Network transmission quality measurement system
Patent Information
- Application Number
- TW114107854
- Authority / Receiving Office
- TW · TW
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-03-04
- Publication Date
- 2026-09-16
- Estimated Expiration
- 2045-03-03
AI Technical Summary
Current network transmission quality measurement methods fail to accurately reflect real-world network performance, especially in dynamic environments, and often interfere with normal traffic, while lacking user experience metrics and protocol compatibility for emerging technologies.
A network application service experience quality calculation system using IoT devices and a Broker server to collect and analyze actual user data, calculating an average opinion score and network load to estimate application service experience quality.
Provides accurate, real-time network quality assessment that aligns with user experience, reducing network interference and supporting diverse protocols, enabling timely issue detection and optimization.
Smart Images
Figure 00000000_0000_ABST
Abstract
Description
[Technical Field]
[0001] This invention relates to a network transmission quality measurement system, and more particularly to a network user experience quality measurement method. [Previous Technology]
[0002] Network transmission quality measurement has become an indispensable part of modern communications, especially in an era of increasing demands for the Internet of Things (IoT), cloud services, streaming media applications, and remote work. However, existing measurement methods and technologies still have many shortcomings, which limit the accurate assessment of network performance and pose challenges to improving user experience. For telecommunications operators, service providers, and end users, overcoming these shortcomings and establishing a more reliable and comprehensive measurement system is a problem that needs to be solved.
[0003] Currently used network transmission quality measurement techniques, such as latency, jitter, and packet loss rate, provide basic network performance indicators. However, the measurement results of these techniques tend to be idealized and fail to highlight problems in real-world applications.
[0004] Insufficient real-time performance is a significant problem. Many measurement methods require periodic data collection and analysis, and this delay can prevent timely detection and resolution of issues in rapidly changing networks. For example, when high traffic causes temporary network congestion, the measurement system may fail to record these transient events in time, thus missing crucial diagnostic opportunities. Furthermore, excessively low measurement frequency can lead to misjudgments of network performance, especially in applications requiring high-frequency monitoring, such as audio / video streaming or instant messaging.
[0005] Another common problem is the increased network load during measurement. Many technologies use active measurement methods, such as injecting test traffic into the network to evaluate its performance. However, this test traffic may interfere with normal business traffic, further affecting the actual operation of the network. At the same time, the traffic from active measurements also increases the network burden, especially under conditions of limited bandwidth or strained network resources, and this additional load may lead to distorted measurement results.
[0006] Furthermore, there is often a disconnect between current measurement metrics and actual user experience. Network performance measurements primarily focus on technical parameters such as latency and packet loss rate, rather than the actual user experience at the application layer. For example, a low-latency network may still cause video playback to stutter or voice calls to drop, reflecting the gap between technical metrics and user experience. This disconnect makes it difficult for service providers to optimize for issues that users truly care about.
[0007] Regarding protocol and technology compatibility, existing measurement methods are still insufficient to fully cover the needs of modern networks. For example, emerging protocols such as HTTP / 3 and QUIC have been widely adopted, but measurement tools for these protocols are still immature. At the same time, low-power protocols in the Internet of Things (IoT) such as MQTT and CoAP have not been adequately studied for performance monitoring, making it difficult to accurately assess the transmission quality of many IoT devices.
[0008] Despite the many shortcomings of existing technologies, some emerging technologies are improving network measurement capabilities. For example, AI-driven network monitoring technology can analyze traffic patterns through machine learning and predict potential network problems in a timely manner. Software-defined networking (SDN) provides a centralized data collection method, allowing administrators to understand the overall network operation in real time. In addition, the development of passive measurement technology reduces network load by analyzing existing service traffic to obtain performance data, avoiding the interference of traditional active measurement on the network. [Summary of the Invention]
[0009] Based on the above objectives, the present invention provides a network application service experience quality calculation system, which obtains the actual transmission values of each real-time network user to estimate the application service experience quality of the network user, comprising: a Broker server, a Client transmission end, a Server transmission end, a network user interface, a database, a first Internet of Things device and a second Internet of Things device.
[0010] As described above, the Broker server is used to store and record detailed historical data of transmission. The Client transmission end is used to collect relevant data on network transmission quality. The Server transmission end is used to aggregate the data transmitted by the Client transmission end and is signal-connected to the Client transmission end. The network user interface is published and set in the Broker server and obtains reports and data lists. The first IoT device is connected to the real-time network user by the Client transmission end to obtain the first actual transmission values of multiple real-time network users, and measures and transmits them back to the Broker server through the Server transmission end. The average opinion score is calculated based on the first actual transmission value. The second IoT device is set in the Broker server and calculates an application service experience quality value based on the first actual transmission value, an algorithm, and the corresponding network load, and stores it in the Broker server.
[0011] Preferably, the first IoT device and the second IoT device are transmitted via a lightweight messaging protocol based on a publish / subscribe model.
[0012] Preferably, the actual transmission value of the real-time network user is the real data when the user is actually using the network.
[0013] Based on the above objectives, according to one embodiment, the present invention further provides a method for estimating the quality of application service experience of network users, which includes the following items: a voice measurement item, a video stream traffic measurement item, a file transfer measurement item, and a text message measurement item.
[0014] As mentioned above, all measurement items use the average opinion score of actual users as the measurement standard. Each measurement item has a different average opinion score calculation method due to its different characteristics. The average opinion score of all items is calculated and weighted by a network usage weight to finally obtain a comprehensive score. The network usage weight will vary depending on the bandwidth occupied by each different measurement item on the network.
[0015] According to one embodiment, each measurement item is scored between 1 and 5 points, with higher numbers representing better quality.
[0016] According to one embodiment, the network usage weight described above is the ratio of the measurement item to all measurement items.
[0017] Based on the above objectives, the present invention further provides a link carrying capacity method, which includes the following items: a voice measurement item, a video stream measurement item, a file transmission measurement item, and a text message measurement item.
[0018] All measurement items use the average opinion score of actual users as the measurement standard. Each measurement item has a different average opinion score calculation method due to its different characteristics. By analyzing the score of each measurement item and the number of connected devices for each score, the capacity carried by this network link is analyzed.
Implementation Method
[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making progressive contributions are within the scope of protection of the present invention.
[0020] Please refer to Figure 1. In Figure 1, a network application service experience quality calculation system 100 obtains the actual transmission values of each real-time network user to estimate the application service experience quality of the network user. It includes: a Broker server 101, a Server transmission end 102, a Client transmission end 103, a database 104, and a network user interface 105, wherein the Client transmission end 103 receives real data from a plurality of Server transmission ends 102.
[0021] As described above, the Broker server 101 is used to store and record detailed historical data of transmission. The Client transmission end 103 collects relevant data on network transmission quality. The Server transmission end 102 aggregates the data from the Client transmission end 103, connects to the Client transmission end 103 by signal, and the user interface is published to the Broker server 101 to provide an interface that allows users to access reports and data lists stored in the Broker server 101.
[0022] As described above, in one embodiment of the network application service experience quality calculation system of the present invention, the architecture of each of the above transmission terminals is implemented through a first Internet of Things device. The Client transmission terminal 103 is connected to a plurality of Server transmission terminals 102 by signal to obtain a first actual transmission value of each real-time network user and transmit it back to the Broker server 101 through the Server transmission terminal 102, and calculate the average opinion score through the first actual transmission value.
[0023] As described above, in one embodiment of the network application service experience quality calculation system of the present invention, a second Internet of Things device is installed in the Broker server 101 to calculate an application service experience quality value based on the first actual transmission value and by a specific algorithm and the corresponding network load, and the value is stored in the Broker server 101.
[0024] As described above, the first IoT device and the second IoT device are devices that use a lightweight messaging protocol based on a publish / subscribe model.
[0025] As mentioned above, the actual transmission value of a real-time network user is the actual data when the user is actually using the network.
[0026] Please refer to Figure 2. In Figure 2, a method for estimating the application service experience quality of network users includes the following steps:
[0027] Step S01: Start the device to be tested, proceed to step S02.
[0028] Step S02: By connecting to the NTP Server and correcting the timing, the test data is published on the Broker server, proceeding to step S03.
[0029] In step S03, a subscriber outputs a command to Broker server 101 to obtain test data. Broker server 101 first transmits the command to server transmission 102 through client transmission terminal 103. Broker server 101 then transmits the test data to client transmission terminal 103 through server transmission terminal 102. Broker server 101 records detailed history when transmitting data between server transmission terminal 102 and client transmission terminal 103. This process continues to step S04.
[0030] Step S04: The above tests will continue to be performed, and tests will be conducted on various different measurement items, including: a voice measurement item, a video stream measurement item, a file transfer measurement item and a text message measurement item. After each test is completed, proceed to step S05.
[0031] Step S05: The above test data is sent back and a report and data list are generated and stored in Broker server 101.
[0032] As mentioned above, all measurement items use the average opinion score of actual users as the measurement standard. Each measurement item has a different average opinion score calculation method due to its different characteristics. The average opinion score of all items is calculated and weighted by a network usage weight to obtain the final comprehensive score. The network usage weight will vary depending on the bandwidth occupied by each different measurement item. The score of each measurement item is between 1 and 5 points. The higher the number, the better the quality. The network usage weight is the ratio of the measurement item to all measurement items.
[0033] Through the above steps, the actual end-to-end network connection quality can be accurately calculated, which can overcome the previous method that network quality parameters could not accurately reflect network quality.
[0034] In one embodiment, a score below 3.5 points will cause a very bad experience for network users, including obvious call delays and intermittent video playback. Therefore, the present invention further sets how many network users can be accommodated if the score must be maintained above 3.5 points, and the link carrying capacity in each network is set under the condition that the network user experience is good.
[0035] Next, please refer to Figure 3. In Figure 3, a link carrying capacity method includes the following steps:
[0036] Step S11: Start the device to be tested, proceed to step S12.
[0037] Step S12: By connecting to the NTP Server and correcting the timing, the test data is published to the Broker server 101, proceeding to step S13.
[0038] In step S13, a subscriber outputs a command to Broker server 101 to obtain test data. Broker server 101 first transmits the command to server transmission 102 through client transmission terminal 103. Broker server 101 then transmits the test data to client transmission terminal 103 through server transmission terminal 102. Broker server 101 records detailed history when transmitting data between server transmission terminal 102 and client transmission terminal 103, up to step S14.
[0039] Step S14: The above tests will continue to be performed, and tests will be conducted on various different measurement items. The measurement items include: a voice measurement item, a video stream measurement item, a file transfer measurement item, and a text message measurement item. By analyzing the score of each measurement item and the number of connected devices for each score, the capacity carried by this network link is analyzed. After each test is completed, proceed to step S15.
[0040] Step S15: The above test data is sent back and a report and data list are generated and stored in Broker server 101.
[0041] Through the above steps, it can be further determined how many users the network can support to maintain a certain quality.
[0042] This invention proposes a method for calculating the quality of network application service experience, used to estimate the quality of application service experience for network users. This method utilizes a first IoT device of the real-time network user to obtain the current actual transmission value through a Client transmission end. This transmission value represents the user's data transmission performance in the current network environment. The Client transmission end also includes a network user interface, through which it connects to a Broker server. Users can query reports and data lists, while the server is responsible for storing and managing historical transmission data.
[0043] The actual transmission value is sent back to the Broker server via the Server transmission end connected to the Client transmission end. The server uses a built-in algorithm to calculate the Mean Opinion Score (MOS) based on the transmission value. This score reflects the user's subjective experience of the network service. In addition, a second IoT device installed in the Broker server further calculates the application service experience quality value based on the actual transmission value, combined with the algorithm and the corresponding network load, and stores the result on the server for subsequent querying.
[0044] This method achieves accurate estimation of the quality of network application service experience through multi-layer data transmission and precise calculation, providing a reliable performance evaluation benchmark for users and service providers.
[0045] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents. [Simplified Explanation of the Diagram]
[0046] Figure 1 is a block diagram of one embodiment of the network application service experience quality calculation system of the present invention.
[0047] Figure 2 is a flowchart of a method for estimating the application service experience quality of network users according to the present invention.
[0048] Figure 3 is a flowchart of a link carrying capacity method according to the present invention.
Claims
1. A network application service experience quality calculation system, which obtains the actual transmission values of each real-time network user to estimate the application service experience quality of the network user, comprising: a client transmission end for collecting relevant data on network transmission quality; a server transmission end for aggregating the data transmitted by the client transmission end, and signal-connected to the client transmission end; a broker server for storing and recording historical transmission data in detail, and signal-connected to the client transmission end and the server transmission end; wherein the client transmission end and the server transmission end are interchangeable, and neither end is limited to being measured; a network user interface, established in the client transmission end, for connecting to the server transmission end and obtaining reports and data lists; A first Internet of Things (IoT) device connects to the real-time network user via the Client transmission terminal to obtain a first actual transmission value from each of the real-time network users and transmits it back to the Broker server via the Server transmission terminal, and calculates an average opinion score based on the first actual transmission value; and a second IoT device is installed in the Broker server, which calculates an application service experience quality value based on the first actual transmission value and by an algorithm and corresponding network load, and stores it in the Broker server.
2. As in request item 1, the network application service experience quality calculation system, wherein, The first IoT device and the second IoT device are devices that use a message transmission protocol based on a publish / subscribe model.
3. As in Request 1, the network application service experience quality calculation system, wherein the actual transmission value of the real-time network user is the real data when the user is actually using the network, and the network protocol used to measure the transmission value is different from the protocol used by the user when actually using the network.
4. The network application service experience quality calculation system of request item 1 further includes a method for estimating the application service experience quality of network users, including: a voice measurement item; a video streaming measurement item; a file transfer measurement item; and a text messaging measurement item; wherein all measurement items use the average opinion score of actual users as the measurement standard, and each measurement item has a different average opinion score calculation method due to different characteristics. The average opinion scores of all items are calculated and weighted by a network usage weight to obtain a comprehensive score, wherein the network usage weight will vary depending on the bandwidth occupied by each different measurement item on the network.
5. As in request item 4, the method for estimating the quality of application service experience for network users, where each measurement item is scored between 1 and 5 points, with higher numbers representing better quality.
6. As in Request 4, the method for estimating the application service experience quality of network users, wherein the weight of network usage is the ratio of this measurement item to all measurement items.
7. The network application service experience quality calculation system of request item 1 further includes a link carrying capacity method, including: a voice measurement item; a video streaming measurement item; a file transfer measurement item; and a text message measurement item; wherein all measurement items use the average opinion score of actual users as the measurement standard, and each measurement item has a different average opinion score calculation method due to different characteristics, and the network link carrying capacity is analyzed by analyzing the score of each measurement item and the number of connected devices for each score.
8. A method for calculating the quality of service experience of a network application, used to estimate the quality of service experience of a network user, comprising the following steps: a first Internet of Things (IoT) device of a real-time network user obtains the actual transmission value of a real-time network user through a Client transmission terminal; a network user interface is established in the Client transmission terminal to connect to a Broker server and obtain reports and data lists, wherein the Broker server stores and records the historical data of the transmission in detail; the actual transmission value is transmitted back to the Broker server through a Server transmission terminal connected to the Client transmission terminal, and the Broker server calculates the average opinion score based on the actual transmission value; and a second IoT device installed on the Broker server calculates an application service experience quality value based on the actual transmission value and by an algorithm and corresponding network load, and stores it on the Broker server.