sampling parameters also include the sampling interval and / or the sampling data packet size.

[0092] Based on the obtained bandwidth prediction results, the sampling parameters of the network to be tested are configured. The sampling parameters include a confidence threshold for determining when to stop sampling; a sampling interval, i.e., the time interval between two samples; and the sampling data packet size, specifying the amount of data in the data packets used for each sample.

[0093] S203. Based on the sampling parameters, perform sampling processing on the velocity network to be measured to obtain multiple sampled velocity values.

[0094] Based on the configured sampling parameters, the network to be measured is sampled. During the sampling process, sampling data packets of a specified size are sent at the set sampling interval, and the network transmission speed obtained in each sampling is recorded. Finally, multiple sampling speed values ​​are obtained to reflect the actual transmission status of the network at different times.

[0095] S204. Calculate the real-time statistical confidence level based on the sampling rate value.

[0096] Using the obtained multiple sampling rate values, a real-time statistical confidence score is calculated. A specific algorithm is used to analyze and process the sampling rate values, quantifying their stability and reliability to derive the real-time statistical confidence score. This score reflects the credibility of the current sampling results and provides a basis for subsequent decisions on whether to stop sampling.

[0097] S205. When the real-time statistical confidence level reaches the confidence level threshold, sampling is stopped, and the speed measurement result is determined based on the sampling speed value and the preset speed measurement prediction model.

[0098] When the real-time statistical confidence level reaches a pre-set confidence threshold, sampling is immediately stopped to avoid unnecessary resource consumption. Subsequently, the multiple speed values ​​obtained from sampling are input into a preset speed prediction model. Through model calculation and analysis, the speed measurement result of the network under test is finally determined, accurately reflecting the actual transmission performance of the network.

[0099] This application provides a network speed measurement method that, by acquiring multi-dimensional raw data and predicting bandwidth based on a specific model and probe data packets, can quickly understand the approximate network performance, providing a foundation for subsequent operations and improving convenience. Sampling parameters are configured according to the bandwidth prediction results, making sampling more targeted, avoiding blind sampling, and improving data acquisition efficiency. Multiple sampling speed values ​​are obtained by sampling according to the sampling parameters, providing rich data for subsequent analysis. Real-time statistical confidence levels are calculated to dynamically monitor the sampling situation. Sampling stops when the confidence level reaches a threshold to avoid invalid sampling and save resources. The speed measurement result is then determined by combining the sampling speed values ​​and the prediction model. The entire process makes network speed measurement more convenient and efficient, thereby achieving the technical effect of improving the convenience of network speed measurement and the efficiency of data acquisition.

[0100] Figure 3 A flowchart illustrating the network speed measurement method provided in this application embodiment. Figure 2 ,like Figure 3 As shown, this embodiment, based on the above embodiments, provides a detailed explanation of the specific process for obtaining the bandwidth prediction result of the network to be tested, including:

[0101] S301. Transmit probe data packets sequentially in ascending order of data size, and obtain the round-trip delay and data transmission rate for each transmission.

[0102] First, transmit probe data packets with progressively increasing data sizes in ascending order, for example, first transmit 10KB, then 100KB, and finally 1MB. For each transmission, record the round-trip time from the sender to the receiver and back, and simultaneously calculate the data transmission rate for that transmission, i.e., the amount of data successfully transmitted per unit time, thereby obtaining basic network transmission performance data.

[0103] S302. Based on the bottleneck bandwidth and round-trip propagation time congestion control model, round-trip delay and data transmission rate, calculate the theoretical bandwidth limit and the effective bandwidth after delay correction.

[0104] Based on the bottleneck bandwidth and round-trip time congestion control model, and combined with the round-trip delay and data transmission rate obtained from S301, calculations are performed. First, the theoretical bandwidth upper limit based on the real-time transmission rate is calculated. At the same time, the impact of network latency on bandwidth is considered, and the latency is corrected to obtain the effective bandwidth. The two results are combined to prepare for subsequent bandwidth determination.

[0105] S303. The smaller of the theoretical bandwidth limit and the effective bandwidth is determined as the bandwidth prediction result.

[0106] The calculated theoretical bandwidth limit is compared with the effective bandwidth after delay correction, and the smaller value is selected as the final bandwidth prediction result. This ensures that the prediction result is closer to the actual network carrying capacity.

[0107] The network speed measurement method provided in this application transmits probe packets sequentially and obtains round-trip time and data transmission rate, providing basic data for subsequent calculations and ensuring data accuracy and comprehensiveness. Based on the model and the acquired data, the theoretical bandwidth upper limit and effective bandwidth are calculated, enabling a more scientific assessment of network bandwidth. Selecting the smaller of the two values ​​as the bandwidth prediction result makes the prediction result more consistent with the actual network carrying capacity, avoiding overestimation and effectively improving the accuracy and reliability of bandwidth prediction.

[0108] Figure 4 A flowchart illustrating the network speed measurement method provided in this application embodiment. Figure 3 ,like Figure 4As shown, this embodiment, based on the above embodiments, provides a detailed explanation of the configuration process for the sampling parameters of the speed measurement network, including:

[0109] S401. Determine the network level of the network to be tested based on the bandwidth prediction results.

[0110] In this embodiment, the network levels include a first network level, a second network level, and a third network level. The bandwidth of the first network level is greater than that of the second network level, and the bandwidth of the second network level is greater than that of the third network level.

[0111] Based on the bandwidth prediction results, the networks to be tested are divided into different levels:

[0112] For example, the first network level is set with a bandwidth greater than 100Mbps, which is suitable for high-speed network scenarios; the second network level has a bandwidth between 20 and 100Mbps, which is for medium-speed networks; and the third network level has a bandwidth less than 20Mbps, which is used for low-speed networks.

[0113] By clearly defining the bandwidth range and accurately defining the network level, a basis is provided for subsequent parameter configuration.

[0114] S402. Configure the sampling parameters of the network to be tested according to the network level.

[0115] In this embodiment, the sampling parameters of the first network level are less than the sampling parameters of the second network level, and the sampling parameters of the second network level are less than the sampling parameters of the third network level.

[0116] Configure sampling parameters according to the defined network levels:

[0117] For example, the first network level uses a 100ms sampling interval, 100KB data packets, and a 2000ms timeout to quickly converge results due to its high network speed; the second network level uses a 200ms sampling interval, 200KB data packets, and a 3000ms timeout to balance speed and accuracy; and the third network level uses a 500ms sampling interval, 500KB data packets, and a 5000ms timeout to improve data stability.

[0118] Optionally, based on the above, a quantum sampling algorithm can be further introduced to optimize the resolution and noise resistance of bandwidth prediction:

[0119] Traditional sampling methods are susceptible to noise interference in low-bandwidth scenarios (such as Level 3), resulting in insufficient frequency resolution (e.g., inability to distinguish between minute fluctuations of 10 Mbps and 15 Mbps). To address this, a quantum sampling algorithm is employed to improve signal discriminability through entangled state coding, specifically as follows:

[0120] Define the entanglement capability index F of a quantum sampling circuit:

[0121]

[0122] Where F is the entanglement capability index. Where N is the sampling frequency and N is the number of qubits.

[0123] By increasing N (e.g., from 4 bits to 8 bits), frequency resolution can be doubled compared to conventional methods, for example, achieving precise differentiation at the 5 Mbps level in the third class.

[0124] By leveraging the superposition property of quantum states, redundant encoding can be performed on the sampled signal. For example, a single sampling point can be mapped to multiple quantum states, and high-frequency noise can be filtered out by measuring the superposition probability distribution of the states, thereby improving the signal-to-noise ratio (SNR) in low-bandwidth scenarios.

[0125] The network speed testing method provided in this application classifies network levels based on bandwidth prediction results, accurately distinguishing networks with different performance levels. This provides direction for subsequent parameter configuration and ensures targeted configuration. By configuring differentiated sampling parameters according to network level, high-speed networks can be tested quickly, and low-speed networks can be tested accurately, effectively reducing traffic waste and improving testing efficiency and data quality.

[0126] Figure 5 A flowchart illustrating the network speed measurement method provided in this application embodiment. Figure 4 ,like Figure 5 As shown, this embodiment, based on the above embodiments, provides a detailed explanation of the process for determining the speed measurement result, including:

[0127] S501. Determine the input parameters based on the network level corresponding to the bandwidth prediction results.

[0128] In this embodiment, the input parameters include round-trip time, packet loss rate, latency jitter, and historical speed value.

[0129] Based on the network tiers defined by bandwidth prediction results, input parameters are determined. For different network tiers, round-trip time (time to and from data packet), packet loss rate (proportion of data packets lost during transmission), latency jitter (variance of latency of consecutive data packets), and historical speed values ​​(weighted average of the three most recent speed measurements) are collected. These parameters comprehensively reflect the current state and past performance of the network, providing data support for subsequent predictions.

[0130] S502. Input the input parameters into the preset speed prediction model, and output the first predicted speed according to the preset speed prediction model.

[0131] In this embodiment, the preset speed measurement prediction model is a lightweight neural network model.

[0132] The collected input parameters are input into a preset lightweight neural network model. This model adopts a "4-input-1-output" structure, taking RTT, packet loss rate, jitter, and historical speed as inputs. After calculation by the hidden layer, the first predicted speed is output. This model is less than 10KB in size, which is suitable for the lightweight requirements of mobile devices and can quickly provide prediction results.

[0133] S503. Calculate the second estimated velocity based on the sampling velocity value.

[0134] Based on multiple sampling rate values ​​obtained during the sampling process, their average or weighted average is calculated using a specific algorithm, and this is used as a second estimated rate, which reflects the actual transmission situation of the network during the sampling phase.

[0135] S504. Perform weighted fusion processing on the first predicted speed and the second estimated speed to obtain the speed measurement result.

[0136] The first predicted speed and the second estimated speed are weighted and fused with a weight ratio of 7:3. The neural network prediction results are combined with the actual sampling results. Taking into account the predictive ability of the model and the actual sampling situation, the final speed measurement result is obtained.

[0137] Optionally, a multimodal training dataset can be pre-constructed for training the speed prediction model in the speed measurement scenario: First, the node connection relationship is extracted from the network topology graph using optical character recognition (OCR) technology, and an initial dataset is generated in combination with the corresponding configuration file; then, an adaptive domain evaluation function is used to calculate the comprehensive classification score, which integrates the prediction probability (weight 0.7) and matching score (weight 0.3) of the text classification model. For example, for text related to "5G base station uplink bandwidth", if the classification probability is 0.92 and the matching score is 0.85, the comprehensive score reaches 0.899; finally, the target training set is selected from 100,000-level raw data through Monte Carlo sampling to ensure that the data ratio of different network levels (such as fiber / 4G / Wi-Fi) conforms to the actual scenario distribution (6:3:1).

[0138] The network speed measurement method provided in this application provides a comprehensive and accurate data foundation for speed measurement prediction by determining input parameters that reflect network status and performance. A lightweight neural network model is used to quickly output a first predicted speed, adapting to the needs of mobile devices. A second estimated speed is calculated based on the sampled speed values, reflecting the actual transmission situation. The two are weighted and fused, combining model prediction and actual sampling, making the speed measurement results more accurate and reliable, effectively improving the quality and stability of speed measurement.

[0139] Figure 6 A flowchart illustrating the network speed measurement method provided in this application embodiment. Figure 5 ,like Figure 6As shown, this embodiment, based on the above embodiments, provides a detailed explanation of the calculation process for real-time statistical confidence, including:

[0140] S601. Calculate the standard deviation based on the sampling rate value.

[0141] During the sampling process, multiple sampling rate values ​​are obtained. These sampling rate values ​​are treated as a set of data. Using the standard deviation calculation formula, the average value of this set of data is first calculated. Then, the sum of squares of the differences between each sampling rate value and the average value is calculated. Next, the sum of squares is divided by the number of sampling rate values ​​(if it is the sample standard deviation, it is divided by the number of values ​​minus one). Finally, the square root of the result is taken to obtain the standard deviation, which reflects the dispersion of the sampling rate values.

[0142] S602. Determine the real-time statistical confidence level based on the ratio of the standard deviation to the maximum value of the sampling rate.

[0143] First, find the maximum value among the sampling rate values. Then, divide the previously calculated standard deviation by this maximum value. The result is the real-time statistical confidence score. The smaller this ratio, the more concentrated the sampling rate values ​​are, the higher the data reliability, and the higher the real-time statistical confidence score; conversely, the larger the ratio, the lower the confidence score.

[0144] Optionally, trimodal noise injection can be used during the sampling data augmentation stage:

[0145] Gaussian noise was used to simulate electromagnetic interference, with the standard deviation set to 5% of the mean of the sampled values.

[0146] Salt and pepper noise simulates packet loss during transmission, and the noise density is dynamically adjusted according to the network level (Level 1: 0.01% / Level 2: 0.05% / Level 3: 0.2%).

[0147] Random occlusion simulates equipment failure, and the occlusion area percentage is set according to the historical failure rate (0.5% for fiber optic links / 2% for 4G base stations / 8% for Wi-Fi hotspots).

[0148] For example, when testing 5G networks, Gaussian noise (σ=0.03) and random occlusion (area 15%) are injected simultaneously to improve the speed measurement stability of the model in complex environments.

[0149] The network speed measurement method provided in this application clearly reflects the dispersion of the sampled data by calculating the standard deviation of the sampled speed values, providing a key indicator for evaluating data reliability. The real-time statistical confidence level is determined by using the ratio of the standard deviation to the maximum value of the sampled speed values, allowing for a direct assessment of data credibility and avoiding inaccurate speed measurement results due to large data fluctuations, thus improving the stability and reliability of the speed measurement results.

[0150] Figure 7This is a schematic diagram of the network speed testing device provided in an embodiment of this application. The device in this embodiment can be in the form of software or hardware. Figure 7 As shown in the embodiment of this application, the network speed testing device 700 includes: a prediction module 701, a configuration module 702, a sampling module 703, a calculation module 704, and a determination module 705.

[0151] The prediction module 701 is used to predict the bandwidth of the network to be tested based on multiple probe data packets of preset data size and a bottleneck bandwidth and round-trip propagation time congestion control model, so as to obtain the bandwidth prediction result of the network to be tested.

[0152] The configuration module 702 is used to configure the sampling parameters of the network to be tested based on the bandwidth prediction results; wherein, the sampling parameters include a confidence threshold for characterizing the termination of sampling, and the sampling parameters also include the sampling interval and / or the sampling data packet size;

[0153] The sampling module 703 is used to sample the network to be measured based on the sampling parameters to obtain multiple sampled velocity values.

[0154] The calculation module 704 is used to calculate the real-time statistical confidence level based on the sampling rate value;

[0155] The determination module 705 is used to stop sampling when the real-time statistical confidence level reaches the confidence level threshold, and to determine the speed measurement result based on the sampling speed value and the preset speed measurement prediction model.

[0156] In one possible implementation, the prediction module 701 is further configured to:

[0157] The probe data packets are transmitted sequentially in ascending order of data size, and the round-trip latency and data transmission rate of each transmission are obtained.

[0158] The bandwidth prediction results are obtained by calculating based on the bottleneck bandwidth and round-trip propagation time congestion control model, round-trip delay, and data transmission rate.

[0159] In one possible implementation, the prediction module 701 is further configured to:

[0160] Based on the bottleneck bandwidth and round-trip propagation time congestion control model, round-trip delay and data transmission rate, the theoretical bandwidth limit and the effective bandwidth after delay correction are calculated.

[0161] The smaller of the theoretical bandwidth limit and the effective bandwidth is determined as the bandwidth prediction result.

[0162] In one possible implementation, the prediction module 701 is further configured to probe data packets, which include a first-level probe packet, a second-level probe packet, and a third-level probe packet with sequentially increasing data sizes; wherein...

[0163] The first-level probe packet has a data size of 10KB, the second-level probe packet has a data size of 100KB, and the third-level probe packet has a data size of 1MB.

[0164] In one possible implementation, the configuration module 702 is further configured to:

[0165] Based on the bandwidth prediction results, the network level of the network to be tested is determined; the network level includes a first network level, a second network level, and a third network level, where the bandwidth of the first network level is greater than that of the second network level, and the bandwidth of the second network level is greater than that of the third network level.

[0166] Configure the sampling parameters of the network to be tested according to the network level; wherein the sampling parameters of the first network level are less than the sampling parameters of the second network level, and the sampling parameters of the second network level are less than the sampling parameters of the third network level.

[0167] In one possible implementation, the determining module 705 is further configured to:

[0168] Determine the input parameters based on the network level corresponding to the bandwidth prediction results;

[0169] Input parameters are fed into a preset speed prediction model to output a first predicted speed based on the preset speed prediction model; wherein, the preset speed prediction model is a lightweight neural network model.

[0170] Calculate the second estimated velocity based on the sampling velocity value;

[0171] The first predicted speed and the second estimated speed are weighted and fused to obtain the speed measurement result.

[0172] In one possible implementation, the determining module 705 is also used to input parameters, including round-trip time, packet loss rate, latency jitter, and historical speed values.

[0173] In one possible implementation, the computing module 704 is further configured to:

[0174] Calculate the standard deviation based on the sampling rate value;

[0175] The real-time statistical confidence level is determined by the ratio of the standard deviation to the maximum value of the sampling rate.

[0176] The network speed testing device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0177] Figure 8 This is a schematic diagram of the network speed testing device provided in an embodiment of this application. Figure 8 As shown, the network speed testing device 800 provided in this embodiment includes at least one processor 801 and a memory 802. Optionally, the device 800 also includes a communication component 803. The processor 801, memory 802, and communication component 803 are connected via a bus.

[0178] In a specific implementation, at least one processor 801 executes computer execution instructions stored in memory 802, causing at least one processor 801 to perform the above-described method.

[0179] The specific implementation process of processor 801 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0180] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0181] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0182] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0183] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0184] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0185] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0186] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0187] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0188] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0189] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0190] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0191] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0192] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A network speed measurement method, characterized in that, include: Based on multiple probe data packets of preset data size, and using a bottleneck bandwidth and round-trip time congestion control model, bandwidth prediction is performed on the network to be tested to obtain the bandwidth prediction result of the network to be tested. Based on the bandwidth prediction results, the sampling parameters of the network to be tested are configured; wherein, the sampling parameters include a confidence threshold for characterizing the termination of sampling, and the sampling parameters also include a sampling interval and / or the size of the sampling data packet; Based on the sampling parameters, the velocity network to be measured is sampled to obtain multiple sampled velocity values; Calculate the real-time statistical confidence level based on the sampling rate value; Sampling stops when the real-time statistical confidence level reaches the confidence level threshold, and the speed measurement result is determined based on the sampling speed value and the preset speed measurement prediction model.

2. The method according to claim 1, characterized in that, The process of predicting the bandwidth of the network to be tested based on multiple probe data packets of preset data sizes and a bottleneck bandwidth and round-trip time congestion control model, to obtain the bandwidth prediction result of the network to be tested, includes: The probe data packets are transmitted sequentially in ascending order of data size, and the round-trip time and data transmission rate for each transmission are obtained. The bandwidth prediction result is obtained by performing calculations based on the bottleneck bandwidth and round-trip time congestion control model, the round-trip delay, and the data transmission rate.

3. The method according to claim 2, characterized in that, The calculation process based on the bottleneck bandwidth and round-trip time congestion control model, the round-trip delay, and the data transmission rate to obtain the bandwidth prediction result includes: Based on the bottleneck bandwidth and round-trip propagation time congestion control model, the round-trip delay, and the data transmission rate, the theoretical bandwidth limit and the effective bandwidth after delay correction are calculated. The smaller of the theoretical bandwidth limit and the effective bandwidth is determined as the bandwidth prediction result.

4. The method according to claim 3, characterized in that, The probe data packet includes a first-level probe packet, a second-level probe packet, and a third-level probe packet with sequentially increasing data sizes; wherein... The first-level probe packet has a data size of 10KB, the second-level probe packet has a data size of 100KB, and the third-level probe packet has a data size of 1MB.

5. The method according to any one of claims 1 to 4, characterized in that, The step of configuring the sampling parameters of the network to be tested based on the bandwidth prediction result includes: Based on the bandwidth prediction results, the network level of the network to be tested is determined; wherein, the network level includes a first network level, a second network level, and a third network level, the bandwidth of the first network level is greater than the bandwidth of the second network level, and the bandwidth of the second network level is greater than the bandwidth of the third network level; Based on the network level, the sampling parameters of the network to be tested are configured; wherein the sampling parameters of the first network level are less than the sampling parameters of the second network level, and the sampling parameters of the second network level are less than the sampling parameters of the third network level.

6. The method according to claim 5, characterized in that, The step of determining the speed measurement result based on the sampling speed value and the preset speed measurement prediction model includes: The input parameters are determined based on the network level corresponding to the bandwidth prediction results; The input parameters are input into the preset speed prediction model to output a first predicted speed according to the preset speed prediction model; wherein, the preset speed prediction model is a lightweight neural network model. Calculate the second estimated velocity based on the sampling velocity value; The first predicted speed and the second estimated speed are weighted and fused to obtain the speed measurement result.

7. The method according to claim 6, characterized in that, The input parameters include round-trip time, packet loss rate, latency jitter, and historical speed values.

8. The method according to any one of claims 1 to 4, characterized in that, The step of calculating the real-time statistical confidence level based on the sampling rate value includes: Calculate the standard deviation based on the sampling rate value; The real-time statistical confidence level is determined based on the ratio of the standard deviation to the maximum value of the sampling rate.

9. A network speed measurement device, characterized in that, include: The prediction module is used to predict the bandwidth of the network to be tested based on multiple probe data packets of preset data size and a bottleneck bandwidth and round-trip time congestion control model, so as to obtain the bandwidth prediction result of the network to be tested. The configuration module is used to configure the sampling parameters of the network to be tested based on the bandwidth prediction results; wherein, the sampling parameters include a confidence threshold for characterizing the termination of sampling, and the sampling parameters also include a sampling interval and / or the size of the sampling data packet; The sampling module is used to perform sampling processing on the network to be tested according to the sampling parameters to obtain multiple sampled velocity values; The calculation module is used to calculate the real-time statistical confidence level based on the sampling rate value; The determination module is used to stop sampling when the real-time statistical confidence level reaches the confidence level threshold, and to determine the speed measurement result based on the sampling speed value and the preset speed measurement prediction model.

10. A network speed testing device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-8.

12. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method described in any one of claims 1-8.

Images