Heterogeneous network-oriented ad hoc network radio station intelligent test system and method
By combining environmental perception units and intelligent test terminal modules with reinforcement learning models, the parameters of ad hoc radios and 5G base stations are dynamically adjusted, solving the problems of low efficiency and poor adaptability in heterogeneous network testing, and achieving efficient adaptive testing in extreme environments.
Patent Information
- Application Number
- CN202511878691.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-12
- Publication Date
- 2026-03-06
AI Technical Summary
Existing self-organizing network radio testing systems cannot adapt to heterogeneous network environments, resulting in low testing efficiency, poor scenario adaptability, and difficulty in covering extreme environmental scenarios.
By employing an environmental sensing unit and an intelligent test terminal module, combined with a reinforcement learning model, the operating parameters of the ad hoc network radio module and the 5G base station module are dynamically adjusted to achieve adaptive testing.
It improves the environmental adaptability and testing efficiency of heterogeneous network testing, and enables efficient and accurate evaluation and monitoring under different environmental conditions.
Smart Images
Figure CN121619604A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of testing technology for self-organizing network communication equipment, and in particular to an intelligent testing system and method for self-organizing network radios in heterogeneous networks. Background Technology
[0002] With the widespread application of Ad Hoc Networks in scenarios such as emergency rescue, tactical communication, and IoT edge coverage, the reliability, environmental adaptability, and cross-network collaboration capabilities of their communication equipment have become crucial to ensuring mission success.
[0003] Currently, existing testing systems for ad hoc network radios mainly focus on closed "pure ad hoc network" environments. This involves setting up several ad hoc network nodes in a laboratory or field, and collecting and evaluating performance metrics such as throughput, latency, and packet delivery rate under manually configured topology and fixed service models. This approach has the following significant drawbacks:
[0004] On the one hand, the test objects are limited to single-hop or multi-hop communication between nodes within the ad hoc network, and do not involve the interconnection and interoperability verification between the ad hoc network and external 5G cellular base stations. Real-world 5G deployments are all heterogeneous networks, including ad hoc network nodes and 5G base stations. Ad hoc networks often need to have links for both internal node communication and 5G base station communication.
[0005] On the other hand, the test environment adaptability is poor. Existing self-organizing network performance tests are mostly in the "one scenario, one structure" mode, that is, the network topology, protocol parameters and hardware platform need to be designed separately for specific application scenarios. During the test, the test strategy cannot be dynamically adjusted according to the real-time wireless environment (such as different temperatures, humidity, air pressure, etc.), resulting in low test efficiency, poor repeatability, and difficulty in covering extreme environmental scenarios. Summary of the Invention
[0006] Based on the above analysis, the embodiments of the present invention aim to provide an intelligent testing system and method for self-organizing radios in heterogeneous networks, in order to solve the problem of poor adaptability to the testing environment of heterogeneous self-organizing networks in the prior art.
[0007] The objective of this invention is mainly achieved through the following technical solutions:
[0008] On one hand, embodiments of the present invention provide an intelligent testing system for self-organizing network radios in heterogeneous networks, comprising:
[0009] An environmental sensing unit is used to collect environmental data.
[0010] The heterogeneous network unit includes a first self-organizing network radio module, a second self-organizing network radio module, and a 5G base station module, which are used to configure the initial operating parameters of each module and build a heterogeneous network environment.
[0011] The intelligent test terminal module is used to send test data packets generated based on the environmental data to the first self-organizing network radio module. Using the first self-organizing network radio module as the data sending end, it performs self-organizing network performance testing by capturing test data from the second self-organizing network radio module or 5G base station module of the heterogeneous network link to obtain performance index values. It is also used to adjust the working parameters of each module according to the test performance index values to obtain the performance extreme values of the heterogeneous network unit in the current environment.
[0012] Furthermore, the performance extremes of the heterogeneous network units in the current environment are obtained through the following process:
[0013] Performance metric values are obtained based on the captured test data;
[0014] Based on the environmental data, performance indicators, and operating parameters of each module, the operating parameters of each module are adjusted using the trained reinforcement learning model to obtain the performance extreme values of the heterogeneous network units in the current environment.
[0015] Furthermore, the operating parameters of each module include the operating frequency band and transmit power of the first and second self-organizing network radio modules, and the link bandwidth of the 5G base station module; adjusting the operating parameters of each module includes:
[0016] Each module in the link is treated as an intelligent agent, and the environmental data and test performance indicators are used as state observations; wherein, the performance indicators include transmission latency and throughput;
[0017] Using the adjustable operating parameters of the self-organizing network radio module and the 5G base station module as the action space, the action selection of the operating parameters is performed with the goal of maximizing the reward based on the performance index setting, so as to obtain the operating frequency band, transmission power and link bandwidth of the 5G base station module.
[0018] Furthermore, the rewards include positive rewards based on throughput, negative rewards based on transmission latency, and fixed penalties for actions that exceed the range of operating parameters and for sudden high latency.
[0019] Furthermore, at least one set each of the second self-organizing network radio module and the 5G base station module is provided; the heterogeneous network link includes:
[0020] First link: The first self-organizing network radio module wirelessly sends test data packets to any second self-organizing network radio module, with that self-organizing network radio module as the receiving end;
[0021] Second link: The first self-organizing network radio module uploads test data packets wirelessly to any 5G base station module, with the 5G base station module serving as the receiving end.
[0022] Furthermore, the test data packet includes a transmission latency test packet and a data transmission rate test packet; wherein,
[0023] The transmission delay is obtained by calculating the difference between the sending and receiving timestamps of the transmission delay test packet; the throughput is obtained by counting the number of data packets successfully received within a preset time and multiplying them by the data packet size.
[0024] Furthermore, the intelligent testing terminal module adaptively acquires the testing scheme and generates the corresponding test data package based on the environmental data and a pre-set testing scheme; wherein, the testing scheme is divided into low temperature, low pressure and high humidity scheme, conventional environment scheme and high temperature, high pressure and low humidity scheme according to the range of environmental data; different testing schemes are matched with corresponding test data packages.
[0025] Furthermore, the environmental data includes ambient temperature, humidity, and air pressure.
[0026] Furthermore, the intelligent test terminal module is connected to the first self-organizing network radio module via a wired connection.
[0027] On the other hand, embodiments of the present invention provide a smart testing method for self-organizing network radios in heterogeneous networks, comprising the following steps:
[0028] Collect environmental data;
[0029] Configure the operating parameters of each module in the heterogeneous network and initialize the heterogeneous network link environment; each module includes a first self-organizing network radio module, a second self-organizing network radio module, and a 5G base station module;
[0030] The test data packet generated based on the environmental data is sent to the first self-organizing network radio module. The self-organizing network performance is tested by capturing the test data of the second self-organizing network radio module or 5G base station module of the heterogeneous network link, with the first self-organizing network radio module as the data sending end.
[0031] Based on the performance metrics tested, the operating parameters of each module are adjusted to obtain the performance extreme values of the heterogeneous network unit under the current environment.
[0032] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:
[0033] 1. The intelligent testing system for self-organizing network radios proposed in this invention is a system that uses an intelligent testing terminal combined with an environmental sensing unit to intelligently test the performance of heterogeneous networks. It not only involves communication between self-organizing network modules but also the linkage testing between self-organizing networks and 5G base stations. At the same time, it solves the problems of low testing efficiency, poor scenario adaptability, and insufficient environmental adaptability caused by the need to build specific network architectures for specific scenarios in traditional performance testing, and improves the comprehensiveness of testing and environmental adaptability.
[0034] 2. By using a reinforcement learning-based testing process and taking the environment as an observation, the working parameters of each module in the heterogeneous network are adaptively adjusted to quickly obtain the extreme values of test performance, thereby improving testing efficiency and realizing automated, efficient and accurate evaluation and monitoring of the performance of ad hoc network radios under different environmental conditions and harsh scenarios.
[0035] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings. Attached Figure Description
[0036] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.
[0037] Figure 1 This is a schematic diagram of the structure of an intelligent test system for self-organizing radios in heterogeneous networks according to an embodiment of the present invention.
[0038] Figure 2 This is a flowchart of an intelligent testing method for self-organizing network radios in heterogeneous networks, according to an embodiment of the present invention. Detailed Implementation
[0039] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.
[0040] Example 1
[0041] A specific embodiment of the present invention discloses an intelligent testing system for self-organizing network radios in heterogeneous networks, such as... Figure 1 As shown, it includes:
[0042] An environmental sensing unit is used to collect environmental data.
[0043] The heterogeneous network unit includes a first self-organizing network radio module, a second self-organizing network radio module, and a 5G base station module, which are used to configure the initial operating parameters of each module and build a heterogeneous network environment.
[0044] The intelligent test terminal module is used to send test data packets generated based on the environmental data to the first self-organizing network radio module. Using the first self-organizing network radio module as the data sending end, it performs self-organizing network performance testing by capturing test data from the second self-organizing network radio module or 5G base station module of the heterogeneous network link to obtain performance index values. It is also used to adjust the working parameters of each module according to the test performance index values to obtain the performance extreme values of the heterogeneous network unit in the current environment.
[0045] Through the above system, the working parameters of each module in the heterogeneous network are adaptively adjusted for different environments, so as to realize intelligent testing that adapts to the environment. This solves the problems of low testing efficiency, poor scenario adaptability and insufficient environmental adaptability caused by traditional performance testing systems not considering the impact of environmental parameters and needing to build specific network architectures for specific scenarios.
[0046] Specifically, the testing process is as follows:
[0047] Step 1: Build a heterogeneous network test environment and deploy intelligent test terminal module, 5G base station module, (first and second) self-organizing network radio module and environmental sensing unit.
[0048] The intelligent test terminal module, as the core control unit of the entire test system, is equipped with a multi-port communication module, a high-performance computing chip, and a sensor data receiving interface. It supports protocol adaptation, data interaction, and sensor parameter parsing with other modules. Its core functions include environmental parameter acquisition, adaptive generation of test plans, test task scheduling, parameter distribution, data capture, AI analysis, and report output.
[0049] The 5G base station module, deployed in the ground communication coverage area, shall be no less than one set. It adopts a combination of macro base stations and micro base stations to simulate the ground cellular network communication scenario. It is connected to the self-organizing network radio module through a wireless link, and the link bandwidth can be dynamically adjusted according to the test plan.
[0050] The ad hoc network radio module serves as both a test subject and a communication node. The deployment quantity can be flexibly adjusted, but no less than two sets. The ad hoc network radio module used as the test subject is defined as the first ad hoc network radio module, and the other ad hoc network radio modules, serving as communication nodes, are defined as the second ad hoc network radio modules. Each radio module supports frequency hopping communication and adaptive modulation / demodulation technology, and also supports cross-network communication with 5G base station modules. Each radio module has a built-in positioning module and status monitoring unit, which can provide real-time feedback of its own operating parameters.
[0051] The environmental sensing unit consists of a high-precision temperature sensor, a barometric pressure sensor, and a humidity sensor. It is used to collect temperature, barometric pressure, and humidity data of the test environment and upload them to the intelligent test terminal module in real time.
[0052] It should be noted that the intelligent test terminal module is interconnected with the first self-organizing network radio module via a wired connection; the sensors of the environmental perception unit are interconnected with the intelligent test terminal module via a wired communication interface and deployed in key areas of the test environment (close to the self-organizing network radio module and the core path of the communication link) to ensure the authenticity and representativeness of the collected data.
[0053] Step 2: Initialize the heterogeneous network and environment sensing unit.
[0054] For example, after the intelligent test terminal module is started, it first completes the initialization of the environmental sensing unit, calibrates the zero point and range of each sensor to ensure the accuracy of data acquisition; then it issues instructions to set the initial operating parameters of the first and second self-organizing network radio modules (operating parameters include operating frequency band or frequency hopping pattern, transmission power, etc.), and synchronously initializes the 5G base station module (including link bandwidth) to complete the network connection establishment between each network module; during the initialization process, the environmental sensing unit synchronously starts to collect temperature, air pressure and humidity data, and feeds them back to the intelligent test terminal module in real time.
[0055] Step 3: The intelligent test terminal module receives real-time temperature, air pressure, and humidity values uploaded by the environmental sensing unit, combines them with the preset environmental test scheme mapping rules, adaptively generates a test scheme, and sends it to the first self-organizing network radio module.
[0056] For example, the test scheme is divided into a low-temperature, low-pressure, high-humidity scheme, a normal environment scheme, and a high-temperature, high-pressure, low-humidity scheme based on the range of environmental data. Different test schemes are matched with corresponding test data packets (with different sending intervals). The mapping rules of the test schemes are shown in Table 1. Based on the test schemes, corresponding test data packets (including transmission delay tests, data transmission rate tests, and anti-interference capability tests) are generated and sent to the first self-organizing network radio module in the order set by the scheme. Among them, the data packet content for the transmission delay test is a standard Ping packet, the data packet content for the data transmission rate test is a simulated iperf traffic packet, and the anti-interference capability test measures the radio's delay performance and transmission rate performance when subjected to interference. By combining environmental sensors to dynamically adapt the test scheme and automatically generate test data packets, this architecture can be used universally in various harsh environments.
[0057] Table 1
[0058]
[0059] Step 4: Using the first self-organizing network radio module as the data transmitter, perform self-organizing network performance testing by capturing test data from the second radio module or 5G base station module in the heterogeneous network link to obtain performance index values. The specific process is as follows:
[0060] S41. Obtain performance index values based on the captured test data; the performance index values include transmission latency and throughput;
[0061] For example, the transmission delay is obtained based on the difference between the sending timestamp and the receiving timestamp of the transmission delay test packet; the number of data packets successfully received within a preset time (30 seconds or 1 minute) is counted, multiplied by the data packet size to obtain the throughput, and the throughput is processed for a preset time to obtain the data transmission rate; the anti-interference capability test is to test the above two performance indicators under interference conditions.
[0062] It should be noted that the heterogeneous network link includes two test links: a first link where the first self-organizing network radio module wirelessly transmits test data packets to any second self-organizing network radio module, with that self-organizing network radio module acting as the receiving end; and a second link where the first self-organizing network radio module wirelessly uploads test data packets to any 5G base station module, with that 5G base station module acting as the receiving end. The intelligent test terminal module is connected to both the first self-organizing network radio module and the receiving end of the heterogeneous network via wired connections. Comprehensive performance testing of the heterogeneous network is achieved through these two-layer links: between the first and second self-organizing network radio modules, or between the first self-organizing network radio module and the 5G base station module.
[0063] S42. Based on environmental data, performance indicators, and the working parameters of each module, the working parameters of each module are adjusted using the trained reinforcement learning model to obtain the performance extreme values of the heterogeneous network units in the current environment.
[0064] To achieve the aforementioned adjustment of operating parameters, a reinforcement learning-based operating parameter allocation network was designed. In the test environment at each time step, the agent selects an operating parameter, receives a reward, and the environment transitions to the next state. To obtain the performance extrema of heterogeneous network units under different environments, the action selection for operating parameters aims to maximize throughput and minimize latency.
[0065] Specifically, each module in the link is treated as an agent. For each agent, temperature, humidity, air pressure, and performance indicators tested under the current environment, obtained through periodic sampling, are used as state observations. The adjustable operating parameters of the first and second self-organizing network radio modules and the 5G base station module are used as the action space. These operating parameters include the operating frequency band and transmit power of the self-organizing network radio module and the link bandwidth of the 5G base station module. The process of training the network to allocate these operating parameters includes:
[0066] The agent inputs its current state observations into the Q-value network to obtain the agent's actions;
[0067] The Q-value network evaluates the current state action based on the input state observation, the obtained action, and the pre-set reward of the environment, and obtains the Q-value function evaluation value;
[0068] Using the accumulated instantaneous reward expectation as the optimal evaluation value, the Q-value network parameters are trained to approximate the optimal evaluation value, and the trained Q-value network is the working parameter allocation network.
[0069] For example, the instantaneous reward includes a positive reward based on throughput, a negative reward based on transmission latency, and fixed penalties for actions exceeding the operating parameter range and sudden high latency. A sliding window is used to normalize the acquired performance metrics to eliminate hardware differences. For sudden high latency, a negative reward is set when the instantaneous latency exceeds the latency increment threshold of the sliding window's maximum value, by pre-setting a latency increment threshold, to suppress occasional high latency spikes.
[0070] The optimal evaluation value based on the above reward mechanism is expressed as:
[0071]
[0072] Among them, Q * (s,a) represents the optimal evaluation value obtained by executing action space a from state s. t Let γ be the immediate reward obtained by performing action a in state s, γ be the future reward discount factor, and γ∈(0,1), s′ be the new state after performing the action, and a′ be the new action space to be performed in the new state s′.
[0073] The process of adjusting the working parameters of each module using the trained working parameter allocation network includes: inputting the current state observation into the trained working parameter allocation network to obtain the state action function Q value, and selecting the action with the largest Q value as the working parameter of the agent under the current observation.
[0074] The above-mentioned intelligent testing scheme based on reinforcement learning improves the environmental adaptability of heterogeneous network testing. In the event of temperature drift, humidity loss, or air pressure frequency deviation, the system can automatically adapt to the test parameters without requiring on-site frequency scanning or manual gain adjustment by the test personnel, thereby improving testing efficiency and reducing maintenance costs.
[0075] Furthermore, the test data undergoes comprehensive analysis. This includes calculating specific values for core indicators such as latency, throughput, and anti-interference capability. It also considers the characteristics of the heterogeneous network scenario and the test environment to rate the overall performance of heterogeneous network units (e.g., excellent / good / qualified / unqualified), generating a performance report. Simultaneously, complete data from each test is automatically stored, providing data support for predicting performance degradation trends of heterogeneous networks under different scenarios and usage durations based on historical test data (e.g., the potential decrease in throughput under long-term high-temperature environments).
[0076] Compared with existing technologies, this embodiment provides an intelligent testing system for self-organizing network radios in heterogeneous networks. By combining the results of intelligent agent testing terminals with environmental perception units and adopting an intelligent testing method based on reinforcement learning, it solves the problems of low testing efficiency, poor scenario adaptability, and insufficient environmental adaptability caused by the need to build specific network architectures for specific scenarios in traditional performance testing. Ultimately, it realizes heterogeneous network element testing under different temperature, humidity, and pressure combinations, improves the environmental adaptability of testing, and improves testing efficiency.
[0077] Example 2
[0078] Another specific embodiment of the present invention discloses an intelligent testing method for self-organizing network radios in heterogeneous networks, such as... Figure 2 As shown, it includes the following steps:
[0079] Step S1: Collect environmental data;
[0080] Step S2: Configure the operating parameters of each module in the heterogeneous network and initialize the heterogeneous network link environment; the modules include a first self-organizing network radio module, a second self-organizing network radio module, and a 5G base station module;
[0081] Step S3: Send the test data packet generated based on the environmental data to the first self-organizing network radio module. Using the first self-organizing network radio module as the data sending end, perform self-organizing network performance testing by capturing test data from the second self-organizing network radio module or 5G base station module of the heterogeneous network link.
[0082] Step S4: Adjust the operating parameters of each module according to the test performance indicators to obtain the performance extreme value of the heterogeneous network unit in the current environment.
[0083] The method described herein can be applied to the system described in any of the schemes in Embodiment 1 to perform self-organizing network testing of heterogeneous networks. Related aspects can be referenced from each other, and are not repeated in this embodiment.
[0084] Compared with existing technologies, this embodiment provides an intelligent testing method for self-organizing network radios in heterogeneous networks. By building a heterogeneous network testing environment integrating temperature, air pressure, and humidity sensors, and adopting a method of adaptively setting test schemes based on sensor values and combining AI algorithms for collaborative work, it achieves automated, efficient, and accurate evaluation and monitoring of self-organizing network radio performance under different environmental conditions and harsh scenarios.
[0085] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.
[0086] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.
Claims
1. A heterogeneous network oriented self-organizing network station intelligent test system, characterized in that, include: An environmental sensing unit is used to collect environmental data. The heterogeneous network unit includes a first self-organizing network radio module, a second self-organizing network radio module, and a 5G base station module, which are used to configure the initial operating parameters of each module and build a heterogeneous network environment. The intelligent test terminal module is used to send the test data packet generated based on the environmental data to the first self-organizing network radio module. Using the first self-organizing network radio module as the data sending end, the self-organizing network performance is tested by capturing the test data of the second self-organizing network radio module or 5G base station module of the heterogeneous network link, and the performance index value is obtained. It is also used to adjust the operating parameters of each module based on the performance index values of the test to obtain the performance extreme values of the heterogeneous network unit in the current environment.
2. The system of claim 1, wherein, The performance extremes of heterogeneous network units in the current environment are obtained through the following process: Performance metric values are obtained based on the captured test data; Based on the environmental data, performance indicators, and operating parameters of each module, the operating parameters of each module are adjusted using the trained reinforcement learning model to obtain the performance extreme values of the heterogeneous network units in the current environment.
3. The system of claim 2, wherein, The operating parameters of each module include the operating frequency band and transmit power of the first and second self-organizing network radio modules, and the link bandwidth of the 5G base station module; adjusting the operating parameters of each module includes: Each module in the link is treated as an intelligent agent, and the environmental data and test performance indicators are used as state observations; wherein, the performance indicators include transmission latency and throughput; Using the adjustable operating parameters of the self-organizing network radio module and the 5G base station module as the action space, the action selection of the operating parameters is performed with the goal of maximizing the reward based on the performance index setting, so as to obtain the operating frequency band, transmission power and link bandwidth of the 5G base station module.
4. The system of claim 3, wherein, The rewards include positive rewards based on throughput, negative rewards based on transmission latency, and fixed penalties for actions that exceed the operating parameter range and for sudden high latency.
5. The system according to any of claims 1-4, characterized in that, At least one set each of the second self-organizing network radio module and the 5G base station module; the heterogeneous network link includes: First link: The first self-organizing network radio module wirelessly sends test data packets to any second self-organizing network radio module, with that self-organizing network radio module as the receiving end; Second link: The first self-organizing network radio module uploads test data packets wirelessly to any 5G base station module, with the 5G base station module serving as the receiving end.
6. The system of claim 3, wherein, The test data packets include transmission latency test packets and data transmission rate test packets; wherein... The transmission delay is obtained by calculating the difference between the sending and receiving timestamps of the transmission delay test packet; the throughput is obtained by counting the number of data packets successfully received within a preset time and multiplying them by the data packet size.
7. The system of claim 4, wherein, The intelligent testing terminal module adaptively acquires the test plan and generates the corresponding test data package based on the environmental data and the preset test plan. The test plan is divided into low temperature, low pressure and high humidity, normal environment and high temperature, high pressure and low humidity according to the range of environmental data. Different test plans are matched with corresponding test data packages.
8. The system of claim 1, wherein, The environmental data includes ambient temperature, humidity, and air pressure.
9. The system of claim 1, wherein, The intelligent test terminal module is connected with the first ad hoc radio module through a wired mode.
10. A method for intelligent testing of a heterogeneous network-oriented ad hoc station, characterized by, The method comprises the following steps: Collecting environmental data; Configuring working parameters of each module of the heterogeneous network and initializing a heterogeneous network link environment; the modules include a first ad hoc radio module, a second ad hoc radio module and a 5G base station module; Distributing test data packets generated according to the environmental data to the first ad hoc radio module, taking the first ad hoc radio module as a data sending end, and performing ad hoc network performance testing through test data of the second ad hoc radio module or the 5G base station module of the heterogeneous network link; According to the tested performance index, adjusting the working parameters of each module to obtain the performance extreme value of the heterogeneous network unit under the current environment.