Satellite laser communication aiming time delay tolerant test method and system

Abstract

The present application relates to the technical field of satellite laser communication, and especially relates to a satellite laser communication aiming time delay tolerance testing method and system. The method constructs a reference loop and a test loop sharing the same error signal source, and the test loop is connected with an adjustable time delay injection module. A dynamic time delay disturbance sequence is applied, and the aiming angle responses of the two loops are collected. The aiming angle deviation residual error is calculated, and a divergence trajectory curve is constructed. According to the slope of the curve, the critical divergence time delay is determined as the upper limit of the aiming time delay tolerance. The method can accurately measure the aiming time delay tolerance capability of the satellite laser communication link.

Description

Technical Field

[0001] This invention relates to the field of satellite laser communication technology, and in particular to a method and system for testing satellite laser communication aiming delay tolerance. Background Technology

[0002] In satellite laser communication systems, the time delay characteristics of the aiming control loop directly affect the stability of link establishment and maintenance. Current conventional practices typically employ static time delay injection, setting several fixed time delay values ​​(e.g., 1ms, 5ms, 10ms) during testing and evaluating the system's tolerance to time delay by observing whether the aiming angle response exceeds the allowable range. Some solutions also set a fixed time delay safety margin based on human experience as a reference for engineering design. This testing method largely relies on offline simulation or simplified mathematical models, lacking simulation of time-varying disturbance scenarios in real links.

[0003] Static time delay injection cannot reflect the transition characteristics of the system response from stability to instability when time delay changes dynamically. It can only obtain a "pass / fail" judgment at discrete points, making it difficult to capture the true boundary of the system's critical divergence. On the other hand, the time delay tolerance upper limit set based on experience lacks quantitative evaluation basis. The actual tolerance capability varies significantly under different satellite platforms and orbital conditions. Fixed thresholds can easily lead to excessive design redundancy or insufficient risk assessment. Existing testing methods are unable to accurately identify the divergence start time of the aiming loop under dynamic continuous time delay disturbances, thus failing to provide reproducible time delay tolerance test results with clear physical meaning. Summary of the Invention

[0004] The present invention provides a method and system for testing satellite laser communication aiming delay tolerance, which can solve the problems in the prior art.

[0005] A first aspect of the present invention provides a method for testing satellite laser communication aiming delay tolerance, comprising: A first aiming control loop and a second aiming control loop are constructed to form a satellite laser communication link. The first aiming control loop and the second aiming control loop share the same error signal source output by the aiming error detection unit. The first aiming control loop transmits the error signal directly to its aiming actuator as a reference loop. The second aiming control loop connects an adjustable time delay injection module in series in the error signal transmission path as a test loop. During the testing phase, a dynamic delay perturbation sequence that changes continuously with time is applied to the adjustable delay injection module. The delay value of the dynamic delay perturbation sequence gradually increases from zero to a preset upper limit value according to a preset perturbation mode. Simultaneously, the aiming angle response of the reference circuit and the aiming angle response of the test circuit are collected. Calculate the aiming angle deviation residual between the reference loop and the test loop at each sampling time, construct the divergence trajectory curve of the aiming angle deviation residual as a function of time, and when the slope of the divergence trajectory curve exceeds the preset divergence threshold and the residual of multiple subsequent sampling times maintains a monotonically increasing trend, record the injection delay value at the corresponding time as the critical divergence delay, use the critical divergence delay as the upper limit of the aiming delay tolerance of the laser communication link, and output the tolerance test result.

[0006] Constructing the first aiming control loop and the second aiming control loop includes: Obtain the output port parameters and signal type parameters of the aiming error detection unit of the laser communication link; A signal splitter is provided at the output port. The signal splitter replicates the error signal source output by the aiming error detection unit into a first error signal and a second error signal. The amplitude and phase characteristics of the two error signals are completely consistent. The first error signal is connected to the aiming actuator of the first aiming control loop to form a reference loop, and the second error signal is connected to the input of the adjustable delay injection module. The output of the adjustable delay injection module is connected to the aiming actuator of the second aiming control loop to form a test loop. The reference circuit and the test circuit are initially zero-point calibrated. The total delay difference between the two circuits from the error signal input to the aiming angle response is measured without injection delay. The total delay difference is used as the reference zero-point offset. When calculating the aiming angle deviation residual at each subsequent sampling time, the angle compensation amount corresponding to the reference zero-point offset is subtracted from the aiming angle response of the test circuit, and then compared with the aiming angle response of the reference circuit to calculate the difference.

[0007] Initial zero-point calibration of the signal transmission delay between the reference circuit and the test circuit includes: When the injection delay value of the adjustable delay injection module is set to zero, a preset calibration excitation signal is input to the aiming error detection unit. The preset calibration excitation signal includes a step component and a sinusoidal sweep frequency component to cover the full-band response characteristics of the aiming control loop. The time sequence of the first angle response output by the aiming actuator of the reference loop and the time sequence of the second angle response output by the aiming actuator of the test loop are recorded synchronously. Calculate the cross-correlation function between the first angle response time series and the second angle response time series, and use the peak offset of the cross-correlation function as the total delay difference; When the delay difference corresponding to the peak offset of the cross-correlation function is greater than the preset delay tolerance, the hardware transmission path of the reference circuit and the test circuit is adjusted for physical delay matching until the total delay difference falls within the preset delay tolerance range, and the final adjusted total delay difference is used as the reference zero point offset. Before the start of each test phase, the initial zero-point calibration is re-performed to obtain the reference zero-point offset for that test, ensuring that the aiming angle deviation residuals of different test batches are repeatable.

[0008] A dynamic delay perturbation sequence that varies continuously with time is applied to the adjustable delay injection module. The delay value of the dynamic delay perturbation sequence gradually increases from zero to a preset upper limit value according to a preset perturbation mode, including: The spectral coverage of the dynamic delay perturbation sequence is determined based on the aiming control bandwidth of the laser communication link, so that the main lobe width of the dynamic delay perturbation sequence covers a preset multiple of the aiming control bandwidth, thereby fully stimulating the dynamic response characteristics of the aiming control loop under different frequency delay perturbations. A composite perturbation waveform is generated, which is composed of a linearly increasing fundamental component and a multi-frequency sinusoidal superposition component. The linearly increasing fundamental component provides a global time delay increasing trend, and the multi-frequency sinusoidal superposition component superimposes time-varying perturbations of different frequencies on the linearly increasing fundamental component to simulate the multi-source time delay coupling effect caused by atmospheric turbulence in the space laser link. The discrete sampling sequence of the composite disturbance waveform is used as the dynamic time delay disturbance sequence. At each sampling time, the corresponding sampling value is written into the injection time delay register of the adjustable time delay injection module, and the injection time delay value of the adjustable time delay injection module is controlled to change continuously according to the composite disturbance waveform. During the increasing process of the dynamic time delay disturbance sequence, the tracking deviation between the actual output time delay of the adjustable time delay injection module and the set injection time delay is monitored synchronously. When the tracking deviation exceeds the preset tracking tolerance, the dynamic time delay disturbance sequence is adaptively corrected in amplitude.

[0009] Generate composite disturbance waveforms, including: The linearly increasing fundamental component is set as a linear ramp function with a time delay value from zero to the preset upper limit value. The slope of the linear ramp function is determined by the ratio of the preset upper limit value to the preset test duration. A multi-frequency sinusoidal superposition component is constructed. The multi-frequency sinusoidal superposition component is composed of a preset number of sinusoidal components with coprime frequencies and different amplitudes, and the frequency of each sinusoidal component is selected within a specified octave range of the aiming control bandwidth. The amplitude of each sinusoidal component is proportional to the square root of the current time delay value of the linear ramp function. The linear ramp function and the multi-frequency sinusoidal superposition component are added point by point to obtain the discrete sampling sequence of the composite disturbance waveform. The discrete sampling sequence is then subjected to anti-aliasing smoothing filtering and output as the dynamic time delay disturbance sequence. The frequency composition and amplitude weight of each sinusoidal component in the multi-frequency sinusoidal superposition component are adjusted according to the actual operating environment parameters of the laser communication link, so that the spectral characteristics of the dynamic time delay disturbance sequence match the time delay statistical distribution characteristics of the target space link environment. The frequency composition and amplitude weight are mapped to the generation parameters of each sinusoidal component through a parameter lookup table and adjusted in real time.

[0010] Construct a divergence trajectory curve of the aiming angle deviation residual over time. When the slope of the divergence trajectory curve exceeds a preset divergence threshold and the residuals at subsequent consecutive sampling times maintain a monotonically increasing trend, record the critical divergence delay, including: The divergence trajectory curve is constructed by plotting the sampling time on the horizontal axis and the difference between the aiming angle response of the reference circuit and the aiming angle response of the test circuit at each sampling time on the vertical axis. The divergence trajectory curve is subjected to piecewise sliding linear fitting. A sliding window with a preset window width moves along the divergence trajectory curve point by point. At each window position, the slope of the fitted straight line of all sampling points in the window is calculated as the local divergence slope, forming a local divergence slope sequence distributed along the time axis. An adaptive threshold decision is made on the local divergence slope sequence. The preset divergence threshold is dynamically adjusted according to the standard deviation of the residual of the stationary segment. When the local divergence slope exceeds the adaptive threshold for the first time and the local divergence slope of the subsequent preset number of sampling points continues to be above the adaptive threshold and the aiming angle deviation residual maintains a monotonically increasing trend, the sampling point that first exceeds the threshold is identified as the divergence start point, and the injection delay value at the time corresponding to the divergence start point is used as the critical divergence delay. The critical divergence delay is compared with the preset design delay tolerance of the spaceborne laser communication terminal, and the upper limit of the aiming delay tolerance and the judgment result of whether the design tolerance is met are output.

[0011] Constructing the divergence trajectory curve and determining the critical divergence delay includes: The curve of the time delay value of the dynamic time delay disturbance sequence changing with time is superimposed on the divergence trajectory curve as an auxiliary reference axis. The corresponding relationship between the divergence process of the aiming angle deviation residual and the growth process of the injected time delay value is presented in the same time coordinate system in the form of dual vertical axes. The divergence trajectory curve is segmented into three feature segments: a stable segment, a transition segment, and a divergence segment. The stable segment corresponds to the gradually changing region of the aiming angle deviation residual, the transition segment corresponds to the acceleration region where the aiming angle deviation residual begins to deviate from the random fluctuation range, and the divergence segment corresponds to the out-of-control region where the aiming angle deviation residual shows a continuous exponential growth trend. The injection delay value corresponding to the boundary point between the transition segment and the divergence segment is used as a candidate value for the critical divergence delay. The candidate value is cross-validated with the critical divergence delay corresponding to the actual divergence start point, and the smaller of the two is taken as the final critical divergence delay. When the difference between the candidate value and the critical divergence delay corresponding to the actual divergence start point exceeds the preset divergence threshold, the injection delay value of the preset protection margin before the divergence segment start time is taken as the final critical divergence delay.

[0012] A second aspect of the present invention provides a satellite laser communication aiming delay tolerance testing system, comprising: A dual-loop construction unit is used to construct a first aiming control loop and a second aiming control loop for a satellite laser communication link. The first aiming control loop and the second aiming control loop share the same error signal source output by the aiming error detection unit. The first aiming control loop transmits the error signal directly to its aiming actuator as a reference loop, and the second aiming control loop connects an adjustable time delay injection module in series in the error signal transmission path as a test loop. The time delay perturbation injection unit is used to apply a dynamic time delay perturbation sequence that changes continuously with time to the adjustable time delay injection module during the testing phase. The time delay value of the dynamic time delay perturbation sequence gradually increases from zero to a preset upper limit value according to a preset perturbation mode, and the aiming angle response of the reference circuit and the aiming angle response of the test circuit are collected simultaneously. The critical delay determination unit is used to calculate the aiming angle deviation residual between the reference loop and the test loop at each sampling time, construct a divergence trajectory curve of the aiming angle deviation residual changing with time, and when the slope of the divergence trajectory curve exceeds a preset divergence threshold and the residual of multiple subsequent consecutive sampling times maintains a monotonically increasing trend, the injection delay value at the corresponding time is recorded as the critical divergence delay, and the critical divergence delay is used as the upper limit of the aiming delay tolerance of the laser communication link, and the tolerance test result is output.

[0013] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0015] This method effectively avoids the shortcomings of traditional static delay testing, which cannot reflect the dynamic characteristics of the link. By using a continuous dynamic delay disturbance sequence, it accurately simulates the time-varying accumulation process of aiming errors in actual communication, making the determination of the delay tolerance upper limit more consistent with the on-orbit working scenario. The design of a shared error signal source eliminates the device differences and noise interference introduced by different detection units. The structure of the reference circuit and the test circuit naturally has the ability to cancel common-mode errors, which greatly improves the reliability and repeatability of the test results.

[0016] The combined logic of the aiming angle deviation residual trajectory curve, the divergence slope threshold, and the monotonically increasing criterion avoids the risk of single-point misjudgment and ensures that the locking of the critical divergence delay has statistically significant robustness. This method does not require additional high-precision inertial measurement or complex optical path modification; it can complete the end-to-end tolerance calibration simply through software-based delay injection, significantly reducing the hardware cost and implementation complexity of the test system.

[0017] During the test, the disturbance mode with the delay value continuously increasing from zero completely covered the entire state region from stable tracking to divergence and loss of lock. It directly outputs the quantized tolerance upper limit characterized by the critical divergence delay, providing a clear engineering boundary for the design of servo controllers in laser communication systems, the allocation of delay budgets for communication protocols, and the assessment of fault tolerance. This test process can be embedded into satellite on-orbit autonomous maintenance or ground-based joint testing processes, enabling rapid retesting and dynamic calibration of delay tolerance characteristics, and improving the adaptive capability of satellite laser communication networks to the slow degradation of the space environment. Attached Figure Description

[0018] Figure 1 A flowchart illustrating the testing method for aiming delay tolerance in satellite laser communication. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0021] refer to Figure 1 The satellite laser communication aiming delay tolerance test method of this invention includes: A first aiming control loop and a second aiming control loop are constructed to form a satellite laser communication link. The first aiming control loop and the second aiming control loop share the same error signal source output by the aiming error detection unit. The first aiming control loop transmits the error signal directly to its aiming actuator as a reference loop. The second aiming control loop connects an adjustable time delay injection module in series in the error signal transmission path as a test loop. During the testing phase, a dynamic delay perturbation sequence that changes continuously with time is applied to the adjustable delay injection module. The delay value of the dynamic delay perturbation sequence gradually increases from zero to a preset upper limit value according to a preset perturbation mode. Simultaneously, the aiming angle response of the reference circuit and the aiming angle response of the test circuit are collected. Calculate the aiming angle deviation residual between the reference loop and the test loop at each sampling time, construct the divergence trajectory curve of the aiming angle deviation residual as a function of time, and when the slope of the divergence trajectory curve exceeds the preset divergence threshold and the residual of multiple subsequent sampling times maintains a monotonically increasing trend, record the injection delay value at the corresponding time as the critical divergence delay, use the critical divergence delay as the upper limit of the aiming delay tolerance of the laser communication link, and output the tolerance test result.

[0022] Constructing the first aiming control loop and the second aiming control loop includes: Obtain the output port parameters and signal type parameters of the aiming error detection unit of the laser communication link; A signal splitter is provided at the output port. The signal splitter replicates the error signal source output by the aiming error detection unit into a first error signal and a second error signal. The amplitude and phase characteristics of the two error signals are completely consistent. The first error signal is connected to the aiming actuator of the first aiming control loop to form a reference loop, and the second error signal is connected to the input of the adjustable delay injection module. The output of the adjustable delay injection module is connected to the aiming actuator of the second aiming control loop to form a test loop. The reference circuit and the test circuit are initially zero-point calibrated. The total delay difference between the two circuits from the error signal input to the aiming angle response is measured without injection delay. The total delay difference is used as the reference zero-point offset. When calculating the aiming angle deviation residual at each subsequent sampling time, the angle compensation amount corresponding to the reference zero-point offset is subtracted from the aiming angle response of the test circuit, and then compared with the aiming angle response of the reference circuit to calculate the difference.

[0023] The process of constructing the first and second aiming control loops begins with obtaining the output port parameters and signal type parameters of the aiming error detection unit in the laser communication link. The aiming error detection unit is typically implemented using a four-quadrant detector or a position-sensitive detector (PSD). Its output port parameters include interface type (analog differential output or digital interface), output impedance, full-scale voltage range, and sampling bandwidth. Signal type parameters include the polarity definition, linear range, and noise floor level of the error signal. Obtaining these parameters provides a matching basis for the subsequent design of the signal splitter, ensuring that the two split signals are electrically identical to the original error signal, and avoiding additional signal distortion due to impedance mismatch or insufficient drive capability.

[0024] After obtaining the port parameters and signal type parameters, a signal splitter is set at the output port of the aiming error detection unit. The core function of the signal splitter is to replicate the single-channel error signal source without distortion into a first-channel error signal and a second-channel error signal, ensuring that the amplitude and phase characteristics of the two outputs are completely consistent. In engineering implementation, the signal splitter typically employs a buffer amplifier structure with high input impedance and low output impedance, based on an operational amplifier voltage follower topology. The two outputs are driven by symmetrical wiring to eliminate phase inconsistencies caused by differences in trace length. For analog differential signals, the splitter must simultaneously replicate both the positive and negative differential signals, ensuring that the common-mode rejection ratio of the two outputs is not lower than the specifications of the original signal. The bandwidth of the splitter should cover at least five times the open-loop bandwidth of the aiming control loop to ensure that the splitting operation within the control bandwidth does not introduce perceptible amplitude or phase frequency distortion. After the hardware installation of the splitter is completed, the static electrical characteristics of the two output signals need to be measured to confirm that the amplitude deviation does not exceed 0.1% of the full scale and the phase deviation does not exceed 0.5° before proceeding to the subsequent connection steps.

[0025] The first error signal is directly connected to the aiming actuator of the first aiming control loop to form a reference loop. In the reference loop, the path of the error signal from the splitter output to the aiming actuator drive input does not introduce any artificial time delays, except for necessary signal conditioning circuits (such as filters and gain stages). The second error signal is connected to the input of an adjustable time delay injection module, and the output of the adjustable time delay injection module is then connected to the aiming actuator drive input of the second aiming control loop to form a test loop. The adjustable time delay injection module can be implemented in hardware using a digital signal processor (DSP) or a field-programmable gate array (FPGA). It utilizes a first-in-first-out (FIFO) buffer queue to perform controllable depth delay readout of the digitized error signal, with a time delay step accuracy reaching the microsecond level, meeting the time delay resolution requirements in satellite laser communication aiming control. The two aiming actuators should be consistent in model, drive parameters, and installation method to ensure that the two loops have the same mechanical response characteristics under the same error signal drive.

[0026] After completing the hardware connection of the two loops, initial zero-point calibration is performed on the reference loop and the test loop. The calibration process is conducted with the adjustable delay injection module's delay value set to zero, meaning the test loop does not introduce any additional artificial delay. Under this condition, a set of standard step error signals is synchronously injected into both loops. The total time from the error signal input moment to the moment the aiming angle response reaches a stable output moment is collected for both the reference loop and the test loop. The difference between the two is calculated, and this difference is defined as the reference zero-point offset, denoted as [missing value]. Reference zero offset The reason for this is that, although the two error signals are consistent at the output of the splitter, the hardware processing links of the adjustable delay injection module in the test loop (including analog-to-digital conversion, FIFO read / write, digital-to-analog conversion, etc.) still introduce inherent processing delays even under zero-delay configuration. These inherent delays differ from the transmission delay of the pure analog path in the reference loop, thus causing the error. .

[0027] Reference zero offset Once determined, it needs to be converted into the corresponding angle compensation amount for correction in subsequent residual calculations. Because the two-loop aiming actuators are in... The test loop will continuously respond to the error signal over the time period, therefore the aiming angle response of the test loop relative to the reference loop will have a certain difference. The corresponding angular offset is denoted as . The specific values ​​are obtained by sampling and averaging the aiming angle responses of the two loops multiple times under zero injection delay conditions, and the average value is stored as a fixed compensation amount for later use. When calculating the aiming angle deviation residual at subsequent sampling times, it is first subtracted from the aiming angle response of the test loop. Then, the aiming angle response of the reference loop is compared and the difference is calculated to obtain the aiming angle deviation residual caused purely by the injected time delay. This compensation operation makes the residual calculation results of the two loops statistically close to zero when the time delay injection amount is zero, thus providing a clean reference starting point for judging the residual divergence in the subsequent time delay increase process.

[0028] In actual engineering deployment, Measurements should be performed under the same temperature and power supply conditions as the formal test to eliminate the influence of environmental factors on inherent time delay. It is recommended to perform measurements before the formal test. Perform at least 10 independent measurements and calculate the standard deviation. If the standard deviation exceeds 5% of the aiming accuracy specification, check the clock stability of the adjustable delay injection module and the aperture jitter of the analog-to-digital converter. After ruling out hardware malfunctions, recalibrate. Complete calibration and confirm. Once stabilized, the dual-loop construction of the reference and test loops was completed, meeting the requirements for entering the dynamic time delay disturbance injection test phase. The entire dual-loop construction process, through rigorous signal splitting, symmetrical hardware design, and precise calibration of the reference zero-point offset, ensured the validity and repeatability of subsequent aiming delay tolerance test results.

[0029] Initial zero-point calibration of the signal transmission delay between the reference circuit and the test circuit includes: When the injection delay value of the adjustable delay injection module is set to zero, a preset calibration excitation signal is input to the aiming error detection unit. The preset calibration excitation signal includes a step component and a sinusoidal sweep frequency component to cover the full-band response characteristics of the aiming control loop. The time sequence of the first angle response output by the aiming actuator of the reference loop and the time sequence of the second angle response output by the aiming actuator of the test loop are recorded synchronously. Calculate the cross-correlation function between the first angle response time series and the second angle response time series, and use the peak offset of the cross-correlation function as the total delay difference; When the delay difference corresponding to the peak offset of the cross-correlation function is greater than the preset delay tolerance, the hardware transmission path of the reference circuit and the test circuit is adjusted for physical delay matching until the total delay difference falls within the preset delay tolerance range, and the final adjusted total delay difference is used as the reference zero point offset. Before the start of each test phase, the initial zero-point calibration is re-performed to obtain the reference zero-point offset for that test, ensuring that the aiming angle deviation residuals of different test batches are repeatable.

[0030] Before performing the aiming delay tolerance test, an initial zero-point calibration of the signal transmission delay between the reference loop and the test loop is required to eliminate the inherent delay bias introduced by the hardware path differences between the two loops. This calibration process is performed with the injection delay value of the adjustable delay injection module set to zero, meaning that no artificial delay disturbance is introduced into the test loop, and the two loops should theoretically have completely identical signal transmission characteristics. If an observable response difference still exists between the two loops at this point, it indicates that there is an inherent delay asymmetry in the hardware path itself, which must be quantified and compensated for before formal testing.

[0031] The design of the calibration excitation signal is a crucial step in zero-position calibration. The preset calibration excitation signal input to the aiming error detection unit consists of two components: a step component and a sinusoidal sweep component. The step component excites the transient response characteristics of the aiming control loop, producing a distinct rising edge characteristic in the time domain, facilitating the capture of the time offset between the two loop responses in subsequent cross-correlation calculations. The sinusoidal sweep component continuously scans from low to high frequencies, covering the full-band response characteristics of the aiming control loop, ensuring that the calibration results reflect not only the delay differences in the DC or low-frequency bands but also the transmission delay characteristics in the high-frequency dynamic response. The superposition of the two components gives the calibration excitation signal rich information in both the time and frequency domains, thus ensuring that the peak value of the cross-correlation function has sufficient resolution and statistical reliability. The frequency range of the sinusoidal sweep component should be preset according to the bandwidth parameters of the actual aiming control loop, typically covering from one-tenth of the lower limit of the loop bandwidth to twice the upper limit of the loop bandwidth, to fully excite the transmission delay characteristics of the loop at different frequencies.

[0032] While applying the calibration excitation signal, the time series of the first angle response output by the aiming actuator of the reference loop and the time series of the second angle response output by the aiming actuator of the test loop are recorded simultaneously. The two acquisition channels must be strictly synchronized, and the sampling clocks must come from the same clock source to avoid introducing additional time deviations at the acquisition end. The length of the acquired time series should be sufficient to cover the complete frequency sweep process and the steady-state portion of the step response, ensuring that the calculation of the cross-correlation function has a sufficient data basis. Let the first angle response time series be... The second angle response time series is ,in This is an index for discrete sampling times.

[0033] calculate and Cross-correlation function between Its definition is: in This represents the cross-correlation delay offset, expressed in units of sampling points. When When the peak value is reached, the corresponding This is the estimated time offset between the two signals. Converting this peak offset to actual time units yields the total delay difference between the two loops. If the delay difference corresponding to the peak offset falls within the preset delay tolerance range, the transmission delays of the two loops are considered sufficiently matched, and this difference can be directly recorded as the reference zero-point offset for subsequent testing.

[0034] If the delay difference corresponding to the peak offset of the cross-correlation function is greater than the preset delay tolerance, physical delay matching adjustment is required for the hardware transmission paths of the reference loop and the test loop. Physical delay matching adjustment can be achieved through various means: for fiber optic transmission paths, the propagation delay of the optical signal can be fine-tuned by increasing or decreasing the length of the fiber optic patch cord; for electrical signal transmission paths, delay differences can be compensated by adjusting the PCB trace length or inserting fixed delay lines; for digital signal processing links, the digital processing delay can be changed by adjusting the data buffer depth or the number of pipeline stages. After adjustment, the calibration excitation signal injection and cross-correlation function calculation are re-executed to verify whether the total delay difference has fallen within the preset delay tolerance range. The above adjustment and verification process is iterated until the total delay difference meets the requirements. The final adjusted total delay difference is determined as the reference zero-point offset. This value will be deducted as an inherent bias in the aiming angle deviation residual calculation in subsequent testing phases to ensure that the residual result only reflects the impact of the injected delay disturbance and does not include system errors caused by hardware path inconsistencies.

[0035] The setting of the preset delay tolerance needs to be comprehensively considered in conjunction with the dynamic characteristics of the aiming control loop. Generally speaking, the delay tolerance should be much smaller than the critical settling delay of the aiming control loop, usually within 5% of the estimated critical settling delay, to ensure that the residual error of the reference zero-point offset does not significantly affect the judgment of the critical divergence delay. If the delay tolerance is set too wide, the uncertainty of the reference zero-point offset will be directly superimposed on the residual curve, resulting in a large system bias in the initial segment of the divergence trajectory curve, affecting the accurate identification of the critical divergence delay.

[0036] Before each testing phase, the initial zero-point calibration procedure described above must be repeated to obtain the reference zero-point offset for that test. This requirement is necessary because the latency characteristics of the hardware transmission path in a satellite laser communication system may drift between different test batches due to factors such as temperature changes, mechanical vibration, or electronic component aging. If calibration is only performed before the first test, subsequent batches will suffer from systematic errors due to hardware latency drift, making the residual aiming angle deviations between different batches incomparable. By re-calibrating before each test, the current state of hardware latency can be captured in real time, ensuring that the reference zero-point offset for each test reflects the current hardware conditions. This guarantees the repeatability and horizontal comparability of the residual data across different test batches, providing a consistent data basis for the statistical analysis of the upper limit of aiming latency tolerance.

[0037] In practical engineering implementation, the execution time of the calibration process should be minimized to reduce test preparation time. A pre-optimized calibration excitation signal length can be used, ensuring the peak resolution of the cross-correlation function while controlling the duration of the calibration excitation signal to within a few times the aiming control loop time constant. Furthermore, the calculation of the cross-correlation function can be accelerated using Fast Fourier Transform, significantly reducing computation time and enabling the entire calibration process to be completed quickly during the test preparation phase without affecting test efficiency.

[0038] A dynamic delay perturbation sequence that varies continuously with time is applied to the adjustable delay injection module. The delay value of the dynamic delay perturbation sequence gradually increases from zero to a preset upper limit value according to a preset perturbation mode, including: The spectral coverage of the dynamic delay perturbation sequence is determined based on the aiming control bandwidth of the laser communication link, so that the main lobe width of the dynamic delay perturbation sequence covers a preset multiple of the aiming control bandwidth, thereby fully stimulating the dynamic response characteristics of the aiming control loop under different frequency delay perturbations. A composite perturbation waveform is generated, which is composed of a linearly increasing fundamental component and a multi-frequency sinusoidal superposition component. The linearly increasing fundamental component provides a global time delay increasing trend, and the multi-frequency sinusoidal superposition component superimposes time-varying perturbations of different frequencies on the linearly increasing fundamental component to simulate the multi-source time delay coupling effect caused by atmospheric turbulence in the space laser link. The discrete sampling sequence of the composite disturbance waveform is used as the dynamic time delay disturbance sequence. At each sampling time, the corresponding sampling value is written into the injection time delay register of the adjustable time delay injection module, and the injection time delay value of the adjustable time delay injection module is controlled to change continuously according to the composite disturbance waveform. During the increasing process of the dynamic time delay disturbance sequence, the tracking deviation between the actual output time delay of the adjustable time delay injection module and the set injection time delay is monitored synchronously. When the tracking deviation exceeds the preset tracking tolerance, the dynamic time delay disturbance sequence is adaptively corrected in amplitude.

[0039] Before determining the design scheme for the dynamic delay perturbation sequence, it is necessary to obtain the control bandwidth parameters of the laser communication link's aiming control loop. The aiming control bandwidth reflects the loop's response capability to angle error signals, and its value directly determines the frequency range that the perturbation sequence needs to cover. If the main lobe width of the perturbation sequence is insufficient to cover the control bandwidth, the delay-sensitive characteristics in some frequency bands will not be fully activated, leading to a conservative or even distorted estimate of the critical divergence delay. Therefore, based on the aiming control bandwidth... To determine the spectral coverage of a dynamic time-delay perturbation sequence, the main lobe width of the spectral spectrum of the dynamic time-delay perturbation sequence is required. satisfy ,in The preset multiplier coefficient typically ranges from 1.5 to 3.0, with the specific value determined based on the actual operating conditions of the link and the required test accuracy. This constraint ensures that the disturbance sequence can fully stimulate the dynamic response characteristics of the aiming control loop under different frequency delay disturbances, thus ensuring the frequency domain completeness of the test results.

[0040] The composite disturbance waveform is composed of two superimposed components. The linearly increasing fundamental component provides the global time delay increasing trend, with its time delay value starting from zero and increasing at a fixed slope. (Unit: nanoseconds per second) Increases monotonically with time until it reaches a preset upper limit. At that moment At that point, the time delay value of the linear fundamental component is This component ensures that the injected delay exhibits a continuously increasing trend at the macroscopic level throughout the entire test process, providing a monotonically traceable time axis reference for subsequent location of the critical divergence delay. The multi-frequency sinusoidal superposition component superimposes several sinusoidal time-varying perturbations of different frequencies onto the linear fundamental wave to simulate the multi-source delay coupling effect caused by atmospheric turbulence in the space laser link. The expression for the multi-frequency sinusoidal superposition component is: ,in To represent the number of superimposed sinusoidal components, For the first The amplitude of each sinusoidal component, For the first The frequency of a sinusoidal component For the first The initial phase of each sinusoidal component. The selected frequency components should be evenly distributed within the aiming control bandwidth to ensure coverage of various time-delay coupling modes such as low-frequency drift, mid-frequency oscillation, and high-frequency jitter. Ultimately, the time delay value of the composite disturbance waveform is... While maintaining an overall increasing trend, the waveform also incorporates rich multi-frequency time-varying components, enabling a comprehensive examination of the stability boundary of the aiming control loop under complex time-delay excitation.

[0041] Discrete sampling of the composite disturbance waveform yields a dynamic time-delay disturbance sequence. The sampling time is based on the sampling period. Divide into equal intervals, the first The injection delay value corresponding to each sampling time is At each sampling time, The injection delay register of the adjustable delay injection module is written to, and the driving module continuously updates its actual output delay according to the composite disturbance waveform. The write operation of the injection delay register is strictly synchronized with the sampling clock to ensure that each write operation is completed on the rising edge of the clock at the corresponding sampling time, avoiding additional unexpected delay errors introduced by write timing jitter. After receiving a new register value, the adjustable delay injection module smoothly transitions the output delay to the target value through internal interpolation or a pass-through switching mechanism. The transition time should be controlled within one sampling period. Within this range, to ensure the continuity of time delay changes.

[0042] During the increasing dynamic delay perturbation sequence, it is necessary to synchronously monitor the following deviation between the actual output delay of the adjustable delay injection module and the set injection delay. Actual output delay Real-time readings are obtained through the delay measurement unit, and the injection delay is set to... Follower deviation is defined as .when Exceeding the preset following tolerance When this happens, an adaptive amplitude correction mechanism is triggered. The core idea of ​​adaptive amplitude correction is: when the following deviation is too large, it indicates that the current delay increment rate or the multi-frequency superposition amplitude exceeds the response capability of the adjustable delay injection module, requiring compression and adjustment of the amplitude of subsequent disturbance sequences. Specifically, the corrected injection delay sequence is then used in subsequent... The amplitude of the multi-frequency sinusoidal superposition component is proportionally reduced within each sampling point, with a reduction factor. It is determined by the ratio of following deviation to following tolerance. The corrected amplitude of the multi-frequency superposition component becomes During the correction process, the slope of the linearly increasing fundamental component... The amplitude of the superimposed high-frequency oscillation component is compressed to ensure that the global latency increase trend remains unchanged, thus ensuring test coverage while taking into account the actual tracking capability of the module.

[0043] When the following deviation is in continuous All sampling points have recovered to When the amplitude correction ends, the amplitude of the multi-frequency sinusoidal superposition component gradually recovers to its original set value. The recovery process employs linear interpolation to avoid additional transient responses caused by abrupt amplitude changes. The entire adaptive correction mechanism operates continuously in a closed-loop manner throughout the entire process of the applied perturbation sequence, ensuring that the adjustable delay injection module always operates within its effective tracking range, while also guaranteeing that the spectral characteristics and increasing trend of the dynamic delay perturbation sequence meet the test design requirements.

[0044] In practical engineering implementation, the bit width of the injected delay register needs to be based on a preset upper limit value. Together with the time delay resolution, this ensures that the register can represent the time delay value across the entire range without truncation. The frequency of the multi-frequency sinusoidal superposition component... When selecting frequencies, integer multiples of harmonics should be avoided to prevent frequency aliasing, which could lead to excessive concentration of excitation energy in certain frequency bands and affect the uniformity of spectral coverage. (Following tolerance) The settings should comprehensively consider the measurement accuracy of the delay measurement unit and the nominal following error of the adjustable delay injection module. Typically, 1.5 times the sum of the two is taken as an engineering margin to avoid false triggering corrections due to measurement noise. Through the aforementioned complete disturbance sequence generation and adaptive correction mechanism, dynamic delay disturbances conforming to design requirements can be stably and accurately applied to the test loop throughout the entire test cycle, providing a reliable excitation basis for the accurate identification of subsequent critical divergence delays.

[0045] Generate composite disturbance waveforms, including: The linearly increasing fundamental component is set as a linear ramp function with a time delay value from zero to the preset upper limit value. The slope of the linear ramp function is determined by the ratio of the preset upper limit value to the preset test duration. A multi-frequency sinusoidal superposition component is constructed. The multi-frequency sinusoidal superposition component is composed of a preset number of sinusoidal components with coprime frequencies and different amplitudes, and the frequency of each sinusoidal component is selected within a specified octave range of the aiming control bandwidth. The amplitude of each sinusoidal component is proportional to the square root of the current time delay value of the linear ramp function. The linear ramp function and the multi-frequency sinusoidal superposition component are added point by point to obtain the discrete sampling sequence of the composite disturbance waveform. The discrete sampling sequence is then subjected to anti-aliasing smoothing filtering and output as the dynamic time delay disturbance sequence. The frequency composition and amplitude weight of each sinusoidal component in the multi-frequency sinusoidal superposition component are adjusted according to the actual operating environment parameters of the laser communication link, so that the spectral characteristics of the dynamic time delay disturbance sequence match the time delay statistical distribution characteristics of the target space link environment. The frequency composition and amplitude weight are mapped to the generation parameters of each sinusoidal component through a parameter lookup table and adjusted in real time.

[0046] The generation of the composite perturbation waveform is the core step in constructing the dynamic time-delay perturbation sequence, and its quality directly determines the reliability and coverage of the aiming time-delay tolerance test results. The linearly increasing fundamental component, serving as the skeleton of the composite waveform, is set to linearly increase from zero to a preset upper limit value. The slope function. Depend on With preset test duration The ratio is determined, that is The unit is nanoseconds per second. This slope setting ensures that the delay value increases smoothly and monotonically from zero to the upper limit throughout the entire test cycle, making the delay load borne by the aiming control loop at each moment verifiable, facilitating the subsequent precise mapping of critical divergence moments to corresponding injection delay values. The linear ramp function is expressed in the continuous time domain as follows: ,in For continuous time variables, .

[0047] Constructing multi-frequency sinusoidal superposition components is a crucial step in endowing composite disturbance waveforms with wideband excitation characteristics. The frequencies of each sinusoidal component must be coprime to avoid periodic in-phase superposition between different components, which could lead to abnormally high local time delay peaks and interfere with the continuity of test results. The frequencies of each sinusoidal component should be within the target control bandwidth. Selected within a specified octave range, specifically, each frequency ( )satisfy ,in and These are the lower and upper bounds of the specified frequency range, respectively, which are preset by the test scheme. The reason for constraining the frequency to near the aiming control bandwidth is that time delay disturbances in this band are most sensitive to the phase margin of the aiming control loop, and can effectively excite the instability boundary of the control loop within a short test duration.

[0048] Amplitude of each sine component It is proportional to the square root of the current time delay value of the linear ramp function, that is... ,in For the first The amplitude weighting coefficients of each sinusoidal component are determined by a parameter lookup table mapping. This represents the linear ramp delay value at the current moment. The physical basis for using the square root relationship is that, in a real space link environment, the amplitude of link delay jitter usually increases with the increase of the total link delay, but the rate of increase gradually slows down. The square root function can precisely approximate this statistical law, making the amplitude envelope of the composite disturbance waveform consistent with the delay statistical distribution characteristics of the real space link, thereby improving the representativeness of the test results to the actual working conditions.

[0049] Multi-frequency sinusoidal superposition components The continuous time expression is ,in For the first The initial phase of each sinusoidal component is determined by a pseudo-random method. The distribution within the range is further reduced to decrease the regularity correlation between components. The continuous-time expression of the composite disturbance waveform is as follows: By directly adding the linear ramp component and the multi-frequency sinusoidal superposition component, a composite time-delay excitation waveform with both monotonically increasing trend and wideband oscillation characteristics is obtained.

[0050] In the discretization stage, with the sampling period Uniform sampling of the composite perturbation waveform yields a discrete sampling sequence. ,in This provides an index for discrete sampling times. The sampling frequency must satisfy the Nyquist criterion, meaning the sampling frequency must not be lower than the highest frequency component of the composite disturbance waveform. To prevent high-frequency component distortion caused by spectral aliasing, the filter is doubled in size. After uniform sampling, an anti-aliasing smoothing filter is applied to the discrete sampling sequence. This filter employs a linear-phase finite-impulse response structure, with a cutoff frequency set at [value missing]. The system filters out high-frequency quantization noise and potential aliasing components introduced by the discretization process, while retaining all effective frequency components of the preset sinusoidal components. The filtered sequence is the final output dynamic time delay perturbation sequence, which is then sent to the adjustable time delay injection module as a time delay setting command.

[0051] Adaptive adjustment of the multi-frequency sinusoidal superposition components based on the actual operating environment parameters of the laser communication link is a crucial mechanism for ensuring test results closely approximate real-world conditions. Actual operating environment parameters include, but are not limited to, orbital altitude, link elevation angle, atmospheric turbulence intensity level, and vibration platform characteristics. These parameters collectively determine the statistical distribution characteristics of the time delay in the target space link environment, i.e., the distribution weights of the time delay power spectral density across different frequency bands. A parameter lookup table, indexed by the combinations of these environmental parameters, pre-stores the corresponding frequency sets and amplitude weighting coefficient sets for each sinusoidal component. Before test initiation, the lookup table is retrieved based on the actual environmental parameters of the current link to obtain the corresponding frequency sets. With amplitude weight set It is then written into the parameter register of the composite disturbance waveform generator in real time to complete the dynamic configuration of frequency composition and amplitude weight.

[0052] The construction of the parameter lookup table relies on prior statistical modeling of the target space link environment. For different orbit types and atmospheric conditions, time delay power spectral density curves are extracted under typical operating conditions through ground-based equivalent experiments or historical on-orbit data fitting. These curves are then decomposed into power weights at several discrete frequency points. Representative frequency points are selected based on frequency coprime constraints, and the corresponding power weights are converted into amplitude weight coefficients and stored in the lookup table. When actual operating environment parameters change, the lookup table mapping mechanism can update the parameters within one control cycle, ensuring that the spectral characteristics of the dynamic time delay disturbance sequence follow the changes in the link environment in real time. This guarantees that the time delay excitation applied in each test accurately reflects the time delay statistical characteristics under the target operating conditions, thus giving the final output targeting time delay tolerance upper limit clear operating condition specificity and engineering usability.

[0053] In practical implementation, the composite disturbance waveform generator can be implemented using a field-programmable gate array (FPGA). The linear ramp function is generated through an accumulator structure. The multi-frequency sine components are generated in parallel by a direct digital frequency synthesizer and then summed. The square root amplitude modulation is implemented by cascading a lookup table and a multiplier. The anti-aliasing filter is implemented using a pipeline structure. The overall delay does not exceed a few sampling periods, which meets the timing requirements for real-time online disturbance injection.

[0054] Construct a divergence trajectory curve of the aiming angle deviation residual over time. When the slope of the divergence trajectory curve exceeds a preset divergence threshold and the residuals at subsequent consecutive sampling times maintain a monotonically increasing trend, record the critical divergence delay, including: The divergence trajectory curve is constructed by plotting the sampling time on the horizontal axis and the difference between the aiming angle response of the reference circuit and the aiming angle response of the test circuit at each sampling time on the vertical axis. The divergence trajectory curve is subjected to piecewise sliding linear fitting. A sliding window with a preset window width moves along the divergence trajectory curve point by point. At each window position, the slope of the fitted straight line of all sampling points in the window is calculated as the local divergence slope, forming a local divergence slope sequence distributed along the time axis. An adaptive threshold decision is made on the local divergence slope sequence. The preset divergence threshold is dynamically adjusted according to the standard deviation of the residual of the stationary segment. When the local divergence slope exceeds the adaptive threshold for the first time and the local divergence slope of the subsequent preset number of sampling points continues to be above the adaptive threshold and the aiming angle deviation residual maintains a monotonically increasing trend, the sampling point that first exceeds the threshold is identified as the divergence start point, and the injection delay value at the time corresponding to the divergence start point is used as the critical divergence delay. The critical divergence delay is compared with the preset design delay tolerance of the spaceborne laser communication terminal, and the upper limit of the aiming delay tolerance and the judgment result of whether the design tolerance is met are output.

[0055] A divergence trajectory curve is plotted point-by-point, with sampling time on the horizontal axis and the difference in aiming angle response between the reference loop and the test loop at each sampling time on the vertical axis. Specifically, at each discrete sampling time, the current angle response value is read from the aiming actuator of the reference loop, and the corresponding angle response value is read from the aiming actuator of the test loop. The difference between the two values ​​yields the aiming angle deviation residual at that moment. The residuals from all sampling times are connected sequentially to form a curve that unfolds over time. This curve typically exhibits a stable segment with low-amplitude fluctuations when the injection delay is small. As the injection delay gradually increases, the curve begins to show a sustained upward divergence trend when the injection delay approaches the tolerance limit of the aiming control loop. The vertical axis of the divergence trajectory curve has the same dimension as the aiming angle response, usually in microradians or arcseconds. The horizontal axis is labeled with the sampling point number or actual timestamp for easy time alignment with the injection delay sequence.

[0056] When performing piecewise sliding linear fitting on the divergence trajectory curve, a preset window width is used. A sliding window (in units of sampling points) moves along the curve point by point. At the... The window position, window coverage starts from the first From the sampling point to the... A total of 1 sampling point There are 10 data points. A least-squares linear fit is performed on the residual data within the window and the corresponding time coordinates. The slope of the fitted line is the local divergence slope at that window location. Let the window contain the first... The time coordinates of each sampling point are The residual value is The local divergence slope is then obtained using the least squares method: As the sliding window moves point by point along the time axis, a set of locally divergent slope sequences distributed along the time axis is obtained. Window width The selection of the window width needs to strike a balance between time resolution and noise suppression: if the window width is too small, residual noise will significantly interfere with slope estimation, easily leading to misjudgments; if the window width is too large, the positioning accuracy of the divergence initiation point will decrease. In practical applications, the window width is usually selected based on the response time constant corresponding to the aiming control bandwidth, generally taking 2 to 5 times the number of sampling points corresponding to the control loop response time constant, to ensure that the window contains enough data points to support linear fitting, while not obscuring the positional information at the divergence initiation moment.

[0057] When performing adaptive thresholding on a locally divergent slope sequence, it is necessary to first determine the range of the stationary phase. The stationary phase corresponds to a period with relatively small injection delays and before significant divergence occurs in the test loop. Within this phase, residual fluctuations mainly originate from measurement noise and inherent jitter in the control loop. All residual values ​​within the stationary phase are then taken, and their standard deviations are calculated. Adaptive threshold The standard deviation of the residuals in the stationary phase and the preset multiplier. The product is determined as follows: in This is the preset sensitivity adjustment coefficient, which is usually set to a value between 3 and 6. When the value is small, the decision sensitivity is high and the divergence trend can be detected earlier, but the probability of false triggering also increases accordingly. When the value is larger, the decision is more conservative, and the probability of false triggering is reduced, but there is a certain lag in the detection of the divergence initiation point. The boundary of the stationary segment can be determined by pre-setting the time interval corresponding to the initial stage in the process of increasing injection delay, or it can be automatically identified by statistical analysis of the first part of the local divergence slope sequence.

[0058] After completing the adaptive threshold calculation, a decision is made point-by-point on the local divergence slope sequence. When a certain sampling point... Local divergence slope at the location First time exceeding the adaptive threshold At that time, the continuous confirmation mechanism is activated: check from The following series Does the local divergence slope of each sampling point remain consistently within a certain range? Above this, simultaneously check whether the aiming angle deviation residual within the corresponding time period maintains a monotonically increasing trend. The method for determining a monotonically increasing trend is: in subsequent... In each sampling point, the residual difference between adjacent sampling points is positive, or local regression points not exceeding a preset proportion are allowed, but the overall trend is upward. If both of the above conditions are met, then... Confirmed as the starting point of divergence; if in the subsequent A local divergence slope occurred at one of the sampling points, decreasing to [a certain value]. If the residual shows a continuous decrease, the current threshold exceedance is determined to be a false trigger caused by noise interference. The current candidate divergence starting point is then cleared, and the search continues for the next threshold exceedance point. Number of consecutive confirmed points. The value needs to be set according to the specific engineering scenario, and usually 5 to 15 sampling points are selected to achieve a reasonable compromise between noise immunity and detection delay.

[0059] Confirm the divergence starting point After that, extract The injection delay value corresponding to the sampling time is the critical divergence delay. Since the injected delay sequence and the residual sequence are strictly aligned on the time axis, they can be directly indexed by sampling points. The corresponding value is found in the injected delay sequence without additional time interpolation. If the injected delay has increased to the preset upper limit during the entire test but the divergence judgment condition is never triggered, it is recorded as "No critical divergence delay detected", indicating that the delay tolerance of this link is better than the current test range upper limit, and it is recommended to expand the preset upper limit and retest.

[0060] Critical divergence delay With respect to the preset time delay tolerance of the spaceborne laser communication terminal Perform a comparison. If... If the aiming delay tolerance of the laser communication link meets the design tolerance requirements, a "qualified" result is output; otherwise... If the design tolerance is not met, the result "Unqualified" will be output, along with the output... and The difference serves as a margin deficiency, providing a quantitative reference for subsequent control loop parameter optimization or hardware delay compression. The final output tolerance test results include the critical divergence delay value and the adaptive threshold. The divergence starting point location, the judgment conclusion on whether the design tolerance is met, and the margin information are fully recorded in the test report for use in the design verification and on-orbit performance evaluation of the spaceborne laser communication terminal.

[0061] Constructing the divergence trajectory curve and determining the critical divergence delay includes: The curve of the time delay value of the dynamic time delay disturbance sequence changing with time is superimposed on the divergence trajectory curve as an auxiliary reference axis. The corresponding relationship between the divergence process of the aiming angle deviation residual and the growth process of the injected time delay value is presented in the same time coordinate system in the form of dual vertical axes. The divergence trajectory curve is segmented into three feature segments: a stable segment, a transition segment, and a divergence segment. The stable segment corresponds to the gradually changing region of the aiming angle deviation residual, the transition segment corresponds to the acceleration region where the aiming angle deviation residual begins to deviate from the random fluctuation range, and the divergence segment corresponds to the out-of-control region where the aiming angle deviation residual shows a continuous exponential growth trend. The injection delay value corresponding to the boundary point between the transition segment and the divergence segment is used as a candidate value for the critical divergence delay. The candidate value is cross-validated with the critical divergence delay corresponding to the actual divergence start point, and the smaller of the two is taken as the final critical divergence delay. When the difference between the candidate value and the critical divergence delay corresponding to the actual divergence start point exceeds the preset divergence threshold, the injection delay value of the preset protection margin before the divergence segment start time is taken as the final critical divergence delay.

[0062] In constructing the divergence trajectory curve, a dual-vertical-axis time coordinate system is used to jointly visualize the aiming angle deviation residual and the injection delay value. The left vertical axis corresponds to the magnitude of the aiming angle deviation residual, in microradians or arcseconds, while the right vertical axis corresponds to the injection delay value of the dynamic delay perturbation sequence, in nanoseconds. Both axes share the same horizontal-axis time coordinate. The curve showing the injection delay value changing over time is superimposed on the coordinate system of the divergence trajectory curve as an auxiliary reference axis, allowing for a direct reading of the correlation between the current injection delay and the corresponding residual amplitude at any given time. The significance of this dual-vertical-axis joint presentation method is that testers do not need to perform time alignment comparisons between two independent charts. Instead, they can directly observe the causal correspondence between the residual divergence behavior and the injection delay growth process in the same time coordinate system, thus providing an intuitive basis for subsequent critical divergence delay determination.

[0063] After the dual-axis graph is plotted, segment-by-segment feature extraction is performed on the divergence trajectory curve, dividing the entire curve into three characteristic segments: a stable segment, a transition segment, and a divergence segment. The stable segment corresponds to the initial stage of the test. At this time, the injection delay value is small, the aiming control loop has sufficient stability margin, and the aiming angle response of the test loop can effectively follow the reference loop. The residual value changes slowly within a small random fluctuation range, without showing a systematic growth trend, exhibiting an overall low-amplitude random walk characteristic. The transition segment corresponds to the stage where the injection delay gradually approaches the stability boundary of the control loop. At this time, the residual value begins to deviate from the random fluctuation range defined by the stable segment, showing an accelerating growth trend, but has not yet entered a completely out-of-control state. The growth rate of the residual is between the slow change level of the stable segment and the exponential growth level of the divergence segment, exhibiting obvious accelerated transition characteristics. The divergence segment corresponds to the stage after the injection delay exceeds the stability limit of the control loop. At this time, the aiming control loop loses its effective ability to suppress delay disturbances, the residual value shows a continuous exponential growth trend, the control error rapidly amplifies to an unacceptable level within a finite time, and the system enters an out-of-control state.

[0064] The automatic division of the three feature segments is achieved using a segmentation algorithm based on local statistical features. In the stationary segment identification stage, several sampling points at the beginning of the divergent trajectory curve's time axis are taken as a baseline statistical window. The mean and standard deviation of the residuals within this window are calculated, and the mean plus a certain multiple of the standard deviation is used as the upper bound of the stationary segment's fluctuation. When the residual value continuously exceeds this upper bound and consistently exceeds a preset number of confirmation points, the stationary segment is determined to have ended, and the transition segment to have begun. The identification of the boundary between the transition segment and the divergent segment relies on the abrupt detection of the residual growth rate. By performing second-order difference analysis on the local slope sequence of the divergent trajectory curve, when the local slope growth rate shows a significant jump and the slope value exceeds a preset divergence threshold, the transition segment is determined to have ended, and the divergent segment to have begun. The corresponding injection delay values ​​are read on the auxiliary reference axis at the times corresponding to the boundary points of the transition segment and the divergent segment, serving as candidate values ​​for the critical divergence delay. .

[0065] Candidate values Once determined, it is compared with the critical divergence delay corresponding to the actual divergence start point determined by the sliding window slope detection method. Cross-validation is performed. The purpose of cross-validation is to reduce the risk of misjudgment caused by noise interference or local anomalies in a single decision method by mutually verifying two independent methods. and The two are compared, and the smaller one is taken as the critical divergence delay of the final output. ,Right now Taking the smaller value reflects a conservative design philosophy, ensuring that the determined upper limit of delay tolerance does not overestimate the actual stability margin of the system, thus providing a safer design margin reference for the engineering application of spaceborne laser communication terminals.

[0066] when and The difference between Exceeding the preset divergence threshold If the two determination methods yield significantly different conclusions, the smaller value will not be directly adopted; instead, a protection margin mechanism will be activated. Specifically, based on the start time of the divergence segment, a preset protection margin will be moved backward along the time axis. The corresponding time length is used to read the injection delay value on the auxiliary reference axis at that moment, and this value is taken as the final critical divergence delay. Protection margin The setting takes into account both the response speed of the control system and the safety requirements of the test scenario. It is usually taken as a multiple of the inverse of the control loop closed-loop bandwidth to ensure that the final determined upper limit of time delay tolerance has a sufficient safety margin relative to the actual divergence starting point, so as to prevent the communication link from becoming unstable due to an overly optimistic estimation of time delay margin in engineering applications.

[0067] Significant discrepancies between the two determination methods typically occur when the transition segment of the divergence trajectory curve is relatively gentle and the divergence process is not steep enough. In this case, candidate values ​​extracted based on segmented features... Judgment shifts may occur due to the blurred boundaries between the transition and divergence segments, while those based on sliding window slope detection... Different conclusions may be drawn due to differences in the slope threshold setting. The protection margin mechanism intervenes when there is a significant divergence between the two methods. By introducing a fixed safety backoff amount, it avoids the need for subjective judgment on which method is superior and provides the final critical divergence delay with a deterministic conservative strategy, thereby ensuring that the test results have reliable engineering applicability in different test scenarios.

[0068] During the test result output phase, the final determined critical divergence delay will be... With respect to the preset time delay tolerance of the spaceborne laser communication terminal To make a comparison, if If the aiming delay tolerance capability of the laser communication link meets the design requirements, then it is determined that the laser communication link meets the design specifications. If the requirement is not met, then it will be determined that the requirement is not met, and will be... and The difference is output as a margin of insufficient time delay tolerance, providing a quantitative reference for subsequent control loop parameter optimization or hardware time delay compression. The dual-vertical-axis divergence trajectory curve and the division results of the three characteristic segments are output as core graphic attachments to the test report for manual review and technical examination by engineers.

[0069] A second aspect of the present invention provides a satellite laser communication aiming delay tolerance testing system, comprising: A dual-loop construction unit is used to construct a first aiming control loop and a second aiming control loop for a satellite laser communication link. The first aiming control loop and the second aiming control loop share the same error signal source output by the aiming error detection unit. The first aiming control loop transmits the error signal directly to its aiming actuator as a reference loop, and the second aiming control loop connects an adjustable time delay injection module in series in the error signal transmission path as a test loop. The time delay perturbation injection unit is used to apply a dynamic time delay perturbation sequence that changes continuously with time to the adjustable time delay injection module during the testing phase. The time delay value of the dynamic time delay perturbation sequence gradually increases from zero to a preset upper limit value according to a preset perturbation mode, and the aiming angle response of the reference circuit and the aiming angle response of the test circuit are collected simultaneously. The critical delay determination unit is used to calculate the aiming angle deviation residual between the reference loop and the test loop at each sampling time, construct a divergence trajectory curve of the aiming angle deviation residual changing with time, and when the slope of the divergence trajectory curve exceeds a preset divergence threshold and the residual of multiple subsequent consecutive sampling times maintains a monotonically increasing trend, the injection delay value at the corresponding time is recorded as the critical divergence delay, and the critical divergence delay is used as the upper limit of the aiming delay tolerance of the laser communication link, and the tolerance test result is output.

[0070] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0071] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0072] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for testing the time delay tolerance of satellite laser communication aiming, characterized in that, include: A first aiming control loop and a second aiming control loop are constructed to form a satellite laser communication link. The first aiming control loop and the second aiming control loop share the same error signal source output by the aiming error detection unit. The first aiming control loop transmits the error signal directly to its aiming actuator as a reference loop. The second aiming control loop connects an adjustable time delay injection module in series in the error signal transmission path as a test loop. During the testing phase, a dynamic delay perturbation sequence that changes continuously with time is applied to the adjustable delay injection module. The delay value of the dynamic delay perturbation sequence gradually increases from zero to a preset upper limit value according to a preset perturbation mode. Simultaneously, the aiming angle response of the reference circuit and the aiming angle response of the test circuit are collected. Calculate the aiming angle deviation residual between the reference loop and the test loop at each sampling time, construct the divergence trajectory curve of the aiming angle deviation residual as a function of time, and when the slope of the divergence trajectory curve exceeds the preset divergence threshold and the residual of subsequent consecutive sampling times maintains a monotonically increasing trend, record the injection delay value at the corresponding time as the critical divergence delay, use the critical divergence delay as the upper limit of the aiming delay tolerance of the laser communication link, and output the tolerance test result.

2. The method according to claim 1, characterized in that, Constructing a first aiming control loop and a second aiming control loop includes: Obtain the output port parameters and signal type parameters of the aiming error detection unit of the laser communication link; A signal splitter is provided at the output port. The signal splitter replicates the error signal source output by the aiming error detection unit into a first error signal and a second error signal. The amplitude and phase characteristics of the two error signals are completely consistent. The first error signal is connected to the aiming actuator of the first aiming control loop to form a reference loop, and the second error signal is connected to the input of the adjustable delay injection module. The output of the adjustable delay injection module is connected to the aiming actuator of the second aiming control loop to form a test loop. The reference circuit and the test circuit are initially zero-point calibrated. The total delay difference between the two circuits from the error signal input to the aiming angle response is measured without injection delay. The total delay difference is used as the reference zero-point offset. When calculating the aiming angle deviation residual at each subsequent sampling time, the angle compensation amount corresponding to the reference zero-point offset is subtracted from the aiming angle response of the test circuit, and then compared with the aiming angle response of the reference circuit to calculate the difference.

3. The method according to claim 2, characterized in that, Initial zero-point calibration of the signal transmission delay between the reference circuit and the test circuit includes: When the injection delay value of the adjustable delay injection module is set to zero, a preset calibration excitation signal is input to the aiming error detection unit. The preset calibration excitation signal includes a step component and a sinusoidal sweep frequency component to cover the full-band response characteristics of the aiming control loop. The time sequence of the first angle response output by the aiming actuator of the reference loop and the time sequence of the second angle response output by the aiming actuator of the test loop are recorded synchronously. Calculate the cross-correlation function between the first angle response time series and the second angle response time series, and use the peak offset of the cross-correlation function as the total delay difference; When the delay difference corresponding to the peak offset of the cross-correlation function is greater than the preset delay tolerance, the hardware transmission path of the reference circuit and the test circuit is adjusted for physical delay matching until the total delay difference falls within the preset delay tolerance range, and the final adjusted total delay difference is used as the reference zero point offset. Before the start of each test phase, the initial zero-point calibration is re-performed to obtain the reference zero-point offset for that test, ensuring that the aiming angle deviation residuals of different test batches are repeatable.

4. The method according to claim 1, characterized in that, A dynamic delay perturbation sequence that varies continuously with time is applied to the adjustable delay injection module. The delay value of the dynamic delay perturbation sequence gradually increases from zero to a preset upper limit value according to a preset perturbation mode, including: The spectral coverage of the dynamic delay perturbation sequence is determined based on the aiming control bandwidth of the laser communication link, so that the main lobe width of the dynamic delay perturbation sequence covers a preset multiple of the aiming control bandwidth, thereby fully stimulating the dynamic response characteristics of the aiming control loop under different frequency delay perturbations. A composite perturbation waveform is generated, which is composed of a linearly increasing fundamental component and a multi-frequency sinusoidal superposition component. The linearly increasing fundamental component provides a global time delay increasing trend, and the multi-frequency sinusoidal superposition component superimposes time-varying perturbations of different frequencies on the linearly increasing fundamental component to simulate the multi-source time delay coupling effect caused by atmospheric turbulence in the space laser link. The discrete sampling sequence of the composite disturbance waveform is used as the dynamic time delay disturbance sequence. At each sampling time, the corresponding sampling value is written into the injection time delay register of the adjustable time delay injection module, and the injection time delay value of the adjustable time delay injection module is controlled to change continuously according to the composite disturbance waveform. During the increasing process of the dynamic time delay disturbance sequence, the tracking deviation between the actual output time delay of the adjustable time delay injection module and the set injection time delay is monitored synchronously. When the tracking deviation exceeds the preset tracking tolerance, the dynamic time delay disturbance sequence is adaptively corrected in amplitude.

5. The method according to claim 4, characterized in that, Generate composite disturbance waveforms, including: The linearly increasing fundamental component is set as a linear ramp function with a time delay value from zero to the preset upper limit value. The slope of the linear ramp function is determined by the ratio of the preset upper limit value to the preset test duration. A multi-frequency sinusoidal superposition component is constructed. The multi-frequency sinusoidal superposition component is composed of a preset number of sinusoidal components with coprime frequencies and different amplitudes, and the frequency of each sinusoidal component is selected within a specified octave range of the aiming control bandwidth. The amplitude of each sinusoidal component is proportional to the square root of the current time delay value of the linear ramp function. The linear ramp function and the multi-frequency sinusoidal superposition component are added point by point to obtain the discrete sampling sequence of the composite disturbance waveform. The discrete sampling sequence is then subjected to anti-aliasing smoothing filtering and output as the dynamic time delay disturbance sequence. The frequency composition and amplitude weight of each sinusoidal component in the multi-frequency sinusoidal superposition component are adjusted according to the actual operating environment parameters of the laser communication link, so that the spectral characteristics of the dynamic time delay disturbance sequence match the time delay statistical distribution characteristics of the target space link environment. The frequency composition and amplitude weight are mapped to the generation parameters of each sinusoidal component through a parameter lookup table and adjusted in real time.

6. The method according to claim 1, characterized in that, Construct a divergence trajectory curve of the aiming angle deviation residual over time. When the slope of the divergence trajectory curve exceeds a preset divergence threshold and the residuals at subsequent consecutive sampling times maintain a monotonically increasing trend, record the critical divergence delay, including: The divergence trajectory curve is constructed by plotting the sampling time on the horizontal axis and the difference between the aiming angle response of the reference circuit and the aiming angle response of the test circuit at each sampling time on the vertical axis. The divergence trajectory curve is subjected to piecewise sliding linear fitting. A sliding window with a preset window width moves along the divergence trajectory curve point by point. At each window position, the slope of the fitted straight line of all sampling points in the window is calculated as the local divergence slope, forming a local divergence slope sequence distributed along the time axis. An adaptive threshold decision is made on the local divergence slope sequence. The preset divergence threshold is dynamically adjusted according to the standard deviation of the residual of the stationary segment. When the local divergence slope exceeds the adaptive threshold for the first time and the local divergence slope of the subsequent preset number of sampling points continues to be above the adaptive threshold and the aiming angle deviation residual maintains a monotonically increasing trend, the sampling point that first exceeds the threshold is identified as the divergence start point, and the injection delay value at the time corresponding to the divergence start point is used as the critical divergence delay. The critical divergence delay is compared with the preset design delay tolerance of the spaceborne laser communication terminal, and the upper limit of the aiming delay tolerance and the judgment result of whether the design tolerance is met are output.

7. The method according to claim 6, characterized in that, Constructing the divergence trajectory curve and determining the critical divergence delay includes: The curve of the time delay value of the dynamic time delay disturbance sequence changing with time is superimposed on the divergence trajectory curve as an auxiliary reference axis. The corresponding relationship between the divergence process of the aiming angle deviation residual and the growth process of the injected time delay value is presented in the same time coordinate system in the form of dual vertical axes. The divergence trajectory curve is segmented into three feature segments: a stable segment, a transition segment, and a divergence segment. The stable segment corresponds to the gradually changing region of the aiming angle deviation residual, the transition segment corresponds to the acceleration region where the aiming angle deviation residual begins to deviate from the random fluctuation range, and the divergence segment corresponds to the out-of-control region where the aiming angle deviation residual shows a continuous exponential growth trend. The injection delay value corresponding to the boundary point between the transition segment and the divergence segment is used as a candidate value for the critical divergence delay. The candidate value is cross-validated with the critical divergence delay corresponding to the actual divergence start point, and the smaller of the two is taken as the final critical divergence delay. When the difference between the candidate value and the critical divergence delay corresponding to the actual divergence start point exceeds the preset divergence threshold, the injection delay value of the preset protection margin before the divergence segment start time is taken as the final critical divergence delay.

8. A satellite laser communication aiming delay tolerance test system, used to implement the method as described in any one of claims 1-7, characterized in that, include: A dual-loop construction unit is used to construct a first aiming control loop and a second aiming control loop for a satellite laser communication link. The first aiming control loop and the second aiming control loop share the same error signal source output by the aiming error detection unit. The first aiming control loop transmits the error signal directly to its aiming actuator as a reference loop, and the second aiming control loop connects an adjustable time delay injection module in series in the error signal transmission path as a test loop. The time delay perturbation injection unit is used to apply a dynamic time delay perturbation sequence that changes continuously with time to the adjustable time delay injection module during the testing phase. The time delay value of the dynamic time delay perturbation sequence gradually increases from zero to a preset upper limit value according to a preset perturbation mode, and the aiming angle response of the reference circuit and the aiming angle response of the test circuit are collected simultaneously. The critical delay determination unit is used to calculate the aiming angle deviation residual between the reference loop and the test loop at each sampling time, construct a divergence trajectory curve of the aiming angle deviation residual changing with time, and when the slope of the divergence trajectory curve exceeds a preset divergence threshold and the residual of multiple subsequent consecutive sampling times maintains a monotonically increasing trend, the injection delay value at the corresponding time is recorded as the critical divergence delay, and the critical divergence delay is used as the upper limit of the aiming delay tolerance of the laser communication link, and the tolerance test result is output.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images