A closed-loop channel restoration system based on a programmable attenuator and a signal simulator

Abstract

This invention relates to the field of wireless communication testing technology and discloses a closed-loop channel restoration system based on a programmable attenuator and a signal simulator. The system includes: a signal source unit for outputting a downlink radio frequency (RF) signal; a programmable attenuation unit whose input is connected to the output of the signal source unit, used to programmatically attenuate the downlink RF signal according to a first control command; a noise injection unit whose output is connected to the output link of the programmable attenuation unit via a coupler, used to inject controllable noise into the link according to a second control command; a signal acquisition terminal connected to the downlink, used to acquire measured RSRP and SINR values ​​in real time; and a closed-loop control unit communicatively connected to the programmable attenuation unit, the noise injection unit, and the signal acquisition terminal, used to execute independent closed-loop control of RSRP and independent closed-loop control of SINR. This invention can simultaneously solve the problems of parameter decoupling and dynamic point-by-point reproduction, is cost-controllable, and has real-time feedback calibration capabilities.

Description

Technical Field

[0001] This invention relates to the field of wireless communication testing technology, and more specifically to a closed-loop channel restoration system based on a programmable attenuator and a signal simulator. Background Technology

[0002] With the large-scale commercialization of 5G and intelligent connected vehicle technologies, the demand for communication performance verification of various terminal devices in real road environments has exploded. However, traditional field testing methods have long been hampered by inherent shortcomings such as high manpower and material costs, uncontrollable environmental variables (such as weather, traffic flow, pedestrian flow, and electromagnetic interference), and the inability to accurately reproduce test scenarios, leading to extended product development cycles and difficulties in problem localization. To address this contradiction, the industry has begun to turn its attention to laboratory channel environment reproduction technology. This technology uses indoor equipment to simulate the wireless propagation characteristics of the field, reproducing the complex operating conditions encountered by terminals in real road tests, such as signal fading and interference changes, under controlled conditions, thereby balancing testing efficiency and repeatability.

[0003] Current mainstream channel reconstruction schemes mostly rely on expensive vector channel simulators. These simulators pre-load theoretical channel models or channel files generated from field-acquired data to control signal attenuation and noise addition in an open-loop manner. While these schemes can simulate macroscopic fading statistical characteristics, they exhibit several shortcomings in practical applications: First, existing methods generally cannot synchronously and accurately restore the two key parameters, the reference signal received power (RSRP) and the signal-to-interference-plus-noise ratio (SINR). The two parameters are mutually constrained during adjustment; adjusting RSRP often leads to changes in SINR, and vice versa, resulting in severe coupling crosstalk.

[0004] Secondly, most commercial solutions employ an open-loop control architecture, lacking a real-time feedback calibration mechanism. They rely solely on pre-calculated channel files to set attenuation values ​​once, failing to dynamically correct based on measured values. This results in significant deviations between the output RSRP and SINR and the design targets. A few attempts to introduce closed-loop control are limited to a single parameter, and independent closed-loop operation is prone to oscillations, slow convergence, and poor synchronization between multiple channels, making it difficult to reproduce the continuous and rapid channel fluctuations in field time series.

[0005] Third, channel reconstruction technology has long faced an irreconcilable contradiction between accuracy and cost. High-precision solutions require top-tier vector channel simulators, which are extremely expensive, and the channel modeling process is complex and has a high operational threshold, limiting their adoption by small and medium-sized R&D teams. Low-cost, simplified solutions typically only include programmable attenuators, lack noise injection capabilities, or only have fixed noise sources, and completely lack closed-loop compensation mechanisms. After long-term operation, errors accumulate due to factors such as equipment temperature drift and changes in link loss, resulting in severely insufficient stability and consistency.

[0006] Fourth, existing solutions generally lack the ability to compensate for inherent link losses and device power drift in real time, and the feedback adjustment mechanism is imperfect. Neither open-loop nor single closed-loop systems can cope with slow drift during the test process, resulting in poor data comparability in long-term batch tests and failing to meet the requirements of high-precision dynamic testing. Summary of the Invention

[0007] The present invention aims to provide a closed-loop channel restoration system based on a programmable attenuator and a signal simulator, which can simultaneously solve the problems of parameter decoupling and dynamic point-by-point reproduction, with controllable cost and real-time feedback calibration capability.

[0008] The basic solution provided by this invention is: a closed-loop channel restoration system based on a programmable attenuator and a signal simulator, comprising: The signal source unit is used to output downlink radio frequency signals; A programmable attenuation unit, the input of which is connected to the output of the signal source unit, is used to programmably attenuate the downlink radio frequency signal according to a first control command; The noise injection unit has its output terminal connected to the output link of the programmable attenuation unit via a coupler, and is used to inject controllable noise into the link according to the second control command. The signal acquisition terminal is connected to the downlink and is used to acquire the measured values ​​of RSRP and SINR in real time. The closed-loop control unit is communicatively connected to the programmable attenuation unit, the noise injection unit, and the signal acquisition terminal, and is used to execute RSRP independent closed-loop control and SINR independent closed-loop control. Specifically, the closed-loop control unit first generates the first control command based on the difference between the target RSRP sequence acquired in the field and the measured RSRP value, using a PID negative feedback algorithm, and adjusts the programmable attenuation unit until the measured RSRP value converges to the target RSRP value; based on RSRP locking, it then generates the second control command based on the difference between the target SINR sequence acquired in the field and the measured SINR value, using a PID negative feedback algorithm, and adjusts the noise injection unit until the measured SINR value converges to the target SINR value; The above process is repeated point by point in chronological order to generate a time-aligned channel restoration control data table. Based on this data table, the programmable attenuation unit and the noise injection unit are controlled synchronously to continuously reproduce the dynamic changes of RSRP and SINR acquired in the field.

[0009] The working principle and advantages of this invention are as follows: This invention discloses a closed-loop channel restoration system based on a programmable attenuator and a signal simulator, which can simultaneously solve the problems of parameter decoupling and dynamic point-by-point reproduction, while maintaining controllable cost and providing real-time feedback calibration capabilities. The key features are: First, this solution fundamentally solves the long-standing coupling crosstalk problem between the two key parameters, RSRP and SINR, in existing channel restoration techniques. By constructing independent dual closed-loop control channels, the RSRP adjustment task is dedicated to the programmable attenuation unit, and the SINR adjustment task is dedicated to the noise injection unit. Furthermore, a serial decoupling logic of "locking the RSRP reference first, then calibrating SINR" is adopted, fundamentally severing the mutual influence path between the two parameters. This ensures that the RSRP and SINR closed loops are strictly separated in time and do not interfere with each other functionally, completely eliminating parameter coupling oscillations during the adjustment process. This ensures that RSRP and SINR at each time point can accurately match the measured values ​​in the field, providing high-quality reference data for subsequent continuous reproduction.

[0010] Secondly, based on this, a PID negative feedback algorithm is introduced, which allows each channel to iteratively correct the control command in real time using the difference between the measured value and the target value. This avoids the problem of the pre-set open-loop command in the open-loop scheme failing due to link drift or device nonlinearity. This not only significantly improves the accuracy of single-point calibration, but also effectively suppresses the adjustment oscillation caused by parameter coupling in the traditional closed-loop scheme. This enables the system to smoothly converge to the target value within a few iterations, significantly improving the stability and reliability of channel restoration.

[0011] Secondly, this solution can completely reproduce the real channel changes in a continuous time series in the field using point-by-point time alignment, demonstrating strong dynamic channel reproduction capabilities. By performing point-by-point closed-loop calibration on the RSRP and SINR time series acquired from the field, a programmable attenuation value and noise injection power value corresponding to each timestamp are generated, ultimately forming a channel reproduction control data table perfectly aligned with the control time axis. During the playback phase, the real-time control module synchronously drives the programmable attenuation unit and noise injection unit according to this data table, ensuring that the downlink signal power and signal-to-noise ratio at each moment remain consistent with the values ​​acquired in the field. This "point-by-point calibration—time-sequence locking—continuous reproduction" technical approach enables the laboratory environment to reproduce the dynamic fluctuations of the signal in the real field (including slow fading, fast fading, and sporadic signal fluctuations—all-scenario characteristics) with near-one-to-one fidelity, providing a realistic testing environment for terminal performance verification in complex mobile scenarios.

[0012] Third, this solution achieves an excellent balance between hardware cost and system flexibility, significantly lowering the entry barrier for high-precision channel reconstruction technology. It utilizes industrially common programmable attenuators and vector signal simulators as core execution components. These devices are highly mature in the field of communication testing and their cost is far lower than that of channel simulators with equivalent performance. More importantly, this solution compensates for the shortcomings of discrete devices in collaborative accuracy through closed-loop control algorithms, achieving reconstruction accuracy superior to traditional open-loop channel simulators. Simultaneously, the system adopts a fully modular design, allowing for flexible hardware customization based on the actual needs of the testing scenario: a complete dual-closed-loop architecture can be deployed for full-function performance testing in the laboratory, while a lightweight version containing only RSRP closed-loop can be tailored for field fixed-point semi-real-vehicle testing or rapid sampling inspection on the production line. A single system is compatible with multiple testing levels, eliminating the need to purchase multiple sets of dedicated equipment for different scenarios and avoiding resource waste caused by hardware redundancy.

[0013] Fourth, this solution effectively solves the problems of error accumulation and equipment drift during long-term operation through a real-time closed-loop feedback mechanism, significantly improving the long-term consistency and comparability of test results. The closed-loop negative feedback control itself has real-time correction capability. During the PID iteration at each time point, the actual measured values ​​of RSRP and SINR will directly participate in the correction of control commands. Any error caused by equipment drift or environmental changes will be captured and compensated by the difference in the next iteration, so that even if the system runs continuously for several hours or even several days, its output parameters can still be stably maintained within the target error range, providing a reliable basic environment for batch terminal comparison testing and regression testing. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the system structure of an embodiment of a closed-loop channel restoration system based on a programmable attenuator and a signal simulator according to the present invention; Figure 2 This is a schematic diagram of the system operation process of an embodiment of a closed-loop channel restoration system based on a programmable attenuator and a signal simulator according to the present invention. Detailed Implementation

[0015] The following detailed explanation illustrates the specific implementation methods: The basic implementation examples are as follows: Figure 1 The system illustrates a closed-loop channel restoration system based on a programmable attenuator and a signal simulator, used to accurately reproduce the wireless channel variation characteristics acquired in the field in a laboratory environment. All hardware components are deployed in an electromagnetically shielded anechoic chamber or a standard laboratory environment. The RF link uses 50Ω impedance-matched coaxial RF cable throughout, and the control link uses Category 5e shielded twisted-pair cable to avoid electromagnetic interference affecting control accuracy and measurement precision.

[0016] The system mainly includes: a signal source unit, a programmable attenuation unit, a noise injection unit, a signal acquisition terminal, and a closed-loop control unit. Detailed descriptions are as follows: The signal source unit is used to output standard cellular mobile communication downlink radio frequency signals, providing a stable basic signal source for channel restoration. In this embodiment, the signal source unit can be a communication test instrument (e.g., a CMX500 communication test instrument) or commercial mobile communication base station equipment (e.g., 4G / 5G pico base stations or micro base stations); the equipment needs to support mainstream frequency bands of LTE FDD / TDD and 5G NR, and the output power stability should be ≤±0.2dB / hour to meet the consistency requirements of long-term testing.

[0017] The RF input terminal of the programmable attenuation unit is connected to the RF output terminal of the signal source unit via a coaxial RF cable, and its RF output terminal is connected to the downlink test link. The programmable attenuation unit is used to programmatically attenuate the power of the downlink RF signal according to the first control command issued by the closed-loop control unit. The attenuation step of the programmable attenuation unit is no greater than 0.1dB, the attenuation accuracy is ±0.2dB, and the response time is no greater than 5ms.

[0018] In this embodiment, the programmable attenuation unit adopts a high-precision programmable attenuator with a frequency coverage of 600MHz to 6GHz, an attenuation range of 0 to 120dB, a step of 0.1dB, an attenuation accuracy of ±0.2dB, supports SCPI standard command control, and has a response time of ≤5ms to meet the real-time requirements of closed-loop control.

[0019] The noise injection unit is used to inject controllable noise (specifically, a controllable Gaussian white noise signal) into the link according to the second control command issued by the closed-loop control unit. The output of the noise injection unit is connected to the output link of the programmable attenuation unit via a coupler. In this embodiment, the noise injection unit is a high-precision vector signal simulator whose frequency coverage matches that of the signal source unit. Its output power range is -120dBm to 0dBm, with a power step of 0.1dB and an accuracy of ±0.2dB. It supports Gaussian white noise signal generation and can adjust the output power in real time via SCPI commands.

[0020] The coupler is a directional RF coupler with a coupling degree of 10dB to 20dB, an isolation degree of ≥25dB, and a frequency coverage range matching the system frequency band. Its main path is connected in series between the output of the programmable attenuation unit and the receiving end of the signal acquisition terminal, and its coupling injection port is connected to the output of the noise injection unit to achieve unidirectional combining of the noise signal and the downlink RF signal. Furthermore, the main path insertion loss of this coupler is ≤0.5dB, ensuring that the noise signal is injected unidirectionally into the downlink without damaging the programmable attenuation unit in the reverse direction, and the impact on the receiving power of the device under test is negligible.

[0021] The signal acquisition terminal is connected to the downlink and is used to collect the measured values ​​of RSRP (Reference Signal Received Power) and SINR (Signal to Interference Plus Noise Ratio) in the current restoration environment in real time, and transmit the collected data back to the closed-loop control unit in real time.

[0022] In this embodiment, the signal acquisition terminal uses the same type of cellular communication module as the field road test data acquisition module (e.g., Quectel AG550Q automotive-grade 5G module, Huawei MH5000 module), and its sampling rate is the same as that of the field acquisition (e.g., 10Hz). It also supports real-time output of RSRP and SINR measured values ​​via AT commands or application programming interface (API) to ensure time synchronization.

[0023] The closed-loop control unit is communicatively connected to the programmable attenuation unit, the noise injection unit, and the signal acquisition terminal, respectively, and is used to execute RSRP independent closed-loop control and SINR independent closed-loop control.

[0024] Specifically, the closed-loop control unit further includes a PC host computer and a real-time control module. The PC host computer is an industrial control computer, running a Windows or Linux operating system and host computer software, and communicates with all the aforementioned hardware units via an Ethernet switch. The PC host computer is responsible for preprocessing the raw field-acquired data (such as parsing and separation), scheduling the entire test process, configuring control parameters, and analyzing test results.

[0025] The real-time control module employs an industrial controller (such as Beckhoff CX series PLC or NI CompactRIO) with a real-time operating system like QNX or RTX. Its control cycle is ≤10ms, and it runs a decoupled dual-closed-loop PID negative feedback control algorithm to perform difference calculation and real-time output of control commands. The PC host computer and the real-time control module establish real-time data interaction via a local area network. The PC host computer sends the pre-processed target sequence to the real-time control module, and the real-time control module sends the measured data, control commands, and calibration results back to the host computer for storage in real time. All devices in the system use the PC host computer as an NTP time server for clock synchronization, with a time synchronization error ≤10ms, ensuring complete alignment of timestamps for all data acquisition and control commands.

[0026] Furthermore, the PC host computer of the closed-loop control unit has a built-in standardized script instruction set to support fully automated testing and parallel scheduling of multiple test nodes. The fully automated testing includes channel restoration, parameter calibration, data acquisition, and report output. Specifically, users can define the test process by writing scripts. For example, the script can contain the following instruction sequence: field data import, link calibration, point-by-point RSRP calibration, point-by-point SINR calibration, channel restoration playback, performance data acquisition of the terminal under test, and report generation. The entire process can achieve unattended automated operation without human intervention. In the parallel scheduling of multiple test nodes, with multiple independent test stations, each station deploys a complete RF link and signal acquisition terminal, and multiple sets of hardware are connected to the same PC host computer through a switch. The PC host computer allocates independent memory space and process resources to each node, and the test tasks of each node can be scheduled in parallel and run without interference. For example, RSRP / SINR restoration tests for the LTE channel can be performed at node A, while restoration tests for the 5G NR channel can be performed at node B. The two nodes operate independently and the data is stored independently, which helps to achieve efficient reuse of test resources and shorten the product development and verification cycle.

[0027] When used for testing multi-mode communication systems (e.g., simultaneously testing TD-LTE and 5G NR), the system can include multiple independent signal source units, programmable attenuation units, and noise injection units, each corresponding to a different communication standard. The closed-loop control unit synchronously coordinates the independent dual-closed-loop control of each communication standard to reproduce signal changes in the multi-mode field environment. In this case, multiple hardware modules are combined via a combiner and then connected to the signal acquisition terminal and the device under test, with the control links remaining independent but clock-synchronized.

[0028] For example, if it is necessary to simultaneously simulate field signal changes in both LTE and 5G NR bands, the system will deploy two completely independent RF links: the first includes an LTE signal source unit (LTE base station or comprehensive tester), an LTE programmable attenuation unit, an LTE noise injection unit, and corresponding directional couplers; the second includes an NR signal source unit, an NR programmable attenuation unit, an NR noise injection unit, and their couplers. The two links are physically independent, with their respective programmable attenuators and signal simulators controlled by the same closed-loop control unit (PC host computer + real-time control module). Each link independently completes point-by-point calibration of its respective field acquisition sequence according to the "RSRP first, SINR later" serial dual closed-loop process under a single standard, and generates its own channel restoration control data table. Since the field acquisition timestamps of the two standards may differ (e.g., LTE sampling period 10ms, NR sampling period 5ms), the closed-loop control unit will use interpolation or resampling during the host computer preprocessing stage to align the time axes of the two sequences to the same reference clock, ensuring that each time point corresponds to the same field geographical location or time during subsequent synchronous playback.

[0029] After completing independent calibration for each standard, the PC host computer merges two (or more) sets of control data tables into a single synchronous scheduling table. The real-time control module triggers playback commands via a global clock. At each time point, the real-time control module simultaneously sends the attenuation value for that moment to the LTE programmable attenuator and the noise power for that moment to the LTE signal simulator, and simultaneously sends the NR attenuation value for the NR programmable attenuator and the NR noise power for the NR signal simulator. All commands are issued in parallel within the same control cycle, and the programmable attenuation and noise addition for each standard are executed independently without interference. Since the RF signals of each standard operate in different frequency bands (e.g., LTE Band 3 and NR Band 78), they will not generate intermodulation interference when transmitted on the same feeder. They can be passively combined using a broadband combiner or a duplexer and then fed together into the signal acquisition terminal and the mobile terminal under test. The signal acquisition terminal is responsible for verifying whether the RSRP / SINR of each signal type simultaneously meets the target value, while the closed-loop control unit only uses the readings of the signal acquisition terminal for verification during the calibration phase; during the actual testing phase, the signal acquisition terminal can be bypassed or used only for monitoring, and the terminal under test can be normally connected to each network type.

[0030] Furthermore, the key to multi-mode reproduction lies in realistically reproducing the interoperability process between different standards of the terminal in the field. During field data acquisition, it is necessary to simultaneously record the RSRP / SINR time series of LTE and NR, and these sequences reflect the handover process as the terminal moves along the test line, with the primary serving cell gradually switching from LTE to NR (or vice versa). For example, within a certain time window, the LTE signal weakens while the NR signal strengthens, and the terminal performs inter-system handover at the signal crossover point. The multi-mode system of this invention can completely reproduce this dynamic: the LTE link gradually attenuates the signal and injects noise according to its sequence, while the NR link synchronously strengthens the signal and reduces noise according to its sequence. In the combined RF environment, the terminal under test will realistically experience the gradual fading of the LTE signal and the gradual rise of the NR signal, thereby triggering the same measurement reports and handover commands as in a real field. Since the parameter changes of the two links are based entirely on the time-aligned data collected in the field, and the RSRP and SINR within each standard are precisely calibrated in a closed loop, the performance indicators such as handover threshold, latency, and success rate exhibited by the terminal in indoor testing are highly consistent with those in field testing.

[0031] Furthermore, this system adopts a modular architecture, allowing for flexible hardware module customization based on test scenario requirements. In full-function laboratory performance testing scenarios, a complete dual-closed-loop architecture comprising a signal source unit, a programmable attenuation unit, a noise injection unit, a signal acquisition terminal, and a closed-loop control unit is employed to achieve the highest accuracy in channel reconstruction. In field, fixed-point semi-real-vehicle testing scenarios, where environmental noise is relatively controllable and equipment portability is paramount, a lightweight architecture omitting the noise injection unit is adopted, retaining only RSRP closed-loop control, while SINR is approximated using a fixed attenuator or a simple noise source. A single system can accommodate multiple testing needs, helping to reduce users' equipment procurement costs and maintenance complexity.

[0032] In addition, before the system is run for the first time, in order to eliminate the impact of the inherent loss of the RF link on the control accuracy, it is necessary to perform link inherent loss calibration through the closed-loop control unit: The PC in the closed-loop control unit first initiates the calibration process: controlling the signal source unit to output a continuous wave signal with a fixed power P0 (e.g., -20dBm), while simultaneously setting the programmable attenuation unit to 0dB attenuation. It then acquires the received power P1 from the signal acquisition terminal and calculates the link loss L based on the difference between the fixed power and the received power. The link loss includes RF cable loss, coupler insertion loss, and connector contact loss. This link loss is used as the initial attenuation calculation benchmark for the PID negative feedback algorithm and written into the configuration file. This calibration process can be executed automatically during system power-on self-test or manually triggered after a test scenario switch to compensate for link drift caused by changes in ambient temperature or equipment aging.

[0033] When the system is running, the closed-loop control unit first generates the first control command based on the difference between the target RSRP sequence collected from the field and the measured RSRP value, using a PID negative feedback algorithm, and adjusts the programmable attenuation unit until the measured RSRP value converges to the target RSRP value; based on RSRP locking, the second control command is then generated based on the difference between the target SINR sequence collected from the field and the measured SINR value, using a PID negative feedback algorithm, and the noise injection unit is adjusted until the measured SINR value converges to the target SINR value; The above process is repeated point by point in chronological order to generate a time-aligned channel restoration control data table. Based on this data table, the programmable attenuation unit and the noise injection unit are controlled synchronously to continuously reproduce the dynamic changes of RSRP and SINR acquired in the field.

[0034] The PID negative feedback algorithm is an incremental PID algorithm. Its proportional coefficient Kp, integral coefficient Ki, and derivative coefficient Kd are tuned on-site so that the RSRP error converges to within ±1dBm within 3 to 5 iterations, and the SINR error converges to within ±1dB within 3 to 5 iterations.

[0035] In specific applications, such as Figure 2 As shown, the operation flow of this system is as follows: Step 1: Raw data acquisition and preprocessing in the field.

[0036] First, users need to collect field data in the target road test scenario using the same equipment as the field test terminal (i.e., the same model of the signal acquisition terminal).

[0037] Design a drive test transmission line that includes at least three coverage scenarios: strong, medium, and poor signal coverage. The field test terminal initiates a service request and accesses the network, recording the RSRP and SINR values ​​at each sampling time using a set sampling frequency (e.g., 10Hz), obtaining the original RSRP and SINR data sequences aligned to timestamps. The collected raw data is imported into a PC host computer. The host computer software preprocesses the raw data: removing outliers and null values, eliminating random noise through moving average filtering, and realigning the RSRP and SINR sequences to ensure that each timestamp corresponds to a unique set of target RSRP and SINR values. Finally, a target RSRP and target SINR sequence arranged in chronological order are generated, with the total sequence length denoted as N.

[0038] Step 2, point-by-point RSRP single closed-loop calibration control.

[0039] The real-time control module extracts the target RSRP value at the nth time point from the preprocessed sequence in chronological order, denoted as . , where the initial value of n is 1.

[0040] The real-time control module determines the link's inherent loss L and the output power of the signal source unit (a known value). ), calculate the initial attenuation: .

[0041] The real-time control module converts the initial attenuation value into an SCPI command recognizable by the programmable attenuation unit, generates a first control command, and sends it to the programmable attenuation unit via the local area network. The programmable attenuation unit executes the command and adjusts its attenuation value.

[0042] The signal acquisition terminal collects the measured RSRP value of the current link in real time, and records it as... The measured values ​​are transmitted back to the real-time control module in real time via the local area network. The real-time control module calculates the difference. Determine whether ΔRSRP is less than or equal to a preset first threshold (in this embodiment, the first threshold is 1dBm).

[0043] like The real-time control module, based on the incremental PID algorithm, calculates the negative feedback adjustment and generates a new first control command, which is then sent to the programmable attenuation unit. The core formula of the incremental PID algorithm is: ; in, The increment is used to adjust the attenuation at the current time, where Kp is the proportional coefficient, Ki is the integral coefficient, Kd is the differential coefficient, and e(k) is the RSRP difference at the current time (i.e., e(k-1) and e(k-2) are the differences between the first two time points. After on-site tuning, the optimal parameters are Kp=0.8, Ki=0.1, and Kd=0.05.

[0044] The real-time control module will generate a new command for the adjusted attenuation and send it out, and then read the data again. The comparison is then performed. This iterative process is repeated until... In actual testing, this scheme can converge the RSRP error to within ±1dBm within 3 to 5 iterations, without overshoot and oscillation.

[0045] when When the real-time control module determines that the RSRP calibration is complete at the current time point, it locks the current attenuation amount of the programmable attenuation unit and records the attenuation value in the channel restoration control data table.

[0046] Subsequently, the real-time control module determines whether n is equal to the total sequence length N. If Then let Return to the beginning of this step and perform RSRP calibration at the next time point; if If the calibration is complete, the entire RSRP sequence calibration is considered complete, and the SINR closed-loop calibration process begins.

[0047] Step 3: Point-by-point SINR single closed-loop calibration control.

[0048] After all RSRP sequences have been calibrated, the real-time control module begins calibration of the second independent closed-loop channel, namely SINR calibration. It is worth noting that SINR calibration is performed only after the RSRP at the corresponding time point has been precisely locked, thus completely eliminating coupling crosstalk between the two parameters.

[0049] The real-time control module extracts the target SINR value at the m-th time point from the preprocessed sequence in chronological order, denoted as . The initial value of m is 1. Simultaneously, the real-time control module reads the calibrated attenuation of the programmable attenuator corresponding to that time point from the channel restoration control data table and maintains this attenuation constant to ensure the stability of the RSRP reference value.

[0050] The real-time control module uses the currently locked RSRP reference value. and goals The required noise injection power is calculated. Specifically, SINR is defined as the ratio (linear value) of signal power to noise power, therefore the required noise power... (Linear value) = Signal power (linear value) / (Linear value). The signal power is the linear power value corresponding to RSRP_locked(m). The calculated noise power is converted to dBm units, and an initial second control command is generated and sent to the noise injection unit. According to the command, the noise injection unit injects Gaussian white noise of the corresponding power into the downlink through a directional RF coupler.

[0051] The signal acquisition terminal collects the measured SINR value of the current link in real time, and records it as follows. The measured values ​​are then transmitted back to the real-time control module in real time. The real-time control module calculates the difference. Determine whether ΔSINR is less than or equal to a preset second threshold (in this embodiment, the second threshold is 1dB).

[0052] like The real-time control module is also based on an incremental PID control algorithm (which can optionally use the same or different PID parameters as the RSRP channel; in this embodiment, the PID parameters of the SINR channel are tuned to Kp=0.6, Ki=0.1, and Kd=0.03). It calculates the negative feedback adjustment amount and generates a new second control command, which is then sent to the noise injection unit. The iterative adjustment process is similar to RSRP, until... Experimental results show that this scheme can converge the SINR error to within ±1dB within 3 to 5 iterations.

[0053] when At that time, the real-time control module determines that the SINR calibration is complete, locks the current output power of the noise injection unit, and records this power value in the channel restoration control data table. Subsequently, the real-time control module determines whether m equals the total sequence length N. If... Then let Return to the beginning of this step and perform SINR calibration at the next time point; if If the SINR sequence calibration is complete, a channel restoration control data table that is fully aligned in chronological order is generated. This data table contains the programmable attenuator attenuation value and noise injection power value for each time point.

[0054] Step 4: Recreate and test the continuous channel scenario.

[0055] After completing the point-by-point calibration, the PC host computer generates a continuous sequence of synchronization control commands based on the final channel restoration control data table and sends it to the real-time control module. The real-time control module synchronously controls the programmable attenuation unit and noise injection unit according to the control commands corresponding to each timestamp, continuously reproducing the dynamic changes in RSRP and SINR acquired in the field. Because the control commands are executed point-by-point in exactly the same time sequence as the field data, the system can reproduce the continuous dynamic changes of the real field channel 1:1, including all-scenario characteristics such as slow fading, fast fading, and occasional signal fluctuations.

[0056] During the channel scenario reconstruction process, the RF antenna port of the mobile terminal under test (such as the 5G mobile phone or vehicle-to-everything (V2X) communication module under test) is connected to the main output port of the directional RF coupler via a coaxial cable (i.e., connected in parallel with the signal acquisition terminal, or connected via a power divider). The terminal under test is connected to the reconstructed channel environment, and the PC host computer synchronously collects communication performance data of the terminal under test, such as throughput, packet loss rate, handover success rate, and call drop rate, to complete the terminal performance test.

[0057] This embodiment provides a closed-loop channel restoration system based on a programmable attenuator and a signal simulator, which can simultaneously solve the problems of parameter decoupling and dynamic point-by-point reproduction, has controllable cost, and has real-time feedback calibration capability.

[0058] The above descriptions are merely embodiments of the present invention. Commonly known structures and characteristics of the solutions are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are aware of all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can, under the guidance of this application, improve and implement this solution in combination with their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent.

Claims

1. A closed-loop channel restoration system based on a programmable attenuator and a signal simulator, characterized in that, include: The signal source unit is used to output downlink radio frequency signals; A programmable attenuation unit, the input of which is connected to the output of the signal source unit, is used to programmably attenuate the downlink radio frequency signal according to a first control command; The noise injection unit has its output terminal connected to the output link of the programmable attenuation unit via a coupler, and is used to inject controllable noise into the link according to the second control command. The signal acquisition terminal is connected to the downlink and is used to acquire the measured values ​​of RSRP and SINR in real time. The closed-loop control unit is communicatively connected to the programmable attenuation unit, the noise injection unit, and the signal acquisition terminal, and is used to execute RSRP independent closed-loop control and SINR independent closed-loop control. Specifically, the closed-loop control unit first generates the first control command based on the difference between the target RSRP sequence acquired in the field and the measured RSRP value, using a PID negative feedback algorithm, and adjusts the programmable attenuation unit until the measured RSRP value converges to the target RSRP value; based on RSRP locking, it then generates the second control command based on the difference between the target SINR sequence acquired in the field and the measured SINR value, using a PID negative feedback algorithm, and adjusts the noise injection unit until the measured SINR value converges to the target SINR value; The above process is repeated point by point in chronological order to generate a time-aligned channel restoration control data table. Based on this data table, the programmable attenuation unit and the noise injection unit are controlled synchronously to continuously reproduce the dynamic changes of RSRP and SINR acquired in the field.

2. The closed-loop channel restoration system based on a programmable attenuator and a signal simulator according to claim 1, characterized in that, The closed-loop control unit includes a PC host computer and a real-time control module; the PC host computer is used for field data preprocessing, test process scheduling and result analysis; the real-time control module runs the PID negative feedback algorithm, and its control cycle is no more than 10ms.

3. The closed-loop channel restoration system based on a programmable attenuator and a signal simulator according to claim 1, characterized in that, The noise injection unit is a vector signal simulator; the coupler is a directional radio frequency coupler, whose main path is connected in series between the output of the programmable attenuation unit and the receiving end of the signal acquisition terminal, and whose coupling injection port is connected to the output of the noise injection unit, for realizing unidirectional combining of noise signal and downlink radio frequency signal.

4. The closed-loop channel restoration system based on a programmable attenuator and a signal simulator according to claim 1, characterized in that, The closed-loop control unit is also used to perform link inherent loss calibration before the first operation: control the signal source unit to output a continuous wave signal with a fixed power, obtain the received power collected by the signal acquisition terminal, calculate the link loss based on the difference between the fixed power and the received power, and use the link loss as the initial attenuation calculation benchmark for the PID negative feedback algorithm.

5. A closed-loop channel restoration system based on a programmable attenuator and a signal simulator according to claim 1, characterized in that, The PID negative feedback algorithm is an incremental PID algorithm. Its proportional coefficient Kp, integral coefficient Ki, and derivative coefficient Kd are tuned on-site so that the RSRP error converges to within ±1dBm within 3 to 5 iterations, and the SINR error converges to within ±1dB within 3 to 5 iterations.

6. A closed-loop channel restoration system based on a programmable attenuator and a signal simulator according to claim 1, characterized in that, The signal acquisition terminal uses the same cellular communication module as the field road test data acquisition module. Its sampling rate is the same as that of the field acquisition, and it outputs the measured RSRP and SINR values ​​in real time.

7. A closed-loop channel restoration system based on a programmable attenuator and a signal simulator according to claim 1, characterized in that, The signal source unit is a communication comprehensive test instrument or a commercial mobile communication base station equipment; the attenuation step of the programmable attenuation unit is no greater than 0.1dB, the attenuation accuracy is ±0.2dB, and the response time is no greater than 5ms.

8. A closed-loop channel restoration system based on a programmable attenuator and a signal simulator according to claim 1, characterized in that, When used for testing multimode communication systems, the system includes multiple independent signal source units, programmable attenuation units, and noise injection units, each corresponding to a different communication standard. The closed-loop control unit synchronously coordinates the independent dual closed-loop control of each communication standard to reproduce the signal changes in the multimode field environment.

9. A closed-loop channel restoration system based on a programmable attenuator and a signal simulator according to claim 1, characterized in that, The system adopts a modular architecture: in the laboratory full-function performance test scenario, it is equipped with a complete dual closed-loop architecture including a signal source unit, a programmable attenuation unit, a noise injection unit, a signal acquisition terminal, and a closed-loop control unit; In the field fixed-point semi-real vehicle test scenario, it is trimmed into a lightweight architecture that omits the noise injection unit.

10. A closed-loop channel restoration system based on a programmable attenuator and a signal simulator according to claim 1, characterized in that, The closed-loop control unit has a built-in standardized script instruction set to support fully automated testing and parallel scheduling of multiple test nodes. The fully automated testing includes channel restoration, parameter calibration, data acquisition, and report output.

Images