Method and system for testing wireless communication performance of 5G terminal

Abstract

The invention relates to the technical field of 5G testing, and discloses a 5G terminal wireless communication performance testing method and system, and the method comprises the steps: initializing a target terminal, and generating matching data for training to obtain a correlation model; based on the correlation model, determining a shielding antenna of the target terminal in the target time period; in the target time period, shielding a non-shielding antenna of the target terminal, collecting communication performance index data of the target terminal, and calculating to obtain a communication performance evaluation value of the target terminal; by copying an extreme shielding scene possibly occurring in the actual use of the terminal, key communication performance indexes such as throughput rate, time delay and bit error rate in a non-ideal working state can be really collected, and a communication performance evaluation value is obtained to reveal the hiding problem caused by self-adaptive adjustment of the terminal in a traditional test method.

Description

Technical Field

[0001] The present invention relates to the technical field of 5G testing, and more specifically, to a method and system for testing the wireless communication performance of a 5G terminal. Background Art

[0002] In the actual application of 5G terminals, due to the dynamic change of the user's holding posture with the application scenario, the antenna may be blocked in real time. For example, the vertical screen single-hand holding during a video call may block the bottom antenna, and the two-hand holding during a horizontal screen game may block the two side antennas, thus directly affecting the stability of wireless communication. Existing terminals use system-level AI to build an association model of "application program and antenna available state", predict the antenna occlusion situation by analyzing features such as application resource occupancy and terminal posture, and accordingly switch the working state of each antenna in advance. However, traditional testing methods only cover typical scenarios, and it is easy for the terminal AI to identify the test mode using training data and actively adapt, resulting in a phenomenon of overestimated evaluation results. At the same time, due to the lack of verification of extreme scenarios where the model prediction completely deviates from the actual occlusion situation, the antenna strategy of the terminal may fail and the performance may drop suddenly in a real complex environment. Existing testing methods are difficult to effectively evaluate the robustness of the system and urgently need to be improved. Summary of the Invention

[0003] The present invention provides a method and system for testing the wireless communication performance of a 5G terminal, which solves the technical problems proposed in the background art.

[0004] The present invention provides a method for testing the wireless communication performance of a 5G terminal, including: Step 1, initialize the target terminal and generate matching data to train an association model; Step 2, based on the association model, determine the occluded antennas of the target terminal within the target time period; Step 3, within the target time period, shield the non-occluded antennas of the target terminal, collect the communication performance index data of the target terminal, and calculate the communication performance evaluation value of the target terminal.

[0005] Further, generating the matching data includes: Perform real number coding on M antennas of the target terminal to obtain M antenna codes; Allocate N occlusion sequences to N application programs of the target terminal respectively, and each occlusion sequence includes m non-repeating antenna codes; where 1 < m < M, m is a positive integer, and the matching error between any two occlusion sequences is greater than the first error threshold.

[0006] Further, training to obtain the association model includes: Within the first preset time period, establish an occlusion unit and a program control unit; Step 11, start the nth application of the target terminal based on the program control unit; where 1 ≤ n ≤ N, and n is a positive integer; Step 12, in response to the startup instruction of the nth application, obtain m antenna codes in the occlusion sequence of the nth application, and shield the corresponding m antennas through the occlusion unit; Step 13, repeatedly execute Step 11 and Step 12 until: During the test time period, obtain the status information of the M antennas of the target terminal respectively; where the status information includes: working status and standby status; Based on the program control unit, sequentially traverse the N applications, and respectively obtain the status information of the M antennas in response to the startup instruction of each application; If the status information is the working status, mark the corresponding antenna as 1, otherwise mark the corresponding antenna as 0, and extract the antennas marked as 1; Determine the set of antennas marked as 1 in response to the startup instruction of the nth application, and calculate the matching error between the antenna set and the occlusion sequence corresponding to the nth application; If the matching errors of the N applications are all less than the second error threshold, the association model converges, stop executing Step 11 and Step 12, otherwise repeatedly execute Step 11 and Step 12.

[0007] Further, the matching error includes the first matching error between the occlusion sequences and the second matching error between the antenna set and the occlusion sequence; The first matching error includes: Associate the same antennas in the occlusion sequence with the occlusion sequence; If the association is successful, respectively remove the corresponding antennas in the occlusion sequence and the occlusion sequence; Obtain the occlusion sequence and the occlusion sequence after the removal is completed, calculate the sum value of the number of remaining units in the occlusion sequence and the occlusion sequence, and use the sum value as the matching error between the occlusion sequence and the occlusion sequence; The second matching error includes: Associate the same antennas in the antenna set with the occlusion sequence; If the association is successful, respectively remove the corresponding antennas in the antenna set and the occlusion sequence; Obtain the antenna set and the occlusion sequence after the removal is completed, calculate the sum value of the number of remaining units in the antenna set and the occlusion sequence, and use the sum value as the matching error between the antenna set and the occlusion sequence.

[0008] Further, determining the occluded antennas of the target terminal within the target time period based on the association model includes: Within the target time period, start the nth application of the target terminal based on the program control unit; Based on the association model, determine that the antenna in the occlusion sequence corresponding to the nth application is the occluded antenna of the target terminal during the target time period. Furthermore, shielding the non-occluded antennas of the target terminal includes: In response to the start instruction of the nth application, the non-occluded antennas among the M occluded antennas of the target terminal are shielded by the occlusion unit.

[0009] Furthermore, the communication performance metric data includes: During the target time period, establish a stable communication connection between the target terminal and the target base station; In response to the start instruction of the nth application, the target terminal sends a data packet to the target base station, and collects the communication performance metric data of the target terminal, including: The number of bits of data sent by the target terminal, the number of bits of data received by the target base station, the number of error bits of data received by the target base station, the timestamp when the target terminal starts to send data, and the timestamp when the target base station finishes receiving data.

[0010] Furthermore, calculate the target communication performance evaluation value of the target terminal, including: based on the communication performance metric data, calculate the throughput, latency, and bit error rate of the target terminal, and perform weighted fusion on the throughput, latency, and bit error rate to obtain the target communication performance evaluation value of the target terminal.

[0011] In a second aspect, a test system for the wireless communication performance of a 5G terminal is applied to the test method for the wireless communication performance of a 5G terminal as described above, and includes: A training module for initializing the target terminal and generating matching data to train and obtain an association model; A prediction module for determining the occluded antennas of the target terminal during the target time period based on the association model; A test module for shielding the non-occluded antennas of the target terminal during the target time period, collecting the communication performance metric data of the target terminal, and calculating the communication performance evaluation value of the target terminal.

[0012] The beneficial effects of the present invention are as follows: By using matching data to induce the 5G terminal to autonomously construct an association model, and then quickly identify the standby antennas and working antennas of the terminal based on the application start instruction, and artificially shield the working antennas during the target time period, so as to simulate the communication redundancy performance of the terminal in the case of AI prediction failure. In this way, not only can extreme occlusion scenarios that may occur in the actual use of the terminal be replicated under experimental conditions, but also key communication performance metrics such as throughput, latency, and bit error rate in non-ideal working states can be truly collected, and the communication performance evaluation value can be obtained to reveal the hidden problems caused by the terminal's adaptive adjustment in traditional test methods. Description of the Drawings

[0013] Figure 1 is a flowchart of a method for testing the wireless communication performance of a 5G terminal according to the present invention; Figure 2 is a block diagram of a method for testing the wireless communication performance of a 5G terminal according to the present invention. Detailed Embodiments

[0014] Now, the subject matter described herein will be discussed with reference to exemplary embodiments. It should be understood that discussing these embodiments is only to enable those skilled in the art to better understand and thus implement the subject matter described herein. Without departing from the scope of protection of the content of this specification, changes can be made to the functions and arrangements of the elements discussed. Each example can omit, substitute, or add various processes or components as needed. Additionally, the features described relative to some examples can also be combined in other examples.

[0015] As Figures 1 to 2 shown, a method for testing the wireless communication performance of a 5G terminal includes: Step 1, initialize the target terminal and generate matching data to train an association model; Step 2, based on the association model, determine the blocked antennas of the target terminal within the target time period; Step 3, within the target time period, shield the unblocked antennas of the target terminal, collect the communication performance index data of the target terminal, and calculate the communication performance evaluation value of the target terminal.

[0016] It should be noted that the target terminal is defined as a 5G mobile phone in this application. A 5G mobile phone usually includes multiple antennas, and each antenna dynamically switches its working state according to its own blocking situation. Existing AI such as deepseek is intelligently integrated into the system bottom layer of the 5G mobile phone by mobile phone manufacturers to build a personalized experience for users. For example, users may have various grasping postures of the mobile phone, including: vertical screen grasping (reading) and horizontal screen grasping (playing games), so the blocked antennas change dynamically based on the change of the application. When the system-level AI obtains the association between a large amount of blocking information and the startup instructions of the application, it will actively build an association model to immediately switch the working state of the antenna as soon as the startup instructions of the application are obtained, ensuring a smooth experience for users.

[0017] Therefore, initialize (format) the 5G terminal to be tested to reset the association model of the 5G terminal. Based on strict startup instructions and fixed antenna occlusion, induce the AI of the 5G terminal to reconstruct the association model. For example, when starting application A, artificially occlude antenna 135. When starting application B, artificially occlude antenna 246. Based on a large number of repeated operations over a long period of time, force the AI of the 5G terminal to actively construct an association model that meets expectations. That is, when any application of the 5G terminal is started, the association model will actively control the antennas that were artificially occluded previously to be switched to the standby state, and the antennas that were not artificially occluded previously to be switched to the working state, then the obtained association model is determined.

[0018] After obtaining the association model, start application A, then artificially occlude 246, so as to simulate the extreme situation of the prediction failure of the association model, and then obtain the corresponding communication performance data, and calculate the communication performance evaluation value of the 5G mobile phone, which can reflect the communication performance redundancy ability of the 5G mobile phone.

[0019] In an embodiment of the present invention, generating matching data includes: Perform real number encoding on the M antennas of the target terminal to obtain M antenna encodings; Allocate N occlusion sequences to the N application programs of the target terminal respectively, and each occlusion sequence includes m non-repeating antenna encodings; where 1 < m < M, m is a positive integer, and the matching error between any two occlusion sequences is greater than the first error threshold.

[0020] Specifically, perform real number encoding on the M antennas in the 5G mobile phone and allocate occlusion sequences to different application programs respectively. Each occlusion sequence consists of multiple non-repeating antenna encodings, and by setting the first error threshold to ensure significant differences between different sequences. The significance of this solution is to provide a structured and highly distinguishable data basis for inducing the mobile phone AI to reconstruct the association model subsequently, ensuring that the antenna occlusion states in different application scenarios can be clearly distinguished during the training process, thus laying a key data support for simulating AI prediction failure and accurately evaluating the communication redundancy performance.

[0021] In an embodiment of the present invention, select a 5G mobile phone with a system-level AI built-in as the test object, and this mobile phone includes 6 independent antennas (i.e., M = 6). Perform real number encoding on these 6 antennas respectively, and assign encoding values 1, 2, 3, 4, 5, and 6 respectively. Subsequently, generate occlusion sequences for the 4 application programs (i.e., N = 4) preset in this 5G mobile phone respectively. Each sequence includes 3 non-repeating antenna encodings (where m = 3, satisfying 1 < m < 6), and by setting the matching error between each sequence to be greater than 0 to ensure obvious distinguishability between sequences. For example, the occlusion sequence of application A can be set as , the sequence of Application B is set as , the sequence of Application C is set as , and the sequence of Application D is set as .

[0022] In an embodiment of the present invention, training an association model includes: Within a first preset time period, establish an occlusion unit and a program control unit; Step 11, start the nth application of the target terminal based on the program control unit; where 1 ≤ n ≤ N and n is a positive integer; Step 12, in response to the start instruction of the nth application, obtain m antenna encodings in the occlusion sequence of the nth application, and shield the corresponding m antennas through the occlusion unit; Step 13, repeatedly execute Step 11 and Step 12 until: During the test time period, obtain the status information of M antennas of the target terminal respectively; where the status information includes: working status and standby status; Based on the program control unit, sequentially traverse N applications, and obtain the status information of M antennas in response to the start instruction of each application respectively; If the status information is the working status, mark the corresponding antenna as 1, otherwise mark the corresponding antenna as 0, and extract the antennas marked as 1; Determine the set of antennas marked as 1 in response to the start instruction of the nth application, and calculate the matching error between the antenna set and the occlusion sequence corresponding to the nth application; If the matching errors of all N applications are less than the second error threshold, the association model converges, stop executing Step 11 and Step 12, otherwise repeatedly execute Step 11 and Step 12.

[0023] In an embodiment of the present invention, the occlusion unit may be a plurality of shielding sheets constructed of metallic iron. After the shielding sheets respond to the start instruction, they are attached to the part of the edge of the 5G mobile phone where the antennas are located based on the transmission component, thereby realizing signal shielding.

[0024] In an embodiment of the present invention, the program control unit may be a macro command clicker, and the macro command clicker is used to click on different areas of the 5G mobile phone at a preset time interval to start different applications.

[0025] In an embodiment of the present invention, obtaining the status information of M antennas of the target terminal includes: After receiving an application startup instruction, the 5G terminal performs radio frequency monitoring on the 5G terminal, and collects key RF parameters of M antennas in real time, such as transmit power, received signal strength, and S parameters, etc.; by comparing these collected data with preset detection thresholds, if the parameters of a certain antenna exceed the set standard, it is determined to be in the working state, otherwise it is regarded as the standby state.

[0026] For example, when receiving the startup instruction of application A, then block , when receiving the startup instruction of application B, then block , and so on.

[0027] In an embodiment of the present invention, the program control unit starts each application in sequence according to a preset order (1≤n≤N), and executes step 11 and step 12 each time it starts. The system immediately collects the current status information of all M antennas of the target terminal, and the information includes whether each antenna is in the working state or the standby state. For each collected status, the system marks the antennas in the working state as "1", and marks the antennas in the standby state as "0". During this process, the system simultaneously extracts all the antennas marked as "1" to form a set of working antennas for the current application. For each application, the system compares the obtained set of working antennas with the preset occlusion sequence of the application, and calculates the matching error between the two. The matching error reflects whether the current shielding operation conforms to the expected occlusion strategy, and the smaller the error value, the higher the matching degree. When the matching errors for all N applications are lower than the preset second error threshold, it means that during the long-term and multi-round testing process, the antenna state switching operation of the terminal has been stable and tends to be consistent. At this time, it is considered that the built-in association model of the terminal has converged, and the testing process terminates; otherwise, the system continues to repeat step 11 and step 12 for further training and data collection until the matching error meets the requirements.

[0028] In an embodiment of the present invention, the matching error includes the first matching error between the occlusion sequences and the second matching error between the antenna set and the occlusion sequence; The first matching error includes: Associate the same antennas in the occlusion sequence with the occlusion sequence; If the association is successful, then remove the corresponding antennas from the occlusion sequence and the occlusion sequence respectively; Obtain the occlusion sequence and the occlusion sequence after the removal is completed, calculate the sum of the number of remaining units in the occlusion sequence and the occlusion sequence, and use the sum value as the matching error between the occlusion sequence and the occlusion sequence; The second matching error includes: Associate the antenna set with the same antennas in the occlusion sequence; If the association is successful, the corresponding antennas are respectively removed from the antenna set and the occlusion sequence; Obtain the antenna set and the occlusion sequence after the removal is completed, calculate the sum of the number of remaining units in the antenna set and the occlusion sequence, and use the sum value as the matching error between the antenna set and the occlusion sequence.

[0029] Specifically, the first matching error (internal matching of the occlusion sequence) includes: Compare two preset occlusion sequences, that is, find the same antenna codes that appear in both occlusion sequences one by one. For the successfully matched antenna codes, remove them from their respective occlusion sequences. The remaining part is the antenna code units that are not matched. Calculate the total number of remaining units in these two sequences after removal as the first matching error. This error reflects the consistency or difference within the preset occlusion sequence, that is, the part that cannot fully correspond between the two occlusion sequences. The lower the value, the higher the matching degree of the two sequences.

[0030] Specifically, the second matching error (matching between the observed antenna set and the preset occlusion sequence) includes: Compare the currently working antenna set collected from the terminal with the preset occlusion sequence, and find the same antenna codes in both. For the successfully matched antenna codes, remove them from both the observed antenna set and the preset occlusion sequence. Calculate the total number of remaining antenna code units in both after removal, and use this sum as the second matching error. This error is used to reflect the deviation between the actually detected working antenna set and the preset occlusion sequence, that is, whether the system is consistent with the expected mode during the switching of shielded and non-shielded antennas.

[0031] For example, for the same application program, the preset occlusion sequence S = ; and the actually detected working antenna set O = . The calculation process is as follows: Match the common antennas: Compare the actual working antenna set O with the preset occlusion sequence S. The common antennas of the two are antenna 1 and antenna 5.

[0032] Remove the matching elements: After removing antenna 1 and antenna 5 from O and S respectively, we get: O remaining = ; S remaining = ; Calculate the error: Add the number of remaining units, that is, the matching error = 1 (O remaining) + 1 (S remaining) = 2. This error reflects the deviation between the actual detection result and the preset strategy. The lower the error, the more consistent the actual system's occlusion processing is with the preset occlusion sequence.

[0033] In one embodiment of the present invention, determining the blocked antennas of a target terminal based on an association model includes: During the target time period, start the nth application program of the target terminal based on the program control unit; Based on the association model, determine the antennas in the occlusion sequence corresponding to the nth application program as the blocked antennas of the target terminal during the target time period. In one embodiment of the present invention, shielding the non-blocked antennas of the target terminal includes: In response to the start instruction of the nth application program, the non-blocked antennas among the M blocked antennas of the target terminal are shielded by the occlusion unit.

[0034] In one embodiment of the present invention, the system uses a previously constructed association model to determine which antennas should be regarded as "blocked antennas" during a specific test time period, so as to verify the communication redundancy ability of the terminal. Start the application program: During the target test time period, the program control unit sequentially starts the nth application program (1 ≤ n ≤ N) preset in the terminal. Obtain the occlusion sequence: For each application program, the terminal has pre-established an "occlusion sequence" based on the association model, which clearly stipulates which antennas should be in the blocked state due to factors such as user holding in this application scenario. The system then determines through this model that for the nth application program, the corresponding antennas in the occlusion sequence are regarded as the blocked antennas during the current test period. Respond to the start instruction: When the nth application program is started, the system executes operations according to the preset strategy through the occlusion unit. Shield the non-blocked antennas: The target terminal has M "blocked antenna" candidate units, and the system judges which antennas do not belong to the current occlusion sequence (that is, the antennas determined by the system to be non-blocked). Subsequently, the occlusion unit shields these non-blocked antennas to make them in a shielded state, so that only the antennas that are expected to be in the blocked state are in the active state. The purpose of this method is to simulate an abnormal scenario, that is, in different application scenarios, although the terminal AI predicts which antennas should be blocked through the association model, the non-blocked antennas are artificially shielded, forcing the system to switch to a state different from the normal working mode, and evaluate the redundancy performance of the terminal when the AI prediction fails or under extreme conditions by collecting its communication performance data. All in all, by starting each application program and obtaining the corresponding occlusion sequence according to the association model, the system can accurately identify the blocked antennas in the current application scenario; then use the occlusion unit to shield the remaining (non-blocked) antennas to construct a specific combination of antenna working states, so as to conduct a more realistic and rigorous test and verification of the communication performance.

[0035] In one embodiment of the present invention, the communication performance index data includes: During the target time period, establish a stable communication connection between the target terminal and the target base station; In response to the startup instruction of the nth application, the target terminal sends a data packet to the target base station and collects the communication performance index data of the target terminal, including: The number of bits of data sent by the target terminal, the number of bits of data received by the target base station, the number of error bits of data received by the target base station, the timestamp when the target terminal starts to send data, and the timestamp when the target base station finishes receiving data.

[0036] It should be noted that a shielding environment is established: within the target time period, the target terminal and the target base station are usually placed in a specially designed darkroom or electromagnetic shielding room. This darkroom is composed of high-efficiency wave-absorbing materials and a metal shielding layer, which can effectively isolate external radio waves and interference sources, preventing external electromagnetic noise from affecting the normal operation of the device under test.

[0037] Isolate external interference: In addition to the physical darkroom shielding measures, the test device is also equipped with a dedicated ground wire and grounding system to ensure the stability of the electromagnetic environment inside the darkroom; at the same time, filters and anti-interference circuits can also be installed inside. These measures work together to minimize the interference signals from the surrounding environment.

[0038] Verify communication stability: In this environment where external interference is isolated, real-time monitoring of parameters such as signal strength, bit error rate, and latency of the communication link is carried out through calibration equipment and test instruments to ensure that the connection established between the target terminal and the base station reaches a stable communication state.

[0039] It should be noted that in response to the startup instruction of the nth application, the target terminal starts to send a data packet to the target base station. At the same time, the system collects multiple key communication metrics, including: The number of bits of data sent by the target terminal, which is used to reflect the total amount of data actually transmitted during the sending process by the terminal. The number of bits of data received by the target base station, which is used to reflect the complete amount of data obtained by the base station during the receiving process. The number of error bits of data received by the target base station, which is used to record the amount of error data caused by interference, signal attenuation, etc. during the receiving process by the base station, and is used to evaluate the bit error rate. The timestamp when the target terminal starts to send data, which is used to mark the start time of data packet transmission. The timestamp when the target base station finishes receiving data, which is used to mark the completion time of data packet reception.

[0040] In an embodiment of the present invention, the target communication performance evaluation value of the target terminal is calculated, including: Based on the communication performance index data, the throughput, latency, and bit error rate of the target terminal are calculated, and the throughput, latency, and bit error rate are weighted and fused to obtain the target communication performance evaluation value of the target terminal.

[0041] In an embodiment of the present invention, the throughput includes: Based on the data collected when the target terminal sends data packets during the test, the calculation formula is: Throughput = Number of bits of transmitted data ÷ (Received completion timestamp - Start transmission timestamp); where the unit of throughput is bits per second (bps). The higher the throughput, the stronger the data transmission ability, and it is an important indicator to measure the effective transmission rate of the communication link.

[0042] In an embodiment of the present invention, the latency includes: Latency is the time interval during data transmission, and the calculation formula is: Latency = Received completion timestamp - Start transmission timestamp; where latency is expressed in milliseconds (ms). The lower the latency, the faster the communication response, and it is a key parameter to measure user experience and real-time communication ability.

[0043] In an embodiment of the present invention, the bit error rate includes: Using the number of error bits collected by the target base station and the number of bits actually received, as follows: Bit error rate = Number of error bits ÷ Number of bits of received data; the lower the bit error rate, the better the communication link quality; a higher bit error rate usually means poor signal quality.

[0044] In an embodiment of the present invention, weighted fusion includes: ; Wherein, represents the target communication performance evaluation value, which is used to reflect the communication performance redundancy ability of the target terminal, 、 and represent the first weight, the second weight, and the third weight respectively, 、 and are all not zero, and 、 and The sum value of is 1, represents the normalized throughput, represents the normalized latency, represents the normalized bit error rate.

[0045] A test system for the wireless communication performance of a 5G terminal, which is applied to the above-mentioned test method for the wireless communication performance of a 5G terminal, includes: A training module, which is used to initialize the target terminal and generate matching data to train and obtain an association model; A prediction module, which is used to determine the blocked antenna of the target terminal within the target time period based on the association model; A test module, which is used to shield the non-blocked antennas of the target terminal within the target time period, collect the communication performance index data of the target terminal, and calculate the communication performance evaluation value of the target terminal.

[0046] The above has described the embodiments of this embodiment, but this embodiment is not limited to the above specific implementation manners. The above specific implementation manners are merely illustrative rather than restrictive. Under the inspiration of this embodiment, those of ordinary skill in the art can also make many forms, all of which fall within the protection scope of this embodiment.

Claims

1. A method for testing the wireless communication performance of a 5G terminal, characterized in that: include: Step 1: Initialize the target terminal and generate matching data to train the association model; Step 2: determining the blocked antennas of the target terminal within the target time period based on the association model; Step 3: During the target time period, shield the non-blocked antenna of the target terminal, collect the communication performance index data of the target terminal, and calculate the communication performance evaluation value of the target terminal.

2. A method for testing the wireless communication performance of a 5G terminal according to claim 1, characterized in that: Generate matching data, including: Perform real number encoding on the M antennas of the target terminal to obtain M antenna codes; N occlusion sequences are respectively allocated to N applications of the target terminal, each occlusion sequence includes m non-repeating antenna codes; wherein 1<m<M, m is a positive integer, and the matching error between any two occlusion sequences is greater than a first error threshold.

3. A method for testing the wireless communication performance of a 5G terminal according to claim 2, characterized in that: The training results in the associated model, including: In a first preset time period, a shielding unit and a program control unit are established; Step 11, starting the nth application program of the target terminal based on the program control unit; wherein 1≤n≤N, and n is a positive integer; Step 12, in response to a start instruction of the nth application, obtaining m antenna codes in a shielding sequence of the nth application, and shielding the corresponding m antennas through a shielding unit; Step 13: Repeat steps 11 and 12 until: During the test period, the state information of the M antennas of the target terminal is respectively obtained; wherein the state information includes: working state and standby state; Sequentially traverse the N applications based on the program control unit, and obtain the state information of the M antennas in response to the start instruction of each application respectively; If the status information is in working state, the corresponding antenna is marked as 1, otherwise the corresponding antenna is marked as 0, and the antenna marked as 1 is extracted; Determine an antenna set marked as 1 in response to a start instruction of an n-th application, and calculate a matching error between the antenna set and an occlusion sequence corresponding to the n-th application; If the matching errors of the N applications are all smaller than the second error threshold, the association model converges, and step 11 and step 12 are stopped; otherwise, step 11 and step 12 are repeated.

4. A method for testing the wireless communication performance of a 5G terminal according to claim 3, characterized in that: Matching errors, including a first matching error between the occlusion sequence and the occlusion sequence, and a second matching error between the antenna set and the occlusion sequence; The first matching error includes: Associating the occlusion sequence with the same antenna in the occlusion sequence; If the association is successful, the corresponding antenna is eliminated in the occlusion sequence and the blocked sequence respectively; Obtain the occlusion sequence and the occlusion sequence after culling, calculate the sum of the number of units remaining in the occlusion sequence and the occlusion sequence, and use the sum as the matching error between the occlusion sequence and the occlusion sequence; The second matching error includes: Associating the antenna set with the same antenna in the occlusion sequence; If the association is successful, the corresponding antenna is removed from the antenna set and the occlusion sequence respectively; The antenna set and the occlusion sequence after elimination are obtained, and the sum of the numbers of the remaining units in the antenna set and the occlusion sequence is calculated, and the sum is used as the matching error between the antenna set and the occlusion sequence.

5. A method for testing the wireless communication performance of a 5G terminal according to claim 4, characterized in that: Determine the blocked antenna of the target terminal within the target time period based on the association model, including: In a target time period, starting an nth application program of a target terminal based on a program control unit; Based on the association model, it is determined that the antenna in the blocking sequence corresponding to the nth application is the blocked antenna of the target terminal in the target time period.

6. A method for testing the wireless communication performance of a 5G terminal according to claim 5, characterized in that: Shielding of unobstructed antennas of target terminals, including: In response to the start-up instruction of the nth application program, the non-blocking antenna among the M blocking antennas of the target terminal is shielded by the blocking unit.

7. A method for testing the wireless communication performance of a 5G terminal according to claim 6, characterized in that: Communication performance indicator data, including: Establishing a stable communication connection between the target terminal and the target base station within the target time period; In response to the start instruction of the nth application, the target terminal sends a data packet to the target base station, and collects communication performance indicator data of the target terminal, including: The number of bits of data sent by the target terminal, the number of bits of data received by the target base station, the number of error bits of data received by the target base station, the timestamp when the target terminal starts sending data, and the timestamp when the target base station completes receiving data.

8. A method for testing the wireless communication performance of a 5G terminal according to claim 7, characterized in that: Calculating a target communication performance evaluation value of the target terminal includes: calculating the throughput, delay and bit error rate of the target terminal based on communication performance indicator data, and weightedly integrating the throughput, delay and bit error rate to obtain the target communication performance evaluation value of the target terminal.

9. A 5G terminal wireless communication performance test system, applied to a 5G terminal wireless communication performance test method according to any one of claims 1 to 8, characterized in that: include: The training module is used to initialize the target terminal and generate matching data to train the association model; A prediction module, for determining the blocked antenna of the target terminal within the target time period based on the association model; The test module is used to shield the non-blocked antenna of the target terminal within the target time period, collect the communication performance index data of the target terminal, and calculate the communication performance evaluation value of the target terminal.

Images