A dynamic load testing method and system for a short message communication system

Abstract

The application discloses a dynamic load test method and system of a short message communication system, and relates to data processing technology, which comprises the following steps: dividing a plurality of priority categories in advance for data messages; obtaining data messages under a plurality of historical emergency scenarios and classifying them according to the priority categories; clock aligning the classified data messages of each secondary node and selecting data message samples of a typical time period; extracting message features of each data message of any typical time period, assigning weights to the corresponding message features according to the priority of each data message, and constructing a dynamic load generator by using LSTM; extracting message features under a current emergency scenario and inputting them into the dynamic load generator; and comparing the output of the dynamic load generator with the load condition of the current communication system to complete dynamic load testing. The method of the application is used for dynamically allocating channel resources according to the result of dynamic load testing and exposing potential resource bottlenecks of the system in advance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a dynamic load testing method and system of a short message communication system. BACKGROUND

[0002] With the wide application of Beidou satellite navigation system, Beidou short message communication technology has become a key communication means in the fields of water conservancy, power, emergency rescue, etc., especially in remote areas or extreme disaster scenarios, its importance as an emergency bottom communication method is increasingly prominent. However, the existing Beidou short message communication system testing method has some shortcomings in dynamic load simulation, channel resource optimization and test scene authenticity, mainly as follows:

[0003] In the mixed networking scene (such as Beidou No. 3 and 5G / LoRa cooperation), low-priority messages may occupy Beidou channel bandwidth, causing high-priority emergency data (such as disaster alarm) transmission delay.

[0004] The current test parameters (such as load intensity, disturbance type) are highly dependent on manual experience, and lack data-driven dynamic adjustment capability. For example, the load curve set by manual experience is difficult to cover complex scenarios (such as "earthquake + communication interruption" superimposed disasters) that have not been seen in history, resulting in insufficient test coverage and potential system bottlenecks that cannot be exposed. SUMMARY

[0005] The embodiments of the present application provide a dynamic load testing method and system of a short message communication system to solve or partially solve the above technical problems, to dynamically allocate channel resources according to the results of dynamic load testing, and to expose potential system resource bottlenecks in advance.

[0006] The embodiments of the present application provide a dynamic load testing method of a short message communication system, applied to an emergency communication scene based on Beidou short message communication, the emergency communication scene includes a plurality of secondary nodes and a master node, the secondary nodes send data messages to the master node based on Beidou short message, and the dynamic load testing method includes:

[0007] Pre-divide the data messages into a plurality of priority categories, wherein the higher the priority of the data message, the more suitable it is for short message communication transmission, and the lower priority message uses hybrid channel transmission;

[0008] Obtain data messages under a plurality of historical emergency scenarios, and classify them according to priority categories;

[0009] Align the classified data messages of each secondary node with the clock, and select data message samples of a typical time period, wherein the typical time period is a time period in which the total data message volume exceeds a preset threshold;

[0010] Extracting the message features of each data message in any typical period, assigning weights to the message features according to the priority of each data message, and constructing a dynamic load generator using LSTM;

[0011] Extracting message features under the current emergency scenario and inputting the dynamic load generator;

[0012] Comparing the output of the dynamic load generator with the load condition of the current communication system to complete the dynamic load test.

[0013] Optionally, the classified data messages of each secondary node are clock-aligned, and the data message samples of the typical period are selected, including:

[0014] Aligning the classified data messages of each secondary node according to the timestamp in the unified time coordinate;

[0015] According to the unified time window, the data messages are intercepted in the time coordinate;

[0016] The period in which the total data message quantity in the intercepted data segment exceeds the preset threshold, and the adjacent previous specified number of periods, are taken as the data message samples of the typical period.

[0017] Optionally, it further includes adding perturbations to part of the data message samples of the typical period in at least one of the following ways:

[0018] Obtaining a plurality of high-priority messages in the data message samples of each typical period, and randomly inserting high-priority messages in the typical period, wherein the frequency of insertion obeys a Poisson distribution satisfying a preset burst strength;

[0019] Adding a random time offset of ±Δt to the message time of high-priority messages in the data message samples of any typical period and / or in the time window before and after the data message samples;

[0020] Adding Gaussian noise to the data field in the data message, and randomly discarding part of the non-critical fields.

[0021] Optionally, it further includes adding spatio-temporal perturbations to the data message samples in the following ways:

[0022] Obtaining the geographical distribution data of historical disaster events in the communication coverage area;

[0023] Determining the intersection area of the target monitoring range in the geographical distribution data;

[0024] Extracting the geographical feature information of the intersection area, and determining the similarity between the extracted geographical feature information and the local geographical features of the target monitoring range;

[0025] Correlate the data message related to the insertion logic based on the local region corresponding to the local geographical feature with a similarity higher than a preset threshold.

[0026] Optionally, in the case where the target monitoring range includes a river channel, the extracted geographical feature information and the local geographical feature include topographical features, hydrological features, and meteorological features, wherein:

[0027] The topographical features are obtained based on DEM data.

[0028] The hydrological features include river channel hydrological data, historical flood frequency, and soil permeability coefficient.

[0029] The meteorological features include extreme rainfall and extreme rainfall event frequency.

[0030] A random forest model is used to dynamically assign weights to the features based on the relevance of historical emergency events and geographical features.

[0031] Optionally, determining the similarity between the extracted geographical feature information and the local geographical feature of the target monitoring range further includes introducing a time dimension and superimposing a time decay factor when calculating the geographical feature similarity, satisfying:

[0032]

[0033] wherein, represents the weight of the i-th geographical feature, i represents the similarity of the i-th geographical feature normalized by cosine similarity or Euclidean distance, i represents the timestamp of the current data message, represents the time of the historical emergency event, is a time decay coefficient, is used to make recent emergency events have a greater impact.

[0034] Optionally, constructing a dynamic load generator using LSTM includes:

[0035] Obtain historical load data corresponding to any typical period and extract load features of the historical load data, wherein the load data includes channel packet loss rate, end-to-end delay, CPU and memory occupancy.

[0036] Input the message features with weights of any typical period and the corresponding load features as a group into the LSTM to generate a predicted load curve using the LSTM to construct a dynamic load generator.

[0037] Optionally, it further includes configuring a comprehensive optimization objective function for load testing:

[0038] ​​

[0039] in, , , For adjustable hyperparameters, This represents the load coverage rate for a typical time period t. The average success rate of high-priority messages in a typical time period t. This indicates the channel overload risk during a typical time period t;

[0040] Solve for the dynamic optimization parameter combination in the comprehensive optimization objective function to balance test efficiency and system stability in real time.

[0041] This application also proposes a dynamic load testing system for a short message communication system, including a processor and a memory. The memory stores a computer program, which, when executed by the processor, implements the steps of the aforementioned dynamic load testing method for a short message communication system.

[0042] The method described in this application can dynamically allocate channel resources based on the results of dynamic load testing and expose potential resource bottlenecks in the system in advance.

[0043] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0044] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0045] Figure 1 This is a basic flowchart illustrating the dynamic load testing method for the short message communication system in this embodiment. Detailed Implementation

[0046] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0047] This application proposes a dynamic load testing method for a short message communication system, applied to emergency communication scenarios based on BeiDou short message communication. In specific examples, the emergency communication scenario may be, for example, a cascade power station scenario or a wide-area power monitoring scenario. The emergency communication scenario includes multiple secondary nodes and one primary node. In a specific example, the primary node may correspond to the centralized control center in the emergency communication scenario, and the secondary nodes may correspond to various monitoring stations. The secondary nodes send data packets to the primary node based on BeiDou short message communication, such as... Figure 1 As shown, the dynamic load testing method includes:

[0048] In step S101, data packets are pre-classified into multiple priority categories. Higher priority data packets are adapted for short message communication transmission, while lower priority packets are transmitted using a hybrid channel. For example, in high-load scenarios, BeiDou-3 channel resources are preferentially allocated to high-priority packets, while low-priority data is transmitted using a hybrid channel (such as 5G / LoRa).

[0049] In step S102, data packets from multiple historical emergency scenarios are acquired and classified according to priority categories. That is, the data packets from historical emergency scenarios are classified according to the aforementioned multiple priority categories.

[0050] In step S103, the data packets classified by each secondary node are clock-aligned, and a sample of data packets from a typical time period is selected. This typical time period is defined as the period during which the total number of data packets exceeds a preset threshold. The dynamic load testing method of this application embodiment is used to test channel resource congestion under extreme conditions. By clock-aligning data packets from historical emergency scenarios, the peak data packet volume that may exist at the same time can be determined.

[0051] In step S104, the message features of each data packet in any typical time period are extracted, weights are assigned to the corresponding message features according to the priority of each data packet, and a dynamic load generator is constructed using LSTM. In a specific example, the dynamic load generator can be constructed by combining parameters such as channel packet loss rate and end-to-end delay.

[0052] In step S105, the message features under the current emergency scenario are extracted and input into the dynamic load generator.

[0053] In step S106, the output of the dynamic load generator is compared with the current load of the communication system to complete the dynamic load test. After the dynamic load test is completed, channel resources can be dynamically allocated according to the test results. Furthermore, the method based on this embodiment can expose potential resource bottlenecks in the system in advance based on the dynamic load generator, forming a "test-analysis-optimization" closed loop and improving the adaptability of the test system.

[0054] In some embodiments, the data packets classified by each secondary node are clock-aligned, and a sample of data packets from a typical time period is selected, including:

[0055] The data packets, categorized by each secondary node, are aligned on a unified time coordinate system based on their timestamps. In a specific example, alignment can be performed based on the data packet's sending timestamp and normal latency.

[0056] The data is extracted according to a unified time window and the time coordinate. For example, a time window of a specified duration can be set, and the classified data packets can be extracted under the aligned unified time coordinate. The extracted data includes multiple data packets within that time window.

[0057] The data segments with a total number of data packets exceeding a preset threshold, along with a specified number of adjacent segments, are used as data packet samples for typical time periods. Based on the long short-term memory effect of LSTM, further using a specified number of adjacent segments as typical time periods can improve the learning and prediction performance of LSTM.

[0058] In some embodiments, the perturbation of data packet samples for a portion of a typical time period is further included by employing at least one of the following methods:

[0059] Obtain multiple high-priority packets from data packet samples of each typical time period, and randomly insert high-priority packets within the typical time period, wherein the insertion frequency follows a preset burst intensity. The Poisson distribution.

[0060] For any typical time period data packet sample, take itself, and / or add a random time offset of ±Δt to the message time of high priority messages in the time windows before and after it;

[0061] Gaussian noise is added to the data fields in the data packets, and some non-critical fields are randomly discarded. Under normal circumstances, since the overall sample size of emergency data packets is small, the above methods can both expand the sample size of emergency data packets and improve the randomness and suddenness of the emergency data packet samples, thereby ensuring that the system has a certain degree of redundancy based on the results of dynamic load testing.

[0062] Based on the foregoing embodiments, since the overall number of emergency data message samples is relatively small, while emergency communication scenarios in actual use may cover a large area, some embodiments further include adding spatiotemporal perturbations to the data message samples in the following manner to further supplement the emergency data message samples:

[0063] Obtain the geographical distribution data of historical disaster events within the communication coverage area.

[0064] The intersection area between the geographical distribution data and the target monitoring range is determined, that is, the historical disaster events that occurred within the target monitoring range are identified. The corresponding historical disaster events may generate emergency data messages.

[0065] Extract the geographic feature information of the intersection area and determine the similarity between the extracted geographic feature information and the local geographic features of the target monitoring range.

[0066] Based on local geographic features with similarity exceeding a preset threshold, logically related data packets are inserted. That is, based on historical disaster events, areas with high similarity are likely to experience similar disaster events, generating data packets. This probability is represented by inserting logically related data packets. In a specific example, a certain probability of event occurrence can be set, and data packets can be inserted according to this probability.

[0067] In some embodiments, where the target monitoring range includes a river, the extracted geographic feature information and local geographic features include: topographic features, hydrological features, and meteorological features, wherein:

[0068] The terrain features are obtained based on DEM data, including terrain slope, elevation standard deviation, and surface roughness.

[0069] The hydrological characteristics include river hydrological data, historical flood frequency, and soil permeability coefficient.

[0070] The meteorological characteristics include extreme rainfall and the frequency of extreme rainfall events.

[0071] A random forest model is used to dynamically assign weights to features based on the correlation between historical emergency events and geographical characteristics. For example, in areas prone to debris flows, the weight of terrain slope is increased to 60%; in areas prone to floods, the weight of hydrological features exceeds 70%.

[0072] In some embodiments, determining the similarity between the extracted geographic feature information and the local geographic features of the target monitoring range further includes introducing a time dimension, superimposing a time decay factor when calculating geographic feature similarity, satisfying:

[0073]

[0074] in, Indicates the first i Geographic feature weights The first digit represents the cosine similarity or Euclidean distance normalized. i Similarity of geographical features Indicates the timestamp of the current data packet. Indicates the time of historical emergency events. The time decay coefficient, This is used to amplify the impact of recent emergencies.

[0075] This design effectively improves the accuracy of similarity calculation, avoids relying solely on coarse-grained matching based on spatial overlap, and enables the model to adapt to different disaster types through dynamic weight allocation.

[0076] In some embodiments, building a dynamic load generator using LSTM includes:

[0077] Obtain historical load data corresponding to any typical time period, and extract the load characteristics of the historical load data, including channel packet loss rate, end-to-end latency, CPU and memory utilization.

[0078] The message features and corresponding load features assigned weights for any typical time period are grouped together and input into an LSTM to generate a predicted load curve, thus constructing a dynamic load generator. In other words, in some examples, the message features and load features grouped together can be set to a certain time interval, and the LSTM will generate the predicted load curve after the corresponding time interval.

[0079] In some embodiments, a comprehensive optimization objective function for configuring load testing is also included:

[0080]

[0081] in, , , For adjustable hyperparameters, This represents the load coverage rate for a typical time period t. The average success rate of high-priority messages in a typical time period t. This indicates the channel overload risk during a typical time period t;

[0082] The dynamic optimization parameter combination in the comprehensive optimization objective function is solved to balance testing efficiency and system stability in real time. This approach achieves an integrated "generate-test-optimize" closed-loop process, reducing manual intervention, and multi-objective optimization ensures that both test coverage and system stability are met.

[0083] The method in this application embodiment achieves the characteristics of suddenness in the test scenario and system stability through feature extraction and adaptive weight allocation, multi-objective optimization and geographic feature coupling perturbation injection, thereby improving test efficiency and providing a comprehensive solution for the reliability assessment of the Beidou short message communication system.

[0084] This application also proposes a dynamic load testing system for a short message communication system, including a processor and a memory. The memory stores a computer program, which, when executed by the processor, implements the steps of the aforementioned dynamic load testing method for a short message communication system.

[0085] Furthermore, although exemplary embodiments have been described herein, their scope includes any and all embodiments based on this disclosure that have equivalent elements, modifications, omissions, combinations (e.g., schemes involving intersections of various embodiments), adaptations, or changes. They are not limited to the examples described in this specification or during the implementation of this application, and such examples are to be construed as non-exclusive.

[0086] The above description is intended to be illustrative and not restrictive. For example, the above examples (or one or more of them) can be used in combination with each other. Other embodiments can be used by those skilled in the art when reading the above description.

[0087] The above embodiments are merely exemplary embodiments of this disclosure. Those skilled in the art can make various modifications or equivalent substitutions to this invention within the scope of the disclosure, and such modifications or equivalent substitutions should also be considered to fall within the protection scope of this invention.

Claims

1. A dynamic load testing method for a short message communication system, characterized in that, The method is applied to emergency communication scenarios based on BeiDou short message service. These scenarios include multiple secondary nodes and one primary node. The secondary nodes send data packets to the primary node using BeiDou short message service. The dynamic load testing method includes: Data packets are pre-classified into multiple priority categories. Data packets with a priority higher than the specified priority are adapted for short message communication transmission, while data packets with a priority lower than the specified priority are transmitted through a mixed channel. Acquire data packets from multiple historical emergency scenarios and categorize them according to priority. The data packets after being classified by each secondary node are clock-aligned, and a sample of data packets from a typical time period is selected, wherein the typical time period is the period in which the total number of data packets exceeds a preset threshold. Extract the message features of each data packet in any typical time period, assign weights to the corresponding message features according to the priority of each data packet, and use LSTM to build a dynamic load generator. Extract message features from the current emergency scenario and input them into the dynamic load generator; The dynamic load test is completed by comparing the output of the dynamic load generator with the current load of the communication system.

2. The dynamic load testing method for a short message communication system as described in claim 1, characterized in that, After classifying the data packets from each secondary node, clock alignment is performed, and data packet samples from typical time periods are selected, including: The data packets from each secondary node are categorized and aligned on a unified time coordinate according to their timestamps. Extract the data according to the unified time window and the time coordinates; The data packets in the segment whose total number of data packets exceeds a preset threshold, and the data packets in the preceding specified number of adjacent segments, are taken as data packet samples for typical segments.

3. The dynamic load testing method for a short message communication system as described in claim 1, characterized in that, This also includes adding perturbations to data packet samples from typical time periods using at least one of the following methods: Multiple high-priority packets are obtained from data packet samples in each typical time period, and high-priority packets are randomly inserted into the typical time period, wherein the insertion frequency follows a Poisson distribution that satisfies the preset burst intensity. For any typical time period data packet sample, take itself, and / or add a random time offset of ±Δt to the message time of high priority messages in the time windows before and after it; Add Gaussian noise to the data fields in the data message, and randomly discard some non-critical fields.

4. The dynamic load testing method for a short message communication system as described in claim 3, characterized in that, This also includes adding spatiotemporal perturbations to data packet samples using the following methods: Obtain the geographical distribution data of historical disaster events within the communication coverage area; Determine the area where the geographic distribution data intersects with the target monitoring range; Extract the geographic feature information of the intersection area and determine the similarity between the extracted geographic feature information and the local geographic features of the target monitoring range; Based on local geographic features with similarity higher than a preset threshold, data packets related to the historical disaster event types or geographic features of the corresponding regions are inserted.

5. The dynamic load testing method for a short message communication system as described in claim 4, characterized in that, When the target monitoring area includes a river, the extracted geographic features and local geographic features include: topographic features, hydrological features, and meteorological features, among which: The terrain features were obtained based on DEM data; The hydrological characteristics include river hydrological data, historical flood frequency, and soil permeability coefficient; The meteorological characteristics include extreme rainfall and the frequency of extreme rainfall events; A random forest model is used to dynamically assign feature weights based on the correlation between historical emergency events and geographical features.

6. The dynamic load testing method for a short message communication system as described in claim 4, characterized in that, Determining the similarity between the extracted geographic features and the local geographic features within the target monitoring area also involves introducing a time dimension. A time decay factor is superimposed when calculating the geographic feature similarity, satisfying the following: in, Indicates the first i Geographic feature weights The first digit represents the cosine similarity or Euclidean distance normalized. i Similarity of geographical features Indicates the timestamp of the current data packet. Indicates the time of historical emergency events. The time decay coefficient, This is used to amplify the impact of recent emergencies.

7. The dynamic load testing method for a short message communication system as described in claim 1, characterized in that, Building a dynamic load generator using LSTM includes: Obtain historical load data corresponding to any typical time period, and extract the load characteristics of the historical load data, including channel packet loss rate, end-to-end latency, CPU and memory utilization. The message features assigned weights for any typical time period and their corresponding load features are grouped together and input into an LSTM to generate a predicted load curve, thereby constructing a dynamic load generator.

8. The dynamic load testing method for a short message communication system as described in claim 7, characterized in that, It also includes a comprehensive optimization objective function for configuring load testing: in, , , For adjustable hyperparameters, This represents the load coverage rate for a typical time period t. The average success rate of high-priority messages in a typical time period t. This indicates the channel overload risk during a typical time period t; Solving for the adjustable hyperparameters in the comprehensive optimization objective function , , The optimal combination is used to balance testing efficiency and system stability in real time.

9. A dynamic load testing system for a short message communication system, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, which, when executed by the processor, implements the steps of the dynamic load testing method for a short message communication system as described in any one of claims 1 to 7.

Images