WO2018018572A1 - 样本数据采集方法、装置及系统 - Google Patents

样本数据采集方法、装置及系统 Download PDF

Info

Publication number
WO2018018572A1
WO2018018572A1 PCT/CN2016/092193 CN2016092193W WO2018018572A1 WO 2018018572 A1 WO2018018572 A1 WO 2018018572A1 CN 2016092193 W CN2016092193 W CN 2016092193W WO 2018018572 A1 WO2018018572 A1 WO 2018018572A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
background information
tested
sample data
server
Prior art date
Application number
PCT/CN2016/092193
Other languages
English (en)
French (fr)
Inventor
胡永平
宋建涛
赵其勇
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2016/092193 priority Critical patent/WO2018018572A1/zh
Publication of WO2018018572A1 publication Critical patent/WO2018018572A1/zh

Links

Images

Definitions

  • the present invention relates to the field of information technology, and in particular, to a sample data collection method, device and system.
  • the DFI Deep/Dynamic Flow Inspection
  • the type identification model obtained by the pre-training is required to determine the service type of the service to be identified.
  • the specific training process includes: the terminal runs the service to be identified multiple times in the specified geographical location, and records the sample data of the service to be identified in the process of running the service to be identified. After the operation is completed, the tester brings the terminal back to the experiment from the specified geographic location. And connecting the terminal and the server through the data line, the server acquires sample data of the service to be identified recorded by the terminal, and trains the type identification model according to the obtained plurality of sample data.
  • the sample data of the service to be identified may include a service type of the service to be identified and a data packet length of the service to be identified.
  • the tester When training the type identification model, the tester is required to bring the terminal that records the sample data back to the laboratory, and connect the terminal to the server of the laboratory, so that the server obtains the sample data, causing the server to obtain the sample data for a long period and the cost is too high.
  • the server When training the type identification model, the tester is required to bring the terminal that records the sample data back to the laboratory, and connect the terminal to the server of the laboratory, so that the server obtains the sample data, causing the server to obtain the sample data for a long period and the cost is too high.
  • an embodiment of the present invention provides a sample data collection method, where the method includes:
  • Obtaining a feature file of the service to be tested during operation where the feature file is used to identify the location a file describing characteristics of a data packet transmitted by the test service during operation;
  • Sample data is sent to the server, the sample data including the background information and the signature file.
  • the background information includes a service type of the service to be tested
  • the method further includes:
  • test report includes: a service type of the service to be tested and the estimated service type;
  • the test report is displayed.
  • the sending sample data to the server includes:
  • the feature file is sent to the server at a preset rate.
  • the background information includes geographic location information of the terminal, a network standard of the terminal, an internet protocol IP address of the terminal, and a model number of the terminal.
  • the signature file includes time information, IP address, port information, protocol type, and packet length.
  • an embodiment of the present invention provides a sample data collection method, where the method includes:
  • sample data of a service to be tested of the terminal where the sample data includes background information and a feature file;
  • the background information and the storage path of the feature file are stored in a feature database.
  • the method further includes:
  • the method further includes:
  • the server acquires at least one background information in the feature database and a storage path of the feature file corresponding to each background information every preset time length;
  • the category recognition model is trained according to each background information and a feature file corresponding to each background information.
  • an embodiment of the present invention provides a sample data collection device, where the device includes:
  • a first acquiring module configured to acquire background information of a service to be tested, where the background information includes a service type of the service to be tested;
  • a service running module configured to run the service to be tested
  • a second obtaining module configured to acquire a feature file of the service to be tested during operation, where the feature file is used to identify a file of a feature of the data packet transmitted by the service to be tested during operation;
  • a sending module configured to send sample data to the server, where the sample data includes the background information and the feature file.
  • the background information includes a service type of the service to be tested
  • the device also includes:
  • a receiving module configured to receive an estimated service type sent by the server, where the estimated service type is determined by the server according to the feature file;
  • a generating module configured to generate a test report according to the estimated service type, where the test report includes: a service type of the service to be tested and the estimated service type;
  • a display module for displaying the test report.
  • the sending module is further configured to send the background information to the server, and send the feature file to the server at a preset rate after the background information is sent.
  • the background information includes geographic location information of the terminal, a network standard of the terminal, an internet protocol IP address of the terminal, and a model number of the terminal.
  • the signature file includes time information, IP address, port information, protocol type, and packet length.
  • an embodiment of the present invention provides a sample data collection device, where the device includes:
  • a first acquiring module configured to acquire sample data of a service to be tested of the terminal, where the sample data includes background information and a feature file;
  • a determining module configured to determine a storage path of the feature file
  • a storage module configured to store the background information and a storage path of the feature file to a feature database.
  • the device further includes:
  • a calling module configured to invoke a type identification model, and determine an estimated service type of the service to be tested according to the sample data
  • a sending module configured to send, to the terminal, an estimated service type determined by the server, so that the terminal generates a test report according to the estimated service type.
  • the device further includes:
  • a second acquiring module configured to acquire at least one background information in the feature database and a storage path of a feature file corresponding to each background information, for each preset duration;
  • a third acquiring module configured to acquire, according to a storage path of the feature file corresponding to each background information, a feature file corresponding to each background information;
  • a training module configured to train the category recognition model according to each background information and a feature file corresponding to each background information.
  • an embodiment of the present invention provides a sample data collection system, where the system includes: at least one terminal and a server;
  • Each of the at least one terminal is configured to obtain background information of a service to be tested, run the service to be tested, obtain a feature file of the service to be tested, and send sample data to the server.
  • the background information includes a service type of the service to be tested, and the feature file is used to identify a file of a feature of the data packet transmitted by the service to be tested during operation, the sample data including the background information and The feature file;
  • the server is configured to obtain sample data of the service to be tested of the terminal, determine a storage path of the feature file, and store the background information and a storage path of the feature file to a feature database.
  • the terminal obtains the background information of the service to be tested, and runs the service to be tested, obtains the feature file during the running process of the service to be tested, and sends sample data including the background information and the feature file to the server, and the server obtains the sample sent by the terminal. Data, and determine the storage path of the feature file in the sample data, and store the background information and the storage path of the feature file to the feature database.
  • the sample data including the background information and the feature file is sent to the server, and the terminal can quickly send the sample data to the server, and the server can also quickly Obtain sample data, reducing the need for the server to obtain sample data The time increases the efficiency of the server to obtain sample data and reduces the cost of the server to obtain sample data.
  • FIG. 1 is a schematic structural diagram of a sample data collection system according to an embodiment of the present invention.
  • FIG. 2 is a flowchart of a sample data collection method according to an embodiment of the present invention.
  • FIG. 3 is a flowchart of a sample data collection method according to an embodiment of the present invention.
  • FIG. 4 is a flowchart of a sample data collection method according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of a sample data collection system according to an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of a sample data collection device according to an embodiment of the present invention.
  • FIG. 7 is a schematic structural diagram of a sample data collection apparatus according to an embodiment of the present invention.
  • FIG. 8 is a schematic structural diagram of a sample data collection apparatus according to an embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram of a sample data collection apparatus according to an embodiment of the present invention.
  • FIG. 10 is a schematic structural diagram of a sample data collection apparatus according to an embodiment of the present invention.
  • FIG. 11 is a schematic structural diagram of a terminal according to an embodiment of the present disclosure.
  • FIG. 12 is a schematic structural diagram of a server according to an embodiment of the present invention.
  • the sample data collection system includes: a server and at least one terminal, and each terminal and the server are respectively connected through a network.
  • each terminal establishes a network connection with the server, and each terminal can run different services to be tested, thereby acquiring different sample data.
  • the embodiment of the present invention is explained by taking only one terminal as an example. The embodiment of the present invention does not limit the number of terminals.
  • the terminal obtains the background information of the pre-set service to be tested, and runs the service to be tested. During the running of the service to be tested, the terminal acquires the feature file of the service to be tested. After the test is completed, the Beijing information and the feature file of the service to be tested are integrated into sample data and sent to the server.
  • the background information includes a service type of the service to be tested, and the feature file is used to identify a file of a feature of the data packet transmitted by the service to be tested during the running process.
  • the server After receiving the sample data sent by the terminal, the server stores the sample data, and stores the background information in the sample data and the storage path of the storage feature file in the feature database, so as to subsequently train the type recognition model.
  • FIG. 2 is a flowchart of a sample data collection method according to an embodiment of the present invention. As shown in FIG. 2, the sample data collection method is applied to a terminal of the sample data collection system shown in FIG. 1, and includes the following steps:
  • step 201 background information of a service to be tested set in advance is obtained.
  • the background information includes a service type of the service to be tested.
  • step 202 the service to be tested is run.
  • step 203 the feature file of the service to be tested is obtained during the running process.
  • the feature file is used to identify a file of the characteristics of the data packet transmitted by the service to be tested during the running process.
  • step 204 sample data is sent to the server.
  • the sample data includes background information and a feature file.
  • the sample data collection method obtaineds the background information of the service to be tested, and runs the service to be tested, and acquires the feature file in the running process of the service to be tested, to the server.
  • the sample data including the background information and the feature file is sent, so that the server acquires the sample data sent by the terminal, and determines the storage path of the feature file in the sample data, and stores the background information and the storage path of the feature file to the feature database.
  • sample data including the background information and the feature file is sent to the server, and the terminal can quickly send the sample data to the server, and the server can also quickly Obtaining sample data reduces the time required for the server to obtain sample data, improves the efficiency of the server to obtain sample data, and reduces the cost of the server to obtain sample data.
  • the background information includes a service type of the service to be tested
  • the method further includes:
  • the estimated service type is the server according to the special Determined by the document;
  • test report includes: a service type of the service to be tested and the estimated service type;
  • the test report is displayed.
  • the sending the sample data to the server includes:
  • the feature file is sent to the server at a preset rate.
  • the background information includes geographic location information of the terminal, a network standard of the terminal, an internet protocol IP address of the terminal, and a model number of the terminal.
  • the feature file includes time information, an IP address, port information, a protocol type, and a packet length.
  • FIG. 3 is a flowchart of a sample data collection method according to an embodiment of the present invention. As shown in FIG. 3, the sample data collection method is applied to a server of the sample data collection system shown in FIG. 1, and includes the following steps:
  • step 301 sample data of the service to be tested of the terminal is obtained.
  • the sample data includes background information and a feature file
  • step 302 a storage path of the signature file is determined.
  • step 303 the background information and the storage path of the feature file are stored in the feature database.
  • the sample data collection method obtaineds the background information of the service to be tested, and runs the service to be tested, and acquires the feature file in the running process of the service to be tested, to the server.
  • the server acquires the sample data sent by the terminal, and determines a storage path of the feature file in the sample data, and stores the background information and the storage path of the feature file to the feature database.
  • sample data including the background information and the feature file is sent to the server, and the terminal can quickly send the sample data to the server, and the server can also quickly Obtaining sample data reduces the time required for the server to obtain sample data, improves the efficiency of the server to obtain sample data, and reduces the cost of the server to obtain sample data.
  • the method further includes:
  • the estimated service type determined by the server is sent to the terminal, so that the terminal generates a test report according to the estimated service type.
  • the method further includes:
  • the server acquires at least one background information in the feature database and a storage path of the feature file corresponding to each background information every preset time period;
  • the category recognition model is trained according to each background information and a feature file corresponding to each background information.
  • FIG. 4 is a flowchart of a sample data collection method according to an embodiment of the present invention.
  • the interaction main body of the sample data collection method is a server and a terminal, and includes the following steps:
  • step 401 the terminal acquires background information of a service to be tested set in advance.
  • the services to be tested may be services such as video play, web browsing, and VoIP (Voice-over-IP), which are not limited in this embodiment of the present invention.
  • the background information may include the service type of the service to be tested, the geographical location information of the terminal, the network standard of the terminal, the IP (Internet Protocol) address of the terminal, and the model number of the terminal.
  • the terminal can obtain the service type of the service to be tested and the geographical location information of the terminal according to the operation triggered by the user.
  • the terminal can also obtain the model of the terminal according to the configuration file of the terminal, and the configuration file is used to describe the brand information of the terminal.
  • the terminal since the terminal accesses the current mobile network when performing the test, the terminal can determine the network standard of the terminal and the IP address of the terminal according to the currently accessed mobile network.
  • the current mobile network may include LTE (Long-Term Evolution) and UMTS (Universal Mobile Telecommunications System), which are not limited in this embodiment of the present invention.
  • the terminal can install the DFI application.
  • the terminal wants to test the test service, and the terminal can test the test service by running the DFI application.
  • the terminal detects the user-triggered option to start DFI
  • the terminal starts the DFI application and starts a shell (command parser) to display an interface for setting background information, and the user can treat the service type of the service and the geographic location of the terminal.
  • the information is selected, and the network standard of the terminal, the IP address of the terminal, and the model of the terminal are automatically obtained, thereby completing the acquisition of the service to be tested. Background Information.
  • step 402 the terminal runs the service to be tested.
  • each type of service to be tested can be run when the service to be tested is run through the terminal, so that multiple types of data are acquired in subsequent steps.
  • the terminal may run the first service to be tested according to the preset number of times.
  • the terminal does not Running the first service to be tested. Instead, the second service to be tested is run until the number of times each type of service to be tested has been run a predetermined number of times.
  • the preset number of times can be adjusted according to the service to be tested, which is not limited by the embodiment of the present invention.
  • the terminal can run the service to be tested according to the operation triggered by the user, and the service to be tested can be automatically obtained after the background information is obtained, which is not limited by the embodiment of the present invention.
  • the terminal launches the DFI application and obtains the background information of the service to be tested
  • the terminal after detecting the option that the user triggers the start of the test, the terminal starts the command to start the service to be tested, thereby starting the service to be tested.
  • step 403 the terminal acquires a feature file during the running process of the service to be tested.
  • the feature file is used to identify a file of the characteristics of the data packet transmitted by the service to be tested during the running process.
  • the profile may include time information, IP address, port information, protocol type, and packet length.
  • the time information may be time information generated by the packet;
  • the IP address may include an IP address of the terminal and an IP address of a service to be accessed by the service to be tested;
  • the port information may include a port address applied by the IP address of the terminal, and
  • the protocol type may include TCP (Transmission Control Protocol) and UDP (User Data Protocol);
  • the length of the data packet may be the packet length of the application layer. But does not include content data.
  • the feature file may also include information such as the domain name to be accessed, which is not limited by the embodiment of the present invention.
  • step 404 the terminal transmits sample data to the server.
  • the sample data may include the background information and the feature file, that is, after acquiring the background information and the feature file of the service to be tested, the terminal sends the background information and the feature file of the service to be tested as sample data to the server.
  • the terminal After the terminal runs the service to be tested, it can send the service to be tested to the server.
  • the sample data may continue to run the service to be tested until the number of times of the operation reaches a preset number of times, and then the sample data is sent to the server together with the sample data, which is not limited in the embodiment of the present invention.
  • the preset number of times of running the service to be tested is 10 times
  • the terminal starts the service to be tested through the DFI application, and obtains the service to be tested through the tcpdump (instruction for intercepting data) instruction during the running of the service to be tested.
  • the feature file when the service to be tested runs once, stops executing the tcpdump command and stores the sample data of the current time, and starts the service to be tested again, until the running of the service to be tested is 10 times, and the sample data of 10 times is sent together. To the server.
  • the terminal when it sends the sample data to the server, it can connect to the port 20 and 21 of the server. After the connection is successful, the terminal first sends the background information of the service to be tested for the first time, and after the background information is sent, The method of transmitting the feature to be tested for the first time is sent to the server, and the background information and the feature file of the service to be tested are sent to the server each time.
  • the preset rate may be adjusted according to the feature file of the service to be tested, and is usually set to 8 KB (kilobytes)/time, and may be set to other rates, which is not limited in this embodiment of the present invention.
  • step 405 the server acquires sample data of the service to be tested of the terminal, and determines a storage path of the feature file.
  • the server receiving terminal transmits sample data and stores background information and a feature file in the sample data.
  • the storage path of the feature file needs to be obtained when the feature file is stored, so that the subsequent step calls the feature file.
  • the server may store the feature file in a preset path, or store the feature file in an arbitrary storage space, and obtain a storage path where the feature file is located, and may also determine the feature by other means.
  • the storage path of the file is not limited by the embodiment of the present invention.
  • step 406 the server stores the background information and the storage path of the signature file to the feature database.
  • the feature database is a preset storage space in the server, and is configured to store sample data according to the background information of the service to be tested and the type of the service to be tested corresponding to the sample data.
  • the server determines the storage path of the feature file of the service to be tested
  • the background information of the service to be tested and the storage path of the feature file are stored in the service feature database, so as to facilitate different types of services to be tested and different background information. Call the sample data.
  • the server will background information and feature files. After the storage path is stored in the feature database, the collection of the sample data is completed. In the embodiment of the present invention, the process of collecting the sample data ends.
  • step 407 the server invokes a type identification model to determine an estimated service type of the service to be tested based on the sample data.
  • the type identification model is obtained by training according to the collected sample data in advance, and can determine the service type of the service to be tested.
  • the server can input the sample data into the type identification model by calling the type identification model, and the type identification model judges according to the length of the data packet in the sample data, thereby determining the estimated service type of the service to be tested corresponding to the sample data.
  • the estimated service type is determined by the server according to the feature file.
  • the server may further compare with the estimated service type determined by the type identification model according to the service type in the background information of the service to be tested, thereby obtaining the accuracy of the type identification model.
  • the server may not compare the service type of the test service with the estimated service type, which is not limited by the embodiment of the present invention.
  • the accuracy is used to indicate that the type identification model identifies the correct probability when identifying the service type of the service to be tested.
  • the server may obtain at least one background information in the feature database and a storage path of the feature file corresponding to each background information, and obtain each background according to the storage path of the feature file corresponding to each background information.
  • the feature file corresponding to the information is trained according to each background information and a feature file corresponding to each background information.
  • the accuracy of the category recognition model can be improved, thus ensuring the correctness of the business type determined by the category recognition model.
  • step 408 the server sends the terminal the estimated service type determined by the server.
  • the server may determine the IP address, model, and other information of the terminal to which the service to be tested belongs, because the background information of the service to be tested includes the background information of the service to be tested.
  • the terminal to which the service to be tested belongs and sends the estimated type of the service to be tested to the terminal, so that the terminal generates a test report according to the estimated service type in the subsequent step.
  • step 409 the terminal receives the estimated service type sent by the server, generates a test report according to the estimated service type, and displays the test report.
  • the test report may include a service type of the service to be tested and the estimated service type.
  • the terminal receives the estimated service type sent by the server, and compares the estimated service type with the service type in the background information of the service to be tested, thereby generating a test report, and displaying the test report, so that the user can conveniently test
  • the location is read to determine the accuracy of the category identification model when the category identification model determines the type of business for the test location.
  • the terminal may run the service to be tested multiple times, the terminal may receive the estimated service type determined by the server for each service to be tested, and the terminal may also run the service to be tested of multiple service types, and the terminal may It is also possible to receive the estimated service type determined by the server for the service to be tested for each type of service.
  • the embodiment of the present invention does not limit the manner in which the terminal receives the estimated service type.
  • the test report generated by the terminal is also for multiple service types and running the service to be tested multiple times.
  • the test report may include the estimated service type of the service to be tested each time the service to be tested is run multiple times.
  • the service to be tested may be marked when the service type of the service to be tested is inconsistent with the estimated service type.
  • the content and format of the test report are not limited in the embodiment of the present invention.
  • the sample data collection system may include a plurality of servers, and may include an interaction server that performs data interaction with the terminal, a storage server that stores sample data, and a training server that trains the type identification model.
  • the embodiment of the present invention is described by using only one server as an example. The number of servers in the embodiment of the present invention is not limited.
  • the terminal acquires the feature file of the service to be tested, and the step 403 may be performed by the server in the embodiment of the present invention. That is, the MBB Insider (Mobile Broad Band Insider) application of the terminal establishes a connection with the server according to the Websocket protocol, and sends background information of the service to be tested of the terminal to the server.
  • the server receives the background information sent by the terminal, parses the background information, determines the IP address of the terminal, starts a packet capture procedure of the Gi port (the interface between the GPRS and the external packet data network), and tracks the data packet of the IP address of the terminal.
  • the server can obtain the feature file of the service to be tested.
  • the device can send the service stop information to the server according to the operation triggered by the user, or automatically according to the preset test parameters.
  • the service stop information is sent to the server, the server receives the service stop information, and stops executing the packet capture process, thereby acquiring the feature file of the service to be tested.
  • the MBB Insider application is connected to the test network. Unlike the existing mobile network, there is no delay or jitter in the test network.
  • the preset test parameters may include the number of times the terminal runs the service to be tested, and the length of time the terminal runs the service to be tested.
  • Business stop information is used to indicate service The server stops running the packet capture program.
  • the terminal when the terminal establishes the operation 10 times and plays the video playback service for 1 minute each time, the terminal sends the background information of the video playback to the server, and the server starts the packet capture process.
  • the video playback service runs, it is determined whether the video needs to be started again. Play the service, if necessary, start the video playback service again. If it is not needed, send the service stop information to the server, so that the server stops running the packet capture program according to the service stop information.
  • the server when obtaining the feature file of the service to be tested through the MBB Insider application of the terminal, the server can simulate the delay, jitter, and the like existing in the mobile network, and can also set the bandwidth occupied by the terminal.
  • the sample data collection method obtaineds the background information of the service to be tested, and runs the service to be tested, and acquires the feature file in the running process of the service to be tested, to the server.
  • the server acquires the sample data sent by the terminal, and determines a storage path of the feature file in the sample data, and stores the background information and the storage path of the feature file to the feature database.
  • sample data including the background information and the feature file is sent to the server, and the terminal can quickly send the sample data to the server, and the server can also quickly Obtaining sample data reduces the time required for the server to obtain sample data, improves the efficiency of the server to obtain sample data, and reduces the cost of the server to obtain sample data.
  • FIG. 6 is a schematic structural diagram of a sample data collection device 60 according to an embodiment of the present invention. As shown in FIG. 6, the sample data collection device 60 is applied to a terminal, and the sample data collection device 60 includes: a first acquisition module 601, The service running module 602, the second obtaining module 603, and the sending module 604.
  • the first obtaining module 601 is configured to obtain preset background information of the service to be tested, where the background information includes a service type of the service to be tested;
  • the service running module 602 is configured to run the service to be tested
  • the second obtaining module 603 is configured to acquire a feature file of the service to be tested during operation, where the feature file is used to identify a file of a feature of the data packet transmitted by the service to be tested during the running process;
  • the sending module 604 is configured to send sample data to the server, where the sample data includes background information and a feature file.
  • the sample data collection device acquires a pre- Setting the background information of the service to be tested, and running the service to be tested, acquiring the feature file during the running process of the service to be tested, sending sample data including the background information and the feature file to the server, and obtaining the sample data sent by the terminal, and determining
  • the storage path of the feature file in the sample data stores the background information and the storage path of the feature file to the feature database.
  • sample data including the background information and the feature file is sent to the server, and the terminal can quickly send the sample data to the server, and the server can also quickly Obtaining sample data reduces the time required for the server to obtain sample data, improves the efficiency of the server to obtain sample data, and reduces the cost of the server to obtain sample data.
  • the background information includes a service type of the service to be tested
  • the apparatus further includes: a receiving module 605, a generating module 606, and a display module 607.
  • the receiving module 605 is configured to receive an estimated service type sent by the server, where the estimated service type is determined by the server according to the feature file;
  • the generating module 606 is configured to generate a test report according to the estimated service type, where the test report includes: a service type of the service to be tested and the estimated service type;
  • the display module 607 is configured to display the test report.
  • the sending module is further configured to send the background information to the server, and send the feature file to the server at a preset rate after the background information is sent.
  • the background information includes geographic location information of the terminal, a network standard of the terminal, an internet protocol IP address of the terminal, and a model number of the terminal.
  • the feature file includes time information, an IP address, port information, a protocol type, and a packet length.
  • FIG. 8 is a schematic structural diagram of a sample data collection device 80 according to an embodiment of the present invention. As shown in FIG. 8, the sample data collection device 80 is applied to a server, and the sample data collection device 80 includes: a first acquisition module 801, The module 802 and the storage module 803 are determined.
  • the first obtaining module 801 is configured to acquire sample data of a service to be tested of the terminal, where the sample data includes background information and a feature file;
  • the determining module 802 is configured to determine a storage path of the feature file
  • the storage module 803 is configured to store the background information and a storage path of the feature file in a feature database.
  • the sample data collection device obtains the background information of the service to be tested set by the terminal, and runs the service to be tested, in the running process of the service to be tested.
  • Obtaining a feature file sending sample data including background information and a feature file to the server, acquiring, by the server, sample data sent by the terminal, determining a storage path of the feature file in the sample data, and storing the background information and the storage path of the feature file to the feature database .
  • sample data including the background information and the feature file is sent to the server, and the terminal can quickly send the sample data to the server, and the server can also quickly Obtaining sample data reduces the time required for the server to obtain sample data, improves the efficiency of the server to obtain sample data, and reduces the cost of the server to obtain sample data.
  • the apparatus further includes: a calling module 804 and a sending module 805.
  • the calling module 804 is configured to invoke a type identification model, and determine an estimated service type of the service to be tested according to the sample data.
  • the sending module 805 is configured to send, to the terminal, an estimated service type determined by the server, so that the terminal generates a test report according to the estimated service type.
  • the apparatus further includes: a second obtaining module 806, a third obtaining module 807, and a training module 808.
  • the second obtaining module 806 is configured to acquire, by the server, at least one background information in the feature database and a storage path of the feature file corresponding to each background information.
  • the third obtaining module 807 is configured to acquire, according to a storage path of the feature file corresponding to each background information, a feature file corresponding to each background information;
  • the training module 808 is configured to train the category recognition model according to each background information and a feature file corresponding to each background information.
  • FIG. 11 is a schematic structural diagram of a terminal 110 according to an embodiment of the present invention.
  • the terminal 110 is applied to the sample data collection system shown in FIG. 1.
  • the terminal 110 includes a receiver 1101, a transmitter 1102, a memory 1103, and a processor 1104.
  • the receiver 1101, the transmitter 1102, and the memory 1103 are respectively connected to the processor 1104.
  • the memory 1103 is configured to store programs and data, and the memory 1103 may include a cache memory, a high-speed RAM memory, and a non-volatile memory, such as at least one disk memory. It can be understood that the memory 1103 can be a non-transient program that can store program code, such as a random access memory (RAM), a magnetic disk, a hard disk, a solid state disk (SSD), or a nonvolatile memory. Non-transitory machine readable medium.
  • RAM random access memory
  • SSD solid state disk
  • the processor 1104 may be a central processing unit CPU or a specific integrated circuit ASIC. (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present invention. For example, processing input/output (I/O) requests to disk, backing up data from production volumes to target volumes, and so on. Therefore, the prompt device of the code conflict can implement functions such as IO operation and data backup.
  • the processor 1104 is configured to execute a program, and specifically, the related steps in the following method embodiments may be performed.
  • Programs can include operating systems, file systems, and other software modules. Different software modules have different functions.
  • the processor 1104 is configured to obtain preset background information of the service to be tested, where the background information includes a service type of the service to be tested.
  • the processor 1104 is further configured to run the service to be tested
  • the processor 1104 is further configured to acquire a feature file of the service to be tested during operation, where the feature file is used to identify a file of a feature of the data packet transmitted by the service to be tested during the running process;
  • the processor 1104 is further configured to send, by the transmitter 1102, sample data to the server, where the sample data includes the background information and the feature file.
  • the processor 1104 is further configured to receive, by the receiver 1101, an estimated service type sent by the server, where the estimated service type is determined by the server according to the feature file;
  • the processor 1104 is further configured to generate a test report according to the estimated service type, where the test report includes: a service type of the service to be tested and the estimated service type;
  • the processor 1104 is also configured to display the test report.
  • the processor 1104 is further configured to send the background information to the server by using the transmitter 1102.
  • the processor 1104 is further configured to send, by the transmitter 1102, the feature file to the server at a preset rate after the background information is sent.
  • FIG. 12 is a schematic structural diagram of a server 120 according to an embodiment of the present invention.
  • the server 120 is applied to the sample data collection system shown in FIG. 1.
  • the server 120 includes a receiver 1201, a transmitter 1202, a memory 1203, and a processor 1204.
  • the receiver 1201, the transmitter 1202, and the memory 1203 are respectively connected to the processor 1204.
  • the memory 1203 is configured to store program data of the program and the at least one terminal of the service to be tested.
  • the memory 1203 may include a cache memory, a high speed RAM memory, and a non-volatile memory. ), for example at least one disk storage. It can be understood that the memory 1203 can be a random access memory (RAM), a magnetic disk, a hard disk, a solid state disk (SSD), or a nonvolatile memory.
  • RAM random access memory
  • SSD solid state disk
  • a variety of non-transitory machine readable media that can store program code.
  • the processor 1204 may be a central processing unit CPU, or an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. For example, processing input/output (I/O) requests to disk, backing up data from production volumes to target volumes, and so on. Therefore, the sample data collection device can implement functions such as IO operation and data backup.
  • the processor 1204 is configured to execute a program, and specifically, the related steps in the following method embodiments may be performed.
  • Programs can include operating systems, file systems, and other software modules. Different software modules have different functions.
  • the processor 1204 is configured to obtain, by using the receiver 1201, sample data of a service to be tested of the terminal, where the sample data includes background information and a feature file;
  • the processor 1204 is further configured to determine a storage path of the feature file
  • the processor 1204 is further configured to store the background information and the storage path of the feature file to the feature database through the memory 1203.
  • the processor 1204 is further configured to invoke a type identification model, and determine, according to the sample data, an estimated service type of the service to be tested;
  • the processor 1204 is further configured to send, by the transmitter 1202, the estimated service type determined by the server to the terminal, so that the terminal generates a test report according to the estimated service type.
  • the processor 1204 is further configured to: acquire, by the server, at least one background information in the feature database and a storage path of the feature file corresponding to each background information;
  • the processor 1204 is further configured to acquire, according to a storage path of the feature file corresponding to each background information, a feature file corresponding to each background information;
  • the processor 1204 is further configured to train the category recognition model according to each background information and a feature file corresponding to each background information.
  • a person skilled in the art may understand that all or part of the steps of implementing the above embodiments may be completed by hardware, or may be instructed by a program to execute related hardware, and the program may be stored in a computer readable storage medium.
  • the storage medium mentioned may be a read only memory, a magnetic disk or an optical disk or the like.

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明公开了一种样本数据采集方法、装置及系统，属于信息技术领域。所述方法包括：终端获取预先设置的待测试业务的背景信息，运行待测试业务，获取待测试业务在运行过程中的特征文件，向服务器发送样本数据。服务器获取终端的待测试业务的样本数据，确定特征文件的存储路径，将背景信息和特征文件的存储路径，存储至特征数据库。本发明通过在终端上运行待测试业务，并记录待测试业务的特征文件，在测试完毕后向服务器发送包括背景信息和特征文件的样本数据，可以快速地向服务器发送样本数据，服务器也可以快速地获取样本数据，减小了服务器获取样本数据所需要的时间，提高了服务器获取样本数据的效率，降低了服务器获取样本数据所花费的成本。

Description

样本数据采集方法、装置及系统 技术领域
本发明涉及信息技术领域，特别涉及一种样本数据采集方法、装置及系统。
背景技术
为了优化无线资源调度，通常会采用DFI(Deep/Dynamic Flow Inspection，深度/动态流检测技术)技术识别无线网络中待识别业务的业务类型，从而根据不同的业务类型对无线资源进行优化。
实际应用中，如果采用DFI技术识别业务类型，则需要通过预先训练得到的类型识别模型确定待识别业务的业务类型。具体的训练过程包括：终端在指定的地理位置多次运行待识别业务，在运行待识别业务的过程中记录待识别业务的样本数据，运行完毕后测试人员将终端从指定的地理位置带回实验室，并将终端与服务器通过数据线连接，服务器获取终端记录的待识别业务的样本数据，根据获取的多个样本数据训练类型识别模型。其中，待识别业务的样本数据可以包括待识别业务的业务类型和待识别业务的数据包长度。
在实现本发明的过程中，发明人发现现有技术至少存在以下问题：
在训练类型识别模型时，需要测试人员将记录样本数据的终端带回实验室，并将终端与实验室的服务器连接，以便服务器获取样本数据，导致服务器获取样本数据的周期过长，成本过高的问题。
发明内容
为了解决现有技术的问题，本发明实施例提供了一种样本数据采集方法、装置及系统。所述技术方案如下：
第一方面，本发明实施例提供了一种样本数据采集方法，所述方法包括：
获取预先设置的待测试业务的背景信息，所述背景信息包括待测试业务的业务类型；
运行所述待测试业务；
获取所述待测试业务在运行过程中的特征文件，所述特征文件用于标识所 述待测试业务在运行过程中传输的数据包的特征的文件；
向服务器发送样本数据，所述样本数据包括所述背景信息和所述特征文件。
在一种可能的设计中，所述背景信息包括所述待测试业务的业务类型；
所述向服务器发送样本数据之后，所述方法还包括：
接收所述服务器发送的预估业务类型，所述预估业务类型是所述服务器根据所述特征文件确定的；
根据所述预估业务类型生成测试报告，所述测试报告包括：所述待测试业务的业务类型和所述预估业务类型；
显示所述测试报告。
在一种可能的设计中，所述向服务器发送样本数据，包括：
向所述服务器发送所述背景信息；
在所述背景信息发送完成后，以预设速率向所述服务器发送所述特征文件。
在一种可能的设计中，所述背景信息包括所述终端的地理位置信息、所述终端的网络制式、所述终端的网际协议IP地址和所述终端的型号。
在一种可能的设计中，所述特征文件包括时间信息、IP地址、端口信息、协议类型和数据包长度。
第二方面，本发明实施例提供了一种样本数据采集方法，所述方法包括：
获取终端的待测试业务的样本数据，所述样本数据包括背景信息和特征文件；
确定所述特征文件的存储路径；
将所述背景信息和所述特征文件的存储路径，存储至特征数据库。
在一种可能的设计中，在所述将所述背景信息和所述特征文件的存储路径，存储至特征数据库之后，所述方法还包括：
调用类型识别模型，根据所述样本数据确定所述待测试业务的预估业务类型；
向所述终端发送所述服务器确定的预估业务类型，以便所述终端根据所述预估业务类型生成测试报告。
在一种可能的设计中，在所述将所述背景信息和所述特征文件的存储路径，存储至特征数据库之后，所述方法还包括：
每隔预设时长，服务器获取所述特征数据库中的至少一个背景信息以及每个背景信息对应的特征文件的存储路径；
根据每个背景信息对应的特征文件的存储路径，获取每个背景信息对应的特征文件；
根据每个背景信息以及每个背景信息对应的特征文件，对所述类别识别模型进行训练。
第三方面，本发明实施例提供了一种样本数据采集装置，所述装置包括：
第一获取模块，用于获取预先设置的待测试业务的背景信息，所述背景信息包括待测试业务的业务类型；
业务运行模块，用于运行所述待测试业务；
第二获取模块，用于获取所述待测试业务在运行过程中的特征文件，所述特征文件用于标识所述待测试业务在运行过程中传输的数据包的特征的文件；
发送模块，用于向服务器发送样本数据，所述样本数据包括所述背景信息和所述特征文件。
在一种可能的设计中，所述背景信息包括所述待测试业务的业务类型；
所述装置还包括：
接收模块，用于接收所述服务器发送的预估业务类型，所述预估业务类型是所述服务器根据所述特征文件确定的；
生成模块，用于根据所述预估业务类型生成测试报告，所述测试报告包括：所述待测试业务的业务类型和所述预估业务类型；
显示模块，用于显示所述测试报告。
在一种可能的设计中，所述发送模块还用于向所述服务器发送所述背景信息，在所述背景信息发送完成后，以预设速率向所述服务器发送所述特征文件。
在一种可能的设计中，所述背景信息包括所述终端的地理位置信息、所述终端的网络制式、所述终端的网际协议IP地址和所述终端的型号。
在一种可能的设计中，所述特征文件包括时间信息、IP地址、端口信息、协议类型和数据包长度。
第四方面，本发明实施例提供了一种样本数据采集装置，所述装置包括：
第一获取模块，用于获取终端的待测试业务的样本数据，所述样本数据包括背景信息和特征文件；
确定模块，用于确定所述特征文件的存储路径；
存储模块，用于将所述背景信息和所述特征文件的存储路径，存储至特征数据库。
在一种可能的设计中，所述装置还包括：
调用模块，用于调用类型识别模型，根据所述样本数据确定所述待测试业务的预估业务类型；
发送模块，用于向所述终端发送所述服务器确定的预估业务类型，以便所述终端根据所述预估业务类型生成测试报告。
在一种可能的设计中，所述装置还包括：
第二获取模块，用于每隔预设时长，服务器获取所述特征数据库中的至少一个背景信息以及每个背景信息对应的特征文件的存储路径；
第三获取模块，用于根据每个背景信息对应的特征文件的存储路径，获取每个背景信息对应的特征文件；
训练模块，用于根据每个背景信息以及每个背景信息对应的特征文件，对所述类别识别模型进行训练。
第五方面，本发明实施例提供了一种样本数据采集系统，所述系统包括：至少一个终端和服务器；
所述至少一个终端中的每个终端，用于获取预先设置的待测试业务的背景信息，运行所述待测试业务，获取所述待测试业务在运行过程中的特征文件，向服务器发送样本数据，其中，所述背景信息包括待测试业务的业务类型，所述特征文件用于标识所述待测试业务在运行过程中传输的数据包的特征的文件，所述样本数据包括所述背景信息和所述特征文件；
所述服务器用于获取终端的待测试业务的样本数据，确定所述特征文件的存储路径，将所述背景信息和所述特征文件的存储路径，存储至特征数据库。
本发明实施例提供的技术方案带来的有益效果是：
通过终端获取预先设置的待测试业务的背景信息，并运行待测试业务，在待测试业务的运行过程中获取特征文件，向服务器发送包括背景信息和特征文件的样本数据，服务器获取终端发送的样本数据，并确定样本数据中特征文件的存储路径，将背景信息和特征文件的存储路径存储至特征数据库。通过在终端上运行待测试业务，并记录待测试业务的特征文件，在测试完毕后向服务器发送包括背景信息和特征文件的样本数据，终端可以快速地向服务器发送样本数据，服务器也可以快速地获取样本数据，减小了服务器获取样本数据所需要 的时间，提高了服务器获取样本数据的效率，降低了服务器获取样本数据所花费的成本。
附图说明
为了更清楚地说明本发明实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本发明实施例提供的一种样本数据采集系统的结构示意图；
图2是本发明实施例提供的一种样本数据采集方法的流程图；
图3是本发明实施例提供的一种样本数据采集方法的流程图；
图4是本发明实施例提供的一种样本数据采集方法的流程图；
图5是本发明实施例提供的一种样本数据采集系统的结构示意图；
图6是本发明实施例提供的一种样本数据采集装置的结构示意图；
图7是本发明实施例提供的一种样本数据采集装置的结构示意图；
图8是本发明实施例提供的一种样本数据采集装置的结构示意图；
图9是本发明实施例提供的一种样本数据采集装置的结构示意图；
图10是本发明实施例提供的一种样本数据采集装置的结构示意图；
图11是本发明实施例提供的一种终端的结构示意图；
图12是本发明实施例提供的一种服务器的结构示意图。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚，下面将结合附图对本发明实施方式作进一步地详细描述。
图1是本发明实施例提供的一种样本数据采集系统的结构示意图，参见图1，该样本数据采集系统包括：服务器和至少一个终端，每个终端与该服务器之间分别通过网络连接。
需要说明的是，本发明实施例中每个终端均与服务器建立了网络连接，而且每个终端可以运行不同的待测试业务，从而获取不同的样本数据。而本发明实施例仅是以一个终端为例进行了解释说明，本发明实施例对终端的数量不做限制。
终端获取预先设置的待测试业务的背景信息，并运行待测试业务。在该待测试业务的运行过程中，终端获取该待测试业务的特征文件，测试完毕后，将待测试业务的北京信息和特征文件整合为样本数据发送至服务器。
其中，该背景信息包括待测试业务的业务类型，该特征文件用于标识该待测试业务在运行过程中传输的数据包的特征的文件。
服务器接收终端发送的样本数据后，对样本数据进行存储，将样本数据中的背景信息和存储特征文件的存储路径存储至特征数据库，以便后续对类型识别模型进行训练。
图2是本发明实施例提供的一种样本数据采集方法的流程图，如图2所示，该样本数据采集方法应用于图1所示的样本数据采集系统的终端中，包括以下步骤：
在步骤201中，获取预先设置的待测试业务的背景信息。
其中，该背景信息包括待测试业务的业务类型。
在步骤202中，运行待测试业务。
在步骤203中，获取待测试业务在运行过程中的特征文件。
其中，该特征文件用于标识待测试业务在运行过程中传输的数据包的特征的文件。
在步骤204中，向服务器发送样本数据。
其中，该样本数据包括背景信息和特征文件。
综上所述，本发明实施例中提供的样本数据采集方法，通过终端获取预先设置的待测试业务的背景信息，并运行待测试业务，在待测试业务的运行过程中获取特征文件，向服务器发送包括背景信息和特征文件的样本数据，以便服务器获取终端发送的样本数据，并确定样本数据中特征文件的存储路径，将背景信息和特征文件的存储路径存储至特征数据库。通过在终端上运行待测试业务，并记录待测试业务的特征文件，在测试完毕后向服务器发送包括背景信息和特征文件的样本数据，终端可以快速地向服务器发送样本数据，服务器也可以快速地获取样本数据，减小了服务器获取样本数据所需要的时间，提高了服务器获取样本数据的效率，降低了服务器获取样本数据所花费的成本。
可选地，该背景信息包括该待测试业务的业务类型；
该向服务器发送样本数据之后，该方法还包括：
接收该服务器发送的预估业务类型，该预估业务类型是该服务器根据该特 征文件确定的；
根据该预估业务类型生成测试报告，该测试报告包括：该待测试业务的业务类型和该预估业务类型；
显示该测试报告。
可选地，该向服务器发送样本数据，包括：
向该服务器发送该背景信息；
在该背景信息发送完成后，以预设速率向该服务器发送该特征文件。
可选地，该背景信息包括该终端的地理位置信息、该终端的网络制式、该终端的网际协议IP地址和该终端的型号。
可选地，该特征文件包括时间信息、IP地址、端口信息、协议类型和数据包长度。
图3是本发明实施例提供的一种样本数据采集方法的流程图，如图3所示，该样本数据采集方法应用于图1所示的样本数据采集系统的服务器中，包括以下步骤：
在步骤301中，获取终端的待测试业务的样本数据。
其中，该样本数据包括背景信息和特征文件；
在步骤302中，确定该特征文件的存储路径。
在步骤303中，将该背景信息和该特征文件的存储路径，存储至特征数据库。
综上所述，本发明实施例中提供的样本数据采集方法，通过终端获取预先设置的待测试业务的背景信息，并运行待测试业务，在待测试业务的运行过程中获取特征文件，向服务器发送包括背景信息和特征文件的样本数据，服务器获取终端发送的样本数据，并确定样本数据中特征文件的存储路径，将背景信息和特征文件的存储路径存储至特征数据库。通过在终端上运行待测试业务，并记录待测试业务的特征文件，在测试完毕后向服务器发送包括背景信息和特征文件的样本数据，终端可以快速地向服务器发送样本数据，服务器也可以快速地获取样本数据，减小了服务器获取样本数据所需要的时间，提高了服务器获取样本数据的效率，降低了服务器获取样本数据所花费的成本。
可选地，在将该背景信息和该特征文件的存储路径，存储至特征数据库之后，该方法还包括：
调用类型识别模型，根据该样本数据确定该待测试业务的预估业务类型；
向该终端发送该服务器确定的预估业务类型，以便该终端根据该预估业务类型生成测试报告。
可选地，在将该背景信息和该特征文件的存储路径，存储至特征数据库之后，该方法还包括：
每隔预设时长，服务器获取该特征数据库中的至少一个背景信息以及每个背景信息对应的特征文件的存储路径；
根据每个背景信息对应的特征文件的存储路径，获取每个背景信息对应的特征文件；
根据每个背景信息以及每个背景信息对应的特征文件，对该类别识别模型进行训练。
图4是本发明实施例提供的一种样本数据采集方法的流程图，如图4所示，该样本数据采集方法的交互主体为服务器和终端，包括以下步骤：
在步骤401中，终端获取预先设置的待测试业务的背景信息。
其中，待测试业务可以为视频播放、网页浏览和VoIP(Voice-over-IP，IP承载的语音)等业务，本发明实施例对此不做限定。背景信息可以包括待测试业务的业务类型、终端的地理位置信息、终端的网络制式、终端的IP(Internet Protocol，互联网协议)地址和终端的型号。
而且，终端可以根据用户触发的操作，获取待测试业务的业务类型和终端的地理位置信息。终端还可以根据终端的配置文件获取终端的型号，该配置文件用于描述终端的品牌信息。
另外，由于终端在进行测试时会接入当前的移动网络，因此终端可以根据当前接入的移动网络确定终端的网络制式、终端的IP地址。其中，当前的移动网络可以包括LTE(Long-Term Evolution，长期演进)和UMTS(Universal Mobile Telecommunications System，通用移动通信系统)，本发明实施例对此不做限定。
例如，终端可以安装DFI应用程序，当检测到用户启动DFI时，则说明用户想要对待测试业务进行测试，终端可以通过运行DFI应用程序对待测试业务进行测试。当终端检测到用户触发的启动DFI的选项时，则终端启动DFI应用程序，并启动一个shell(命令解析器)，显示设置背景信息的界面，用户可以对待测试业务的业务类型和终端的地理位置信息进行选择，并自动获取终端的网络制式、终端的IP地址和终端的型号等信息，从而完成获取待测试业务的 背景信息。
在步骤402中，终端运行待测试业务。
由于待测试业务包括多种类型，在通过终端运行待测试业务时，可以将每种类型的待测试业务都运行，以便在后续步骤中获取多种类型的数据。
具体地，当终端获取了待测试业务的背景信息后，终端可以按照预先设定的次数运行第一待测试业务，当第一待测试业务的运行次数到达预先设定的次数后，则不再运行该第一待测试业务。而是运行第二待测试业务，直到每种类型的待测试业务的运行次数都达到了预先设定的次数。
其中，该预先设定的次数可以根据待测试业务进行调整，本发明实施例对此不做限定。而且，终端可以根据用户触发的操作运行待测试业务，也可以在获取背景信息后自动获取待测试业务，本发明实施例对此不做限定。
例如，终端在启动DFI应用程序并获取了待测试业务的背景信息后，当检测到用户触发开始测试的选项后，执行启动待测试业务的命令，从而启动待测试业务。
在步骤403中，终端获取待测试业务在运行过程中的特征文件。
其中，该特征文件用于标识待测试业务在运行过程中传输的数据包的特征的文件。该特征文件可以包括时间信息、IP地址、端口信息、协议类型和数据包长度。
具体地，该时间信息可以为报文产生的时间信息；IP地址可以包括终端的IP地址和待测试业务所要访问的业务的IP地址；端口信息可以包括终端的IP地址所应用的端口地址，以及待测试业务所要访问的业务的端口地址；协议类型可以包括TCP(Transmission Control Protocol，传输控制协议)和UDP(User Data Protocol，用户数据报协议)；数据包长度可以为应用层的数据包长度，但是不包括内容数据。
而且，该特征文件还可以包括所要访问的域名等信息，本发明实施例对此不做限定。
在步骤404中，终端向服务器发送样本数据。
其中，该样本数据可以包括该背景信息和该特征文件，也即是终端在获取了待测试业务的背景信息和特征文件后，将待测试业务的背景信息和特征文件作为样本数据向服务器发送。
当终端运行完毕一次待测试业务后，可以向服务器发送本次待测试业务的 样本数据，也可以继续运行待测试业务，直到运行的次数达到了预先设定的次数，之后再将多次的样本数据一同发送至服务器，本发明实施例对此不做限定。
例如，预先设定的运行待测试业务的次数为10次，终端通过DFI应用程序启动待测试业务，并在待测试业务运行的过程中，通过tcpdump(截获数据的指令)指令获取待测试业务的特征文件，当待测试业务运行完毕一次后，停止执行tcpdump指令并存储本次的样本数据，并且再次启动待测试业务，直到运行待测试业务运行完毕10次后，将10次的样本数据一同发送至服务器。
而且，终端在向服务器发送样本数据时，可以连接服务器的20、21端口，在连接成功后先向服务器发送第一次运行待测试业务的背景信息，在该背景信息发送完毕后，再以预设速率循环多次发送的方式，向服务器发送第一次运行待测试业务的特征文件，以此方式将每次运行待测试业务的背景信息和特征文件发送至服务器。
其中，该预设速率可以根据待测试业务的特征文件进行调整，通常设置为8KB(千字节)/次，当然也可以设置为其他速率，本发明实施例对此不做限定。
在步骤405中，服务器获取终端的待测试业务的样本数据，确定特征文件的存储路径。
服务器接收终端发送样本数据，并存储样本数据中的背景信息和特征文件。其中，在存储特征文件时，需要获取特征文件的存储路径，以便后续步骤对特征文件进行调用。
例如，在存储特征文件时，服务器可以将特征文件存储在预设的路径，也可以将特征文件随机存储在任意存储空间，并获取该特征文件所在的存储路径，当然也可以通过其他方式确定特征文件的存储路径，本发明实施例对此不足限定。
在步骤406中，服务器将背景信息和特征文件的存储路径，存储至特征数据库。
其中，该特征数据库为服务器内预设的存储空间，用于根据样本数据所对应的待测试业务的背景信息和待测试业务的类型，存储样本数据。
在服务器确定了待测试业务的特征文件的存储路径后，就可以将待测试业务的背景信息和特征文件的存储路径存储至业务特征数据库，方便后续根据不同类型的待测试业务和不同的背景信息对样本数据进行调用。
需要说明的是，在执行步骤406后，也即是服务器将背景信息和特征文件 的存储路径存储至特征数据库后，则完成了对样本数据的采集，本发明实施例中对样本数据的采集过程结束。
在步骤407中，服务器调用类型识别模型，根据样本数据确定待测试业务的预估业务类型。
其中，该类型识别模型是预先根据采集的样本数据训练得到的，可以确定待测试业务的业务类型。
服务器通过调用类型识别模型，可以将样本数据输入该类型识别模型，该类型识别模型根据样本数据中的数据包长度进行判断，从而确定该样本数据对应的待测试业务的预估业务类型。其中，该预估业务类型是该服务器根据该特征文件确定的。
在确定待测试业务的预估业务类型后，服务器还可以根据待测试业务的背景信息中的业务类型，与类型识别模型确定的预估业务类型进行比较，从而得到类型识别模型的准确率。当然，服务器也可以不对待测试业务的业务类型和预估业务类型进行比较，本发明实施例对此不做限定。
其中，该准确率用于指示类型识别模型在识别待测试业务的业务类型时，识别正确的概率。
另外，服务器每隔预设时长，可以获取该特征数据库中的至少一个背景信息以及每个背景信息对应的特征文件的存储路径，根据每个背景信息对应的特征文件的存储路径，获取每个背景信息对应的特征文件，根据每个背景信息以及每个背景信息对应的特征文件，对该类别识别模型进行训练。
在经过长时间的训练后，类别识别模型的准确率即可得到提高，从而保证了通过类别识别模型确定业务类型的正确性。
在步骤408中，服务器向终端发送该服务器确定的预估业务类型。
在服务器确定了待测试业务的预估业务类型后，由于待测试业务的背景信息中包括该待测试业务所属的终端的IP地址、型号等信息，服务器可以根据该待测试业务的背景信息确定该待测试业务所属的终端，并向该终端发送该待测试业务的预估类型，以便在后续步骤中终端据该预估业务类型生成测试报告。
在步骤409中，终端接收该服务器发送的预估业务类型，根据该预估业务类型生成测试报告，并显示该测试报告。
其中，该测试报告可以包括该待测试业务的业务类型和该预估业务类型。
终端接收服务器发送的预估业务类型，并可以根据该预估业务类型与待测试业务的背景信息中的业务类型进行比较，进而生成测试报告，并显示该测试报告，以便用户能够方便地在测试地点进行阅读，从而确定类别识别模型在针对该测试地点确定业务类型时，该类别识别模型的准确率。
而且，由于终端可能运行多次待测试业务，则终端可能接收到服务器发送的针对每次待测试业务确定的预估业务类型，而且终端还有可能运行多种业务类型的待测试业务，则终端还可能接收到服务器发送的针对每种业务类型的待测试业务确定的预估业务类型，本发明实施例对终端接收预估业务类型的方式不做限定。
相应的，终端生成的测试报告也是针对多种业务类型和多次运行待测试业务的，该测试报告中可以包括多次运行待测试业务时，每次确定的该待测试业务的预估业务类型，还可以当待测试业务的业务类型与预估业务类型不一致时，对该待测试业务进行标示，本发明实施例对测试报告的内容、形式不做限定。
在实际应用中，如图5所示，样本数据采集系统可能包括多个服务器，可以包括与终端进行数据交互的交互服务器，存储样本数据的存储服务器以及训练类型识别模型的训练服务器。而本发明实施例仅是以一个服务器为例进行了说明，本发明实施例对服务器的数量不做限定。
需要说明的是，在步骤403中是终端获取待测试业务的特征文件，本发明实施例中还可以通过服务器来执行步骤403。也即是，通过终端的MBB Insider(Mobile Broad Band Insider，内部移动宽带)应用程序，根据Websocket协议与服务器建立连接，向服务器发送终端的待测试业务的背景信息。服务器接收终端发送的背景信息，对背景信息进行解析，确定终端的IP地址，启动Gi口(GPRS与外部分组数据网之间的接口)的抓包程序，跟踪终端的IP地址的数据包。当终端启动待测试业务后，服务器即可获取待测试业务的特征文件，当终端运行完毕待测试业务后，可以根据用户触发的操作向服务器发送业务停止信息，也可以根据预设的测试参数自动向服务器发送业务停止信息，服务器接收该业务停止信息，并停止执行抓包程序，从而获取了待测试业务的特征文件。
其中，MBB Insider应用程序是接入测试网络的，与接入现有的移动网络不同，测试网络不存在延迟、抖动等情况。预设的测试参数可以包括终端运行待测试业务的次数，终端运行待测试业务的时长。业务停止信息用于指示服 务器停止运行抓包程序。
例如，在终端建立运行10次，每次播放1分钟的视频播放业务，则终端向服务器发送视频播放的背景信息，服务器启动抓包程序，当视频播放业务运行完毕后，判断是否需要再次启动视频播放业务，如果需要则再次启动视频播放业务，如果不需要则向服务器发送业务停止信息，以便服务器根据该业务停止信息停止运行抓包程序。
另外，在通过终端的MBB Insider应用程序获取待测试业务的特征文件时，服务器可以模拟移动网络中存在的延迟、抖动等情况，还可以对终端所占用的带宽进行设置。
综上所述，本发明实施例中提供的样本数据采集方法，通过终端获取预先设置的待测试业务的背景信息，并运行待测试业务，在待测试业务的运行过程中获取特征文件，向服务器发送包括背景信息和特征文件的样本数据，服务器获取终端发送的样本数据，并确定样本数据中特征文件的存储路径，将背景信息和特征文件的存储路径存储至特征数据库。通过在终端上运行待测试业务，并记录待测试业务的特征文件，在测试完毕后向服务器发送包括背景信息和特征文件的样本数据，终端可以快速地向服务器发送样本数据，服务器也可以快速地获取样本数据，减小了服务器获取样本数据所需要的时间，提高了服务器获取样本数据的效率，降低了服务器获取样本数据所花费的成本。
上述所有可选技术方案，可以采用任意结合形成本公开的可选实施例，在此不再一一赘述。
图6是本发明实施例提供的一种样本数据采集装置60的结构示意图，如图6所示，该样本数据采集装置60应用于终端，该样本数据采集装置60包括：第一获取模块601、业务运行模块602、第二获取模块603和发送模块604。
该第一获取模块601，用于获取预先设置的待测试业务的背景信息，该背景信息包括待测试业务的业务类型；
该业务运行模块602，用于运行该待测试业务；
该第二获取模块603，用于获取该待测试业务在运行过程中的特征文件，该特征文件用于标识该待测试业务在运行过程中传输的数据包的特征的文件；
该发送模块604，用于向服务器发送样本数据，该样本数据包括背景信息和特征文件。
综上所述，本发明实施例中提供的样本数据采集装置，通过终端获取预先 设置的待测试业务的背景信息，并运行待测试业务，在待测试业务的运行过程中获取特征文件，向服务器发送包括背景信息和特征文件的样本数据，服务器获取终端发送的样本数据，并确定样本数据中特征文件的存储路径，将背景信息和特征文件的存储路径存储至特征数据库。通过在终端上运行待测试业务，并记录待测试业务的特征文件，在测试完毕后向服务器发送包括背景信息和特征文件的样本数据，终端可以快速地向服务器发送样本数据，服务器也可以快速地获取样本数据，减小了服务器获取样本数据所需要的时间，提高了服务器获取样本数据的效率，降低了服务器获取样本数据所花费的成本。
可选地，该背景信息包括该待测试业务的业务类型；
参见图7，该装置还包括：接收模块605、生成模块606和显示模块607。
该接收模块605，用于接收该服务器发送的预估业务类型，该预估业务类型是该服务器根据该特征文件确定的；
该生成模块606，用于根据该预估业务类型生成测试报告，该测试报告包括：该待测试业务的业务类型和该预估业务类型；
该显示模块607，用于显示该测试报告。
可选地，该发送模块还用于向该服务器发送该背景信息，在该背景信息发送完成后，以预设速率向该服务器发送该特征文件。
可选地，该背景信息包括该终端的地理位置信息、该终端的网络制式、该终端的网际协议IP地址和该终端的型号。
可选地，该特征文件包括时间信息、IP地址、端口信息、协议类型和数据包长度。
图8是本发明实施例提供的一种样本数据采集装置80的结构示意图，如图8所示，该样本数据采集装置80应用于服务器，该样本数据采集装置80包括：第一获取模块801、确定模块802和存储模块803。
该第一获取模块801，用于获取终端的待测试业务的样本数据，该样本数据包括背景信息和特征文件；
该确定模块802，用于确定该特征文件的存储路径；
该存储模块803，用于将该背景信息和该特征文件的存储路径，存储至特征数据库。
综上所述，本发明实施例中提供的样本数据采集装置，通过终端获取预先设置的待测试业务的背景信息，并运行待测试业务，在待测试业务的运行过程 中获取特征文件，向服务器发送包括背景信息和特征文件的样本数据，服务器获取终端发送的样本数据，并确定样本数据中特征文件的存储路径，将背景信息和特征文件的存储路径存储至特征数据库。通过在终端上运行待测试业务，并记录待测试业务的特征文件，在测试完毕后向服务器发送包括背景信息和特征文件的样本数据，终端可以快速地向服务器发送样本数据，服务器也可以快速地获取样本数据，减小了服务器获取样本数据所需要的时间，提高了服务器获取样本数据的效率，降低了服务器获取样本数据所花费的成本。
可选地，参见图9，该装置还包括：调用模块804和发送模块805。
该调用模块804，用于调用类型识别模型，根据该样本数据确定该待测试业务的预估业务类型；
该发送模块805，用于向该终端发送该服务器确定的预估业务类型，以便该终端根据该预估业务类型生成测试报告。
可选地，参见图10，该装置还包括：第二获取模块806、第三获取模块807和训练模块808。
该第二获取模块806，用于每隔预设时长，服务器获取该特征数据库中的至少一个背景信息以及每个背景信息对应的特征文件的存储路径；
该第三获取模块807，用于根据每个背景信息对应的特征文件的存储路径，获取每个背景信息对应的特征文件；
该训练模块808，用于根据每个背景信息以及每个背景信息对应的特征文件，对该类别识别模型进行训练。
图11是本发明实施例提供的一种终端110的结构示意图，该终端110应用于图1所示样本数据采集系统中。参见图11，该终端110包括：接收器1101、发射器1102、存储器1103和处理器1104，该接收器1101、该发射器1102、该存储器1103分别与该处理器1104连接。
存储器1103，用于存放程序和数据，存储器1103可以包括高速缓存存储器(cache)，也可以包括高速RAM存储器，还可以包括非易失性存储器(non-volatile memory)，例如至少一个磁盘存储器。可以理解的是，存储器1103可以为随机存储器(Random-Access Memory，RAM)、磁碟、硬盘、固态硬盘(Solid State Disk，SSD)或者非易失性存储器等各种可以存储程序代码的非短暂性的(non-transitory)机器可读介质。
处理器1104可能是一个中央处理器CPU，或者是特定集成电路ASIC (Application Specific Integrated Circuit)，或者是被配置成实施本发明实施例的一个或多个集成电路。例如，处理对磁盘的输入/输出(Input/Output，I/O)请求，将生产卷中的数据备份到目标卷中等等。从而使代码冲突的提示装置可以实现IO操作、数据备份等功能。在本发明实施例中，处理器1104用于执行程序，具体可以执行下述方法实施例中的相关步骤。
程序可以包括操作系统，文件系统以及其他软件模块。不同的软件模块具有不同的功能。
其中，处理器1104用于获取预先设置的待测试业务的背景信息，该背景信息包括待测试业务的业务类型；
处理器1104还用于运行该待测试业务；
处理器1104还用于获取该待测试业务在运行过程中的特征文件，该特征文件用于标识该待测试业务在运行过程中传输的数据包的特征的文件；
处理器1104还用于通过发射器1102向服务器发送样本数据，该样本数据包括该背景信息和该特征文件。
可选地，处理器1104还用于通过接收器1101接收该服务器发送的预估业务类型，该预估业务类型是该服务器根据该特征文件确定的；
处理器1104还用于根据该预估业务类型生成测试报告，该测试报告包括：该待测试业务的业务类型和该预估业务类型；
处理器1104还用于显示该测试报告。
可选地，处理器1104还用于通过发射器1102向该服务器发送该背景信息；
处理器1104还用于通过发射器1102在该背景信息发送完成后，以预设速率向该服务器发送该特征文件。
图12是本发明实施例提供的一种服务器120的结构示意图，该服务器120应用于图1所示的样本数据采集系统中。参见图12，该服务器120包括：接收器1201、发射器1202、存储器1203和处理器1204，该接收器1201、该发射器1202、该存储器1203分别与该处理器1204连接。
存储器1203，用于存放程序和至少一个终端的待测试业务的样本数据，存储器1203可以包括高速缓存存储器(cache)，也可以包括高速RAM存储器，还可以包括非易失性存储器(non-volatile memory)，例如至少一个磁盘存储器。可以理解的是，存储器1203可以为随机存储器(Random-Access Memory，RAM)、磁碟、硬盘、固态硬盘(Solid State Disk，SSD)或者非易失性存储器 等各种可以存储程序代码的非短暂性的(non-transitory)机器可读介质。
处理器1204可能是一个中央处理器CPU，或者是特定集成电路ASIC(Application Specific Integrated Circuit)，或者是被配置成实施本发明实施例的一个或多个集成电路。例如，处理对磁盘的输入/输出(Input/Output，I/O)请求，将生产卷中的数据备份到目标卷中等等。从而使样本数据采集装置可以实现IO操作、数据备份等功能。在本发明实施例中，处理器1204用于执行程序，具体可以执行下述方法实施例中的相关步骤。
程序可以包括操作系统，文件系统以及其他软件模块。不同的软件模块具有不同的功能。
其中，处理器1204用于通过接收器1201，获取终端的待测试业务的样本数据，该样本数据包括背景信息和特征文件；
处理器1204还用于确定该特征文件的存储路径；
处理器1204还用于通过存储器1203，将该背景信息和该特征文件的存储路径，存储至特征数据库。
可选地，处理器1204还用于调用类型识别模型，根据该样本数据确定该待测试业务的预估业务类型；
处理器1204还用于通过发射器1202，向该终端发送该服务器确定的预估业务类型，以便该终端根据该预估业务类型生成测试报告。
可选地，处理器1204还用于每隔预设时长，服务器获取该特征数据库中的至少一个背景信息以及每个背景信息对应的特征文件的存储路径；
处理器1204还用于根据每个背景信息对应的特征文件的存储路径，获取每个背景信息对应的特征文件；
处理器1204还用于根据每个背景信息以及每个背景信息对应的特征文件，对该类别识别模型进行训练。
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成，也可以通过程序来指令相关的硬件完成，所述的程序可以存储于一种计算机可读存储介质中，上述提到的存储介质可以是只读存储器，磁盘或光盘等。
以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (17)

  1. 一种样本数据采集方法，其特征在于，应用于客户端，所述方法包括：
    获取预先设置的待测试业务的背景信息，所述背景信息包括待测试业务的业务类型；
    运行所述待测试业务；
    获取所述待测试业务在运行过程中的特征文件，所述特征文件用于标识所述待测试业务在运行过程中传输的数据包的特征的文件；
    向服务器发送样本数据，所述样本数据包括所述背景信息和所述特征文件。
  2. 根据权利要求1所述的方法，其特征在于，所述背景信息包括所述待测试业务的业务类型；
    所述向服务器发送样本数据之后，所述方法还包括：
    接收所述服务器发送的预估业务类型，所述预估业务类型是所述服务器根据所述特征文件确定的；
    根据所述预估业务类型生成测试报告，所述测试报告包括：所述待测试业务的业务类型和所述预估业务类型；
    显示所述测试报告。
  3. 根据权利要求1所述的方法，其特征在于，所述向服务器发送样本数据，包括：
    向所述服务器发送所述背景信息；
    在所述背景信息发送完成后，以预设速率向所述服务器发送所述特征文件。
  4. 根据权利要求1所述的方法，其特征在于，所述背景信息包括所述终端的地理位置信息、所述终端的网络制式、所述终端的网际协议IP地址和所述终端的型号。
  5. 根据权利要求1所述的方法，其特征在于，所述特征文件包括时间信息、IP地址、端口信息、协议类型和数据包长度。
  6. 一种样本数据采集方法，其特征在于，用于服务器，所述方法包括：
    获取终端的待测试业务的样本数据，所述样本数据包括背景信息和特征文件；
    确定所述特征文件的存储路径；
    将所述背景信息和所述特征文件的存储路径，存储至特征数据库。
  7. 根据权利要求6所述的方法，其特征在于，在所述将所述背景信息和所述特征文件的存储路径，存储至特征数据库之后，所述方法还包括：
    调用类型识别模型，根据所述样本数据确定所述待测试业务的预估业务类型；
    向所述终端发送所述服务器确定的预估业务类型，以便所述终端根据所述预估业务类型生成测试报告。
  8. 根据权利要求6所述的方法，其特征在于，在所述将所述背景信息和所述特征文件的存储路径，存储至特征数据库之后，所述方法还包括：
    每隔预设时长，服务器获取所述特征数据库中的至少一个背景信息以及每个背景信息对应的特征文件的存储路径；
    根据每个背景信息对应的特征文件的存储路径，获取每个背景信息对应的特征文件；
    根据每个背景信息以及每个背景信息对应的特征文件，对所述类别识别模型进行训练。
  9. 一种样本数据采集装置，其特征在于，所述装置包括：
    第一获取模块，用于获取预先设置的待测试业务的背景信息，所述背景信息包括待测试业务的业务类型；
    业务运行模块，用于运行所述待测试业务；
    第二获取模块，用于获取所述待测试业务在运行过程中的特征文件，所述特征文件用于标识所述待测试业务在运行过程中传输的数据包的特征的文件；
    发送模块，用于向服务器发送样本数据，所述样本数据包括所述背景信息和所述特征文件。
  10. 根据权利要求9所述的装置，其特征在于，所述背景信息包括所述待测试业务的业务类型；
    所述装置还包括：
    接收模块，用于接收所述服务器发送的预估业务类型，所述预估业务类型是所述服务器根据所述特征文件确定的；
    生成模块，用于根据所述预估业务类型生成测试报告，所述测试报告包括：所述待测试业务的业务类型和所述预估业务类型；
    显示模块，用于显示所述测试报告。
  11. 根据权利要求9所述的装置，其特征在于，所述发送模块还用于向所述服务器发送所述背景信息，在所述背景信息发送完成后，以预设速率向所述服务器发送所述特征文件。
  12. 根据权利要求9所述的装置，其特征在于，所述背景信息包括所述终端的地理位置信息、所述终端的网络制式、所述终端的网际协议IP地址和所述终端的型号。
  13. 根据权利要求9所述的装置，其特征在于，所述特征文件包括时间信息、IP地址、端口信息、协议类型和数据包长度。
  14. 一种样本数据采集装置，其特征在于，所述装置包括：
    第一获取模块，用于获取终端的待测试业务的样本数据，所述样本数据包括背景信息和特征文件；
    确定模块，用于确定所述特征文件的存储路径；
    存储模块，用于将所述背景信息和所述特征文件的存储路径，存储至特征数据库。
  15. 根据权利要求14所述的装置，其特征在于，所述装置还包括：
    调用模块，用于调用类型识别模型，根据所述样本数据确定所述待测试业务的预估业务类型；
    发送模块，用于向所述终端发送所述服务器确定的预估业务类型，以便所 述终端根据所述预估业务类型生成测试报告。
  16. 根据权利要求14所述的装置，其特征在于，所述装置还包括：
    第二获取模块，用于每隔预设时长，服务器获取所述特征数据库中的至少一个背景信息以及每个背景信息对应的特征文件的存储路径；
    第三获取模块，用于根据每个背景信息对应的特征文件的存储路径，获取每个背景信息对应的特征文件；
    训练模块，用于根据每个背景信息以及每个背景信息对应的特征文件，对所述类别识别模型进行训练。
  17. 一种样本数据采集系统，其特征在于，所述系统包括：至少一个终端和服务器；
    所述至少一个终端中的每个终端，用于获取预先设置的待测试业务的背景信息，运行所述待测试业务，获取所述待测试业务在运行过程中的特征文件，向服务器发送样本数据，其中，所述背景信息包括待测试业务的业务类型，所述特征文件用于标识所述待测试业务在运行过程中传输的数据包的特征的文件，所述样本数据包括所述背景信息和所述特征文件；
    所述服务器用于获取终端的待测试业务的样本数据，确定所述特征文件的存储路径，将所述背景信息和所述特征文件的存储路径，存储至特征数据库。
PCT/CN2016/092193 2016-07-29 2016-07-29 样本数据采集方法、装置及系统 WO2018018572A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2016/092193 WO2018018572A1 (zh) 2016-07-29 2016-07-29 样本数据采集方法、装置及系统

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2016/092193 WO2018018572A1 (zh) 2016-07-29 2016-07-29 样本数据采集方法、装置及系统

Publications (1)

Publication Number Publication Date
WO2018018572A1 true WO2018018572A1 (zh) 2018-02-01

Family

ID=61015690

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/092193 WO2018018572A1 (zh) 2016-07-29 2016-07-29 样本数据采集方法、装置及系统

Country Status (1)

Country Link
WO (1) WO2018018572A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023036280A1 (zh) * 2021-09-10 2023-03-16 华为技术有限公司 一种模型测试方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102315974A (zh) * 2011-10-17 2012-01-11 北京邮电大学 基于层次化特征分析的tcp、udp流量在线识别方法和装置
CN104486143A (zh) * 2014-12-01 2015-04-01 中国联合网络通信集团有限公司 一种深度报文检测方法、检测系统
US20150172159A1 (en) * 2013-12-17 2015-06-18 Cisco Technology, Inc. Sampled deterministic finite automata for deep packet inspection
CN105657001A (zh) * 2015-12-28 2016-06-08 中国联合网络通信集团有限公司 一种分析通信大数据的方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102315974A (zh) * 2011-10-17 2012-01-11 北京邮电大学 基于层次化特征分析的tcp、udp流量在线识别方法和装置
US20150172159A1 (en) * 2013-12-17 2015-06-18 Cisco Technology, Inc. Sampled deterministic finite automata for deep packet inspection
CN104486143A (zh) * 2014-12-01 2015-04-01 中国联合网络通信集团有限公司 一种深度报文检测方法、检测系统
CN105657001A (zh) * 2015-12-28 2016-06-08 中国联合网络通信集团有限公司 一种分析通信大数据的方法及装置

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023036280A1 (zh) * 2021-09-10 2023-03-16 华为技术有限公司 一种模型测试方法及装置

Similar Documents

Publication Publication Date Title
KR102158557B1 (ko) 응답 시간 확정 방법 및 장치
US9645910B1 (en) Cross platform mobile HTML debugging
KR101722628B1 (ko) 라우터에 기반한 네트워킹 제어 방법, 장치, 라우터, 프로그램 및 저장매체
JP6314252B2 (ja) ネットワークビデオ再生方法及び装置
CN105207832B (zh) 一种服务器压力测试方法和装置
EP3754998B1 (en) Streaming media quality monitoring method and system
US20140101739A1 (en) Semantic challenge for audio human interactive proof
US20210400114A1 (en) Methods and apparatus for census and panel matching using http headers
US20170132112A1 (en) Test method and device for business system
CN106658666B (zh) 一种用于建立无线连接的方法与设备
CN107370628B (zh) 基于埋点的日志处理方法及系统
CN110913038B (zh) Ip地址判定方法、装置、服务器及计算机可读存储介质
CN108733545B (zh) 一种压力测试方法及装置
Pawlaszczyk et al. Alexa, tell me-a forensic examination of the amazon echo dot 3 rd generation
CN111506496A (zh) 一种测试数据获取方法、装置、电子设备及存储介质
CN113676926B (zh) 用户网络感知画像方法及装置
WO2018018572A1 (zh) 样本数据采集方法、装置及系统
CN103885877B (zh) 一种http的模拟浏览器测试脚本生成方法及装置
CN111143650B (zh) 获取页面数据的方法、装置、介质及电子设备
CN110784364B (zh) 一种数据监测方法、装置、存储介质及终端
CN106899843B (zh) 一种视频业务质量评估方法及装置
CN114531345A (zh) 流量比对结果存储方法、装置、设备及存储介质
WO2017185632A1 (zh) 数据传输的方法及电子设备
US11243866B1 (en) Method and system for conducting application quality tests on communication devices
CN112769627B (zh) 网络环境模拟方法、系统及计算机程序产品

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16910153

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 16910153

Country of ref document: EP

Kind code of ref document: A1