Calibration efficiency evaluation system and calibration efficiency evaluation method suitable for single efficiency index

Abstract

The invention provides a calibration performance evaluation system and a calibration performance evaluation method suitable for a single performance index, and relates to the technical field of spaceflight. The method comprises the steps of receiving a remote control instruction, identifying a task intention and triggering an evaluation mode; acquiring a real-time physical state of a satellite, and calculating a dynamic calibration reference value of a single efficiency index based on a physical model; obtaining an actual measurement result of the index probe type measurement, and calculating the deviation between the actual measurement result and a reference value; and combining the task intention and the deviation adjustment evaluation parameter, and outputting a self-calibration calibration evaluation result. According to the method, the problems that dynamic response is lagged and errors are introduced by manual intervention in traditional multi-dimensional evaluation can be solved, rapid and reliable calibration of a single efficiency index in space-based micro-nano satellite networking is achieved, and safe switching of high-value services is guaranteed.

Description

TECHNICAL FIELD

[0001] The application provides a single performance index scaling performance evaluation system and a single performance index scaling performance evaluation method, and belongs to the technical field of aerospace. BACKGROUND

[0002] In the complex operating environment of space-based micro-nano satellite networking, accurate evaluation of the performance of each satellite is crucial to ensure the smooth progress of the task. However, existing evaluation methods often face challenges, especially when it comes to quickly and reliably scaling evaluation for a single performance indicator in multiple dimensions such as communication, computation, and energy. Due to the complex interaction between multi-dimensional indicators and the strong data correlation, traditional methods often struggle to quickly and accurately recalibrate specific indicators when faced with subtle changes in the operating environment, resulting in delayed or inaccurate evaluation results. In addition, when evaluation is difficult, artificially adjusting evaluation parameters may introduce new errors and uncertainties, making the evaluation results even less reliable.

[0003] For example, in the daily operation of space-based micro-nano satellite networking, in order to ensure the quality of service, the system operator will establish a set of fine evaluation system for different key performance indicators, such as communication throughput, computation resource utilization or energy efficiency. Taking communication throughput as an example, in order to accurately evaluate it, system maintenance personnel will regularly establish and update a set of "reference data set for performance scaling". This data set is not generated randomly, but is determined based on long-term observation and analysis of specific, stable satellite orbit configuration, inter-satellite link topology and pre-set communication traffic pattern. It represents the communication performance benchmark that the network can achieve under ideal or standard operating conditions, and is an important reference for measuring subsequent performance. For example, under a certain fixed satellite formation configuration, by transmitting data packets of known size within a certain time period and recording their transmission rate, packet loss rate and delay, a reference data set reflecting the upper limit of communication throughput and stability under this configuration can be established.

[0004] Suppose that in order to optimize and enhance the high-value communication link service between two specific ground stations on Earth, a sub-network in the satellite network performs an orbit adjustment task according to the predetermined plan. This task involves multiple micro-nano satellites performing precise orbital maneuvers, with the core purpose being to shorten the communication distance and optimize the link geometry configuration by changing the relative positions of the satellites, thereby improving the communication throughput capacity of the specific link. For example, by adjusting a group of relay satellites to a closer orbit plane of the ground station, or adjusting their phase on the orbit, to ensure more direct and stable line-of-sight communication within a longer time window. This orbit adjustment is carefully calculated and planned to achieve specific performance gains.

[0005] However, despite the successful completion of the above-mentioned orbit adjustment task, the subtle changes in the relative positions and attitudes of the satellites in actual operation cause deviations in the actual propagation path, signal strength, and time delay characteristics of the inter-satellite laser communication link from the environment reflected by the "reference data set for performance scaling" before adjustment. This deviation is not caused by equipment failure, but by the natural evolution of the physical operating environment. For example, even a micron-level attitude angle deviation can cause the light beam to deviate from the target receiver center by several meters or even tens of meters on a laser communication link of thousands of kilometers long, resulting in a significant decrease in received optical power. In addition, the unevenness of the Earth's gravitational field, the slight disturbance of the solar radiation pressure, and the residual jitter of the satellite's own attitude control system will accumulate over a long period of time, resulting in subtle differences between the actual orbit and attitude of the satellite and the theoretical values that are difficult to completely eliminate. These physical changes greatly reduce the representativeness of the original reference data set, as the "specific, stable satellite orbit configuration" premise on which it is based is no longer fully established. At this time, if the old reference data set is still used to evaluate the communication throughput, the results will not accurately reflect the true performance of the current network.

[0006] In the face of such changes in the physical characteristics of the communication link, traditional performance evaluation systems face significant challenges in re-scaling. Such systems typically use a multi-dimensional evaluation framework, with a "composite calculation formula for performance indicators" that takes into account a variety of heterogeneous indicators, such as signal-to-noise ratio, bit error rate, data packet delay, network congestion level, and even some energy consumption data. There are complex interactions and nonlinear relationships between these indicators. For example, a decrease in signal-to-noise ratio not only directly affects the bit error rate, but also can trigger a link adaptation mechanism, changing the modulation and coding scheme, and thus affecting the actual throughput. When the physical characteristics of the communication link change, this multi-dimensional, strongly correlated evaluation logic makes it difficult for the system to quickly and accurately re-scale for the single indicator of communication throughput. The system exhibits significant lag in identifying and quantifying such changes and updating the evaluation reference. This is because, to accurately update a composite formula involving multiple interrelated indicators, it is necessary to comprehensively perceive and accurately measure the current state of all related indicators and re-calculate their weights and interactions, which is itself a computationally intensive and time-consuming process. For example, if the signal strength of a laser communication link decreases due to attitude deviation, the traditional system may need to adjust multiple parameters related to signal strength, such as reception sensitivity threshold, forward error correction code strength, and even consider whether to switch to a backup frequency. The interactions of these adjustments make accurate evaluation of a single indicator complex and slow.

[0007] Following the orbital readjustment, the subnetwork was scheduled to take over the previously planned high-value communication services as quickly as possible, according to the operational plan. However, due to the slow response of the aforementioned assessment system to environmental changes, the ground control center was unable to obtain a reliable assessment of the current communication throughput. This meant that the operator could not confirm whether the orbital readjustment had truly achieved the expected communication enhancement effect, nor could it determine the actual carrying capacity of the current network. Without confirming the actual communication capacity, hastily switching high-value services (e.g., those involving financial transactions, emergency rescue, or sensitive scientific data transmission) to this unverified subnetwork posed a serious risk of communication interruption or data loss. This uncertainty prevented the timely realization of the value of the entire orbital readjustment operation, causing not only severe operational delays but also potentially significant economic losses and reputational risks. For example, a mission to transmit Earth observation data, originally scheduled to be completed within a specific time window, might fail to start on time due to a delayed assessment, potentially leading to expired observation data or inability to be processed synchronously with the ground, thus losing its scientific value.

[0008] To complete network verification before the service switchover deadline, operators, under immense pressure, attempted to force the system to output a seemingly reasonable throughput assessment by manually adjusting multiple settings in the "Quantitative Model of Communication Link Quality." For example, they might, based on experience, artificially increase the value of a "link gain" parameter in the model or lower the "noise tolerance" threshold, hoping the assessment result would appear more consistent with expectations. However, this experience-based intervention not only failed to solve the fundamental problem but also introduced new assessment errors and uncertainties due to a lack of understanding of the complex relationships between different settings. The Quantitative Model of Communication Link Quality is a highly complex mathematical model with intricate interdependencies among its internal parameters (such as modulation order, coding efficiency, antenna pointing accuracy, atmospheric attenuation coefficient, etc.). An empirical adjustment to one parameter can trigger a chain reaction within the model, causing implicit assumptions about other parameters to be broken, thus causing the entire model's output to deviate from reality. For example, if an operator blindly increases the model's "effective bandwidth" parameter without understanding the actual channel conditions, the system might calculate an artificially high throughput, when in reality, due to physical channel limitations, this throughput is impossible to achieve. This kind of artificial, unsystematic intervention makes the assessment results less reliable, and may even be more dangerous than no assessment at all, because it provides a false sense of "certainty".

[0009] Ultimately, after the planned orbital adjustments aimed at improving performance during the deployment of space-based micro-nano satellites, the existing evaluation system faced a serious dilemma when confronted with a ground mission demanding extremely high timeliness and accuracy. It was unable to provide a reliable, effectively calibrated evaluation result for a single performance metric—communication throughput—within a short timeframe. Traditional multi-dimensional, interdependent evaluation logic, due to its inherent complexity and sluggish response, proved inadequate in rapidly adapting to environmental changes and focusing on a single key indicator. Worse still, human intervention further exacerbated the uncertainty of the evaluation, creating severe technical difficulties for mission decision-making. Therefore, a method for rapidly and accurately calibrating and evaluating a single performance metric was urgently needed to cope with this complex and constantly changing operational environment. Summary of the Invention

[0010] Based on this, the present invention provides a calibration performance evaluation method applicable to a single performance indicator. The present invention transforms the "behavioral purpose" inherent in the mission instructions issued by the ground control center into the intrinsic driving force for the system to conduct performance evaluation and parameter adjustment. By intelligently parsing this high-level semantic information, the system can proactively predict the impact of environmental changes on specific performance indicators and immediately initiate a customized, physically-based reference value generation and rapid measurement mechanism. On this basis, the system can autonomously and accurately calibrate its internal evaluation parameters according to the deviation between the mission intent and measured data and theoretical references, thereby eliminating reliance on human experience and ensuring that the evaluation of a single key performance indicator is both rapid and reliable in the complex and ever-changing space-based environment.

[0011] This invention provides a calibration performance evaluation method applicable to a single performance index, aiming to achieve rapid and reliable evaluation of a single performance index in space-based micro-nano satellite networking through an intelligent approach. The method includes the following steps: (1) Receive remote control command, identify the task intent contained in the remote control command, and trigger evaluation mode according to the task intent; (2) In the evaluation mode, real-time physical state information of the satellite is obtained, and dynamic calibration reference values ​​of a single performance index under the current environment are calculated based on the physical model. (3) Perform probe-type measurement on the single performance index to obtain the measured results, and calculate the deviation between the measured results and the dynamic calibration reference value; (3) Based on the task intent and the deviation, adjust the evaluation parameters and output the calibration evaluation result of the single performance index after self-calibration.

[0012] Preferably, step (1) is specifically performed as follows: The instruction parsing unit receives remote control instruction data packets and performs an integrity check on the remote control instruction data packets; When the instruction parsing unit detects that the semantic matching degree of the remote control instruction data packet is lower than a preset threshold, the instruction parsing unit sends an auxiliary information request to the operation situation awareness module. The operational situation awareness module collects real-time satellite orbital position information, attitude information, onboard laser communication terminal working status information, and future mission plan information. The intent inference logic cross-compares the instruction semantic fragments of the remote control instruction data packet with the multi-source operational status information to determine the consistency of the context and infer the task intent. When the intent inference logic infers that the task intent is to improve the communication throughput of a specific link, the intent inference logic generates an intent confirmation signal. The main control logic unit receives the intent confirmation signal and triggers the evaluation system to enter the evaluation mode.

[0013] Preferably, step (2) has the following operations: In the evaluation mode, real-time orbital position and attitude information of the satellite are acquired; The receiving satellite monitors the instantaneous fluctuations in the power of the optical signal entering the laser communication receiver or its electrical conversion signal; Extract the instantaneous pointer jitter index; The instantaneous jitter index is transmitted to the physical link characteristic calculation unit; The physical link characteristic calculation unit uses the satellite's real-time orbital position information, the attitude information, and the preset inter-satellite laser communication link physical design parameters, combined with the instantaneous pointing jitter index, to correct the pointing loss part in the theoretical link loss calculation; A dynamic calibration reference value for communication throughput in the current environment is calculated based on the corrected link loss.

[0014] Preferably, the specific operation of step (3) is as follows: Monitor the real-time bit error rate, signal strength fluctuations, and instantaneous changes in voltage and current of the satellite's internal power bus of the inter-satellite link; Identify whether there is instantaneous spatial environmental interference or internal electromagnetic interference during the measurement; When a transient interference event is detected, pause the current probe measurement; A new round of probe measurements is restarted to perform probe-based measurements on the single performance index and obtain the measured results. Calculate the deviation between the measured result and the dynamic calibration reference value; Record the time, type, and duration of the interference; Interference information is used as an auxiliary basis for judgment during the evaluation parameter self-calibration process.

[0015] Preferably, the specific operation of step (4) is as follows: S1. Identify satellite subsystems or link performance indicators that have a nonlinear coupling relationship with the evaluation parameters to be adjusted, and obtain a list of potential impacts; S2. Obtain real-time operational constraint data from the satellite health and safety management unit to obtain multi-dimensional constraint conditions; S3. Combining the mission intent, the deviation, the potential impact list, and the multi-dimensional constraints, simulate the impact of different adjustment strategies on the entire satellite system and select the optimal adjustment strategy. S4. Adjust the evaluation parameters according to the optimal adjustment strategy, and output the calibration evaluation result of the single performance index after self-calibration, and output a comprehensive report on the system health status.

[0016] More preferably, the step of identifying satellite subsystems or link performance indicators that have a nonlinear coupling relationship with the evaluation parameters to be adjusted, and obtaining a list of potential impacts, includes: In the evaluation mode, the system actively injects a preset micro-amplitude perturbation signal into the evaluation parameter to be adjusted; Simultaneously, the system monitors in real time the response data of multiple satellite subsystems or link performance indicators related to the evaluation parameters to be adjusted; Based on the response data, the system identifies nonlinear correlation patterns between the evaluation parameters to be adjusted and the performance indicators of the multiple satellite subsystems or links; The system dynamically updates the coupling relationship between evaluation parameters and satellite subsystem or link performance indicators based on the identified nonlinear correlation patterns; Based on the updated coupling relationships, the system generates a list of potential impacts that reflects the current actual operating status.

[0017] More preferably, the step of identifying the nonlinear correlation pattern between the evaluation parameter to be adjusted and the performance indicators of the plurality of satellite subsystems or links includes: Multi-scale time-frequency analysis was performed on the response data to identify correlation patterns within different time scales and frequency ranges; Feature extraction is performed on the identified association patterns; Based on the differences in feature vectors, the association patterns are clustered to decouple complex patterns into independent association patterns. Based on the independent association pattern, the coupling relationship between the evaluation parameters and the performance indicators of the satellite subsystem or link is dynamically updated; Based on the updated coupling relationships, a list of potential impacts reflecting the current actual operating status is generated.

[0018] More preferably, the step of obtaining real-time operational constraint data from the satellite health and safety management unit to obtain multi-dimensional constraint conditions includes: Obtain real-time operational constraint data from the satellite health and safety management unit; Perform data integrity verification and data range validity checks on the real-time operational constraint data; When the data integrity verification or the data range validity check fails, a data retransmission request is sent to the satellite health and safety management unit. When the satellite health and safety management unit is unable to provide effective data retransmission, it retrieves backup constraint data that matches the current satellite operation mode from the preset backup constraint database. The backup constraint data is fused with the acquired valid real-time operational constraint data to obtain multi-dimensional constraint conditions.

[0019] More preferably, the step of obtaining real-time operational constraint data from the satellite health and safety management unit includes: In the evaluation mode, real-time operational constraint data is obtained from the satellite health and safety management unit; The real-time operational constraint data is sampled over multiple cycles to obtain a set of sampled data; Perform statistical analysis on the sampled data to determine whether there are any abnormal fluctuations in the sampled data that exceed a preset statistical threshold; When abnormal fluctuations are found in the sampled data, the constraint data item corresponding to the abnormal fluctuation is identified. Based on the constraint data items corresponding to the abnormal fluctuations, the abnormal data points are removed from the sampled data; The sampled data after removing the abnormal data points is smoothed to obtain corrected real-time running constraint data. The corrected real-time running constraint data is used as a multi-dimensional constraint condition.

[0020] This invention also provides a calibration performance evaluation system applicable to a single performance indicator, comprising: The instruction receiving module is used to receive remote control instructions; The intent recognition module is used to identify the task intent contained in the remote control command; The mode triggering module is used to trigger the evaluation mode according to the task intent; The status acquisition module is used to acquire real-time physical status information of the satellite in the evaluation mode. The reference value calculation module is used to calculate the dynamic calibration reference value of a single performance index under the current environment based on the physical model. The measurement module is used to perform probe-type measurements on the single performance index to obtain the actual measurement results; The deviation calculation module is used to calculate the deviation between the measured result and the dynamic calibration reference value; The parameter adjustment module is used to adjust the evaluation parameters based on the task intent and the deviation. The results output module is used to output the calibration evaluation results of the single performance index after self-calibration.

[0021] Compared with the prior art, the present invention has the following beneficial effects: This invention departs from the traditional assessment system's passive response to environmental changes, instead adopting a proactive strategy of understanding mission intent and performing intelligent self-calibration. In space-based micro-nano satellite constellations, predetermined orbit adjustments are frequent and critical, directly impacting the physical operating environment. Furthermore, the physical characteristics of inter-satellite laser communication links are subject to imperceptible deviations due to subtle changes in satellite relative positions and attitudes. Traditional assessment systems, due to their multi-dimensional framework and complex interactions between indicators, exhibit inherent complexity and sluggish response when rapidly adapting to these changes and focusing on the single indicator of communication throughput. Worse still, when assessments face difficulties, operator intervention based on experience introduces additional assessment errors and uncertainties. The representativeness of the original performance calibration reference dataset decreases significantly after changes in the physical environment, leading to inaccurate assessment data.

[0022] This invention elevates the "mission intent" inherent in commands issued by the ground control center from background information to the core driving force for system evaluation and calibration. It doesn't simply automate manual operations, but rather empowers the system to "understand" the purpose of operations, proactively and intelligently adjusting its evaluation logic accordingly. This shift from "command execution" to "intent understanding" allows the system to autonomously adjust and optimize evaluation parameters based on clear business objectives and real-time physical feedback. It abandons the traditional mindset that "a comprehensive evaluation is necessary to draw conclusions," instead focusing on the most critical single indicator and using a "surgical" precision detection to quickly pinpoint the core of the problem. This fundamental shift in reference sources ensures that evaluation is no longer constrained by the lag of environmental changes, but rather synchronized with physical changes. In this way, this solution solves the problems of dimensional redundancy, dynamic response lag, and errors introduced by human intervention caused by the high heterogeneity of indicators and strong data correlation in traditional methods within a space-based micro-nano satellite network environment, achieving rapid, reliable, and effective calibration of single performance indicators.

[0023] This invention significantly improves the speed and reliability of calibrating and evaluating single communication throughput metrics after orbital adjustments in space-based micro-nano satellite networks. It intelligently identifies mission intent, generates physical-driven evaluation references in real time, and combines this with rapid probe measurements for parameter self-calibration. This completely eliminates errors and uncertainties introduced by manual intervention, ensuring that the evaluation results accurately reflect the independent impact of subtle changes in the physical environment on communication throughput, thereby supporting rapid and secure switching of high-value services. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the preferred embodiments of this invention will be described in further detail below with reference to the examples. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without inventive effort are within the scope of protection of this invention.

[0025] This calibration performance evaluation method, applicable to a single performance indicator, mainly includes the following core logic: 1. Task Intent Perception and Evaluation Mode Trigger Logic: The system's internal command parsing unit, such as a software module based on keyword matching and syntax analysis, continuously scans remote control command data packets received from the ground control center. These data packets are typically transmitted in specific protocol formats, such as the CCSDS (Advisory Committee on Space Data Systems) protocol. When the parsing unit detects specific semantic patterns in the command data packets, such as "orbital maneuver command: target link throughput optimization" or "attitude adjustment: communication performance enhancement," it determines that the current operation's goal is to optimize communication throughput. Once a clear mission intent to "improve specific link communication throughput" is identified, the parsing unit generates an internal event signal. This event signal is then transmitted to the master control unit, triggering it to enter "communication performance self-calibration mode." Activation of this mode is manifested by setting a specific position in the system's internal status register to a high level, suspending the reference to old evaluation parameters, and preparing to initiate a new parameter calculation process. The rationale for this is that during satellite operation, the semantics of commands are explicit. By parsing these semantics, the system can proactively anticipate upcoming environmental changes and their impact on performance, rather than passively waiting for data anomalies to respond. This is crucial for the frequent orbit adjustment tasks in space-based micro-nano satellite networks.

[0026] 2. Dynamic reference throughput generation logic: In the "communication performance self-calibration mode," the system immediately instructs all microsatellites and nanosatellites participating in orbit adjustments to acquire and broadcast their latest, high-precision orbital position information (e.g., position vectors [X, Y, Z] in the J2000 coordinate system) and attitude information (e.g., quaternions [q0, q1, q2, q3] or Euler angles) through their onboard navigation and positioning systems (e.g., high-precision GPS receivers or inter-satellite ranging systems). This information is transmitted in real time via inter-satellite or satellite-to-ground telemetry links. A "physical link characteristic calculation unit" located at the ground control center or a designated computing node within the network receives and processes this information in real time. This unit utilizes pre-stored physical design parameters of the inter-satellite laser communication link, such as the laser's transmit power (P_tx, in watts), the receiver's sensitivity (S_rx, in watts), the beam divergence angle (θ_div, in radians), and the modulation and coding scheme used (e.g., QPSK, BPSK, and their corresponding bit rates). Based on the current satellite's actual geometry, this unit calculates the theoretical maximum communication throughput (R_max) that this inter-satellite link can achieve under current conditions using the following physical formula: First, calculate the inter-satellite distance (d): d = sqrt((X1-X2)^2 + (Y1-Y2)^2 + (Z1-Z2)^2) Where (X1, Y1, Z1) and (X2, Y2, Z2) are the real-time position coordinates of the two communication satellites, respectively.

[0027] Next, calculate the link loss (L_path): L_path = (λ / (4πd))^2 G_tx G_rx L_pointing Where λ is the laser wavelength, G_tx and G_rx are the gains of the transmitting and receiving antennas, respectively, and L_pointing is the pointing loss, which is calculated based on attitude information and beam divergence angle. For example, L_pointing = exp(-2 (θ_offset / θ_div)^2), where θ_offset is the actual pointing deviation angle.

[0028] Then, calculate the received optical power (P_rx): P_rx = P_tx L_path Next, based on the received optical power and receiver sensitivity, and in conjunction with the modulation and coding scheme, the theoretical maximum signal-to-noise ratio (SNR_max) and the corresponding theoretical maximum bit rate (R_bit_max) are calculated. For example, for a specific modulation scheme, R_bit_max = B log2(1 + SNR_max), where B is the bandwidth.

[0029] Ultimately, the theoretical maximum communication throughput is R_max = R_bit_max.

[0030] This real-time calculated theoretical maximum throughput will serve as a "dynamic calibration reference value" under the current environment, used for subsequent performance evaluation and comparison. This method eliminates the reliance on static historical data, ensuring that the evaluation reference value accurately reflects the latest physical reality after frequent orbit adjustments in space-based micro-nano satellite networking.

[0031] 3. Single-index rapid probe measurement logic: After the dynamic reference values ​​are generated, the system immediately instructs the relevant satellites to perform a series of "probe-based" data transmission tests on the target inter-satellite laser communication link. These tests are not comprehensive operational data transmissions, but rather the transmission of small-scale test data packets in a preset pattern. For example, fixed-size (e.g., 1KB) UDP or TCP test data packets can be sent, and their transmission timestamps are recorded. Upon receiving the data packets, the receiving satellite immediately sends back an acknowledgment and records the reception timestamp. The system directly measures the transmission rate of these test data packets (e.g., by calculating the amount of data successfully transmitted per unit time), the packet loss rate (by comparing the number of data packets sent and received), and the end-to-end transmission delay (by calculating the difference between the transmission timestamp and the reception timestamp). These measurement data are transmitted back to the ground control center or a designated computing node via a telemetry link. A "measured data comparison unit," such as a software module running on a general-purpose processor, quickly compares these measured results with the "dynamic calibration reference values" generated in step 2, calculating the deviation between the actual performance of the current communication throughput and the theoretical optimal value. For example, the deviation = (R_max - R_actual) / R_max, where R_actual is the measured throughput. This fast, focused probe measurement avoids the response lag caused by comprehensive data collection and complex calculations in traditional multi-dimensional evaluations, and is particularly suitable for scenarios that require rapid verification of a single communication throughput metric.

[0032] 4. Evaluation parameter self-calibration and output logic: The "behavioral intent recognition module" combines the identified task intent (e.g., increasing communication throughput) with the deviation between the measured data and the dynamic reference value in step 3 to automatically adjust the evaluation parameters related to communication throughput within the evaluation system. Specifically, if the measured throughput R_actual is lower than the dynamic reference value R_max, the system will automatically correct the weights or benchmark values ​​of parameters such as link loss and signal-to-noise ratio in the current environment according to the magnitude of the deviation, making them closer to the actual physical conditions. This adjustment is based on preset calibration rules and physical constraints, rather than empirical guesswork. For example, if the measured throughput is significantly lower than the theoretical value, the system will determine that there may be link attenuation that has not been fully perceived, and adjust the parameter of "effective link gain" in the evaluation model accordingly to reflect the current actual transmission capacity. The calibration rule can be a lookup table or an iterative optimization process based on a physical model. For example, if the deviation exceeds a certain threshold (e.g., 5%), the system will attempt to adjust the link loss parameter L_path_effective = L_path_calculated + ΔL, where ΔL is the correction amount calculated based on the deviation and preset calibration coefficients. Simultaneously, the system updates its internal "current link state parameter set" to ensure that subsequent evaluations are based on these calibrated parameters. Ultimately, the system outputs a self-calibrated, calibrated evaluation result for the current communication throughput. This result accurately reflects the actual communication capability after orbital adjustments, along with the confidence level of the evaluation (e.g., by calculating the goodness of fit of measured data to the calibrated model). This self-calibration mechanism completely eliminates manual intervention based on operator experience, ensuring the quality and consistency of evaluation results, especially providing reliable assessments when subtle deviations in the physical characteristics of inter-satellite links occur due to minute changes in satellite relative position and attitude.

[0033] The working principle of this invention can be figuratively understood as the system possessing the ability to "understand intent" and "self-adjust." In space-based micro-nano satellite networking, satellite orbit adjustment tasks are frequent and critical, each with a clear purpose, such as improving the throughput of a specific communication link. Traditional evaluation methods, like a passive observer, wait for orbit adjustments to complete, collect various data, and then perform calculations within a large and complex evaluation framework to try to find the cause of performance changes. This process is often time-consuming and easily affected by various irrelevant factors, especially when the physical characteristics of inter-satellite laser communication links deviate imperceptibly due to subtle changes in the relative position and attitude of satellites. In such cases, old evaluation standards quickly become ineffective. When the evaluation results are unclear, operators may adjust system parameters based on experience, which may introduce new uncertainties.

[0034] This invention first employs an "intent recognition" mechanism to actively "listen" to instructions issued by the ground control center. Once it "understands" the purpose behind the instruction (e.g., "This orbital adjustment is to improve communication throughput"), the system immediately enters a "focus mode." In this mode, it no longer waits for comprehensive data collection but directly requests the satellite to broadcast the latest precise position and attitude information. Based on this latest physical data and utilizing known physical laws of communication links, it calculates in real time the theoretically optimal communication throughput that this link can achieve under the current environment. This is akin to tailoring a completely new and most accurate "ideal standard" for the current environment.

[0035] Then, the system instructs the satellite to perform a rapid and precise "probe-based" test, measuring only the communication throughput metric—much like a doctor using a stethoscope to examine heart function, avoiding the time-consuming process of a comprehensive systemic examination. The system then compares this measured result with the previously calculated "ideal standard." If a discrepancy is found, the system intelligently adjusts its internal evaluation parameters, such as correcting the estimated link loss, to better reflect the current situation, based on the initially identified mission intent. This adjustment process is automated, eliminating the need for manual guesswork or intervention.

[0036] The working method of this invention contrasts sharply with generic evaluation schemes that lack "intent understanding" capabilities. Generic schemes are typically suitable for scenarios with relatively stable environments, slow changes, or low requirements for evaluation timeliness. They may rely on long-term accumulated data to establish a general performance benchmark, only initiating comprehensive troubleshooting when performance generally declines. However, in scenarios like space-based micro-nano satellite networking, where predetermined orbit adjustments are frequent, the physical environment changes rapidly, and there is an urgent need for calibration of a single communication throughput metric, the sluggish response and complex processing of multi-dimensional indicators in generic schemes can lead to the inability to obtain key information in a timely and accurate manner, thus affecting the deployment of high-value services. This solution, by proactively understanding mission intent and focusing on rapid, physically driven evaluation and self-calibration of a single indicator, can quickly adapt to environmental changes and provide reliable evaluation results—this is its unique advantage in specific scenarios.

[0037] Example Suppose we have a space-based micro-nano satellite network, where two satellites, A and B, need to establish a high-bandwidth laser communication link to transmit Earth observation data. To optimize the communication throughput of this link, the ground control center plans to perform a predetermined orbital adjustment on satellites A and B.

[0038] If this solution is not adopted: 1. Before orbit adjustment: The evaluation system relies on a "performance calibration reference dataset" based on past stable orbital configurations and attitudes. This dataset may have been obtained through long-term observation and testing when satellites A and B were in a certain standard relative position, for example, when the theoretical maximum throughput was measured to be 120 Mbps.

[0039] 2. Orbit Adjustment Execution: The ground control center sends orbit adjustment commands to satellites A and B. The satellites then perform maneuvers to change their orbits.

[0040] 3. Post-Adjustment Evaluation: After the orbital adjustment is completed, the relative positions and attitudes of satellites A and B undergo subtle changes. For example, the distance between them may change slightly, or there may be a minor deviation in laser pointing. Traditional evaluation systems continue to use the old 120 Mbps as a reference. When it begins collecting new telemetry data (such as signal strength, bit error rate, and latency), it finds that these data do not match the old reference values.

[0041] 4. Evaluation Challenges: Because the old reference dataset no longer represents the current environment, the evaluation system cannot quickly provide accurate communication throughput calibration results. It may need to wait for a large amount of new data to accumulate, or perform complex, multi-dimensional recalculation of metrics, a process that may take several hours or even longer.

[0042] 5. Manual Intervention: In order to obtain an evaluation result before the Earth observation data backhaul mission begins, operators may manually adjust the parameters in the "communication link quality quantification model" within the evaluation system based on experience. For example, they may artificially increase the value of "effective link gain" in an attempt to make the evaluation result look better. However, such adjustments are blind and may lead to inflated or understated evaluation results. For example, the system may output 110 Mbps, but this value is unreliable.

[0043] 6. Operational Risks: Due to the lack of reliable assessment results, the ground control center cannot determine whether the current link can safely carry high-value Earth observation data. Initiating data backhaul prematurely may result in data loss or transmission interruption, causing mission failure.

[0044] After adopting this solution: 1. Triggering of task intent perception and evaluation mode: The ground control center sends orbit adjustment commands to satellites A and B. This command data packet explicitly contains the semantic information "orbit maneuver command: target link throughput optimization." The command parsing units on satellites A and B (e.g., a software module running on the onboard processor, containing a predefined list of keywords and a set of syntax rules) immediately recognize this intent. Upon successful recognition, the module sends an internal signal to the satellite's master control unit, switching the evaluation system status of satellites A and B from "routine monitoring mode" to "communication performance self-calibration mode." This switch is automatic and real-time, avoiding manual judgment and operation.

[0045] 2. Dynamic reference throughput generation: Upon entering the "communication performance self-calibration mode," satellites A and B immediately acquire their latest orbital positions (e.g., satellite A: [X_A, Y_A, Z_A], satellite B: [X_B, Y_B, Z_B]) and attitude information (e.g., quaternions [q0_A, q1_A, q2_A, q3_A]) through their high-precision GPS receivers. This information is broadcast to each other in real time via the inter-satellite telemetry link and simultaneously sent to the "physical link characteristic calculation unit" at the ground control center.

[0046] Upon receiving this real-time data, the computing unit (e.g., a computing program running on a high-performance server) immediately performs calculations using preset physical parameters of the laser communication link (e.g., laser power P_tx=1W, receiver sensitivity S_rx=10^-12W, beam divergence angle θ_div=10 microradians, and modulation method QPSK).

[0047] First, calculate the real-time distance d between the two satellites.

[0048] Then, based on the distance d, laser wavelength λ, antenna gain G_tx, G_rx, and pointing loss L_pointing calculated from the attitude information, the link loss L_path is calculated.

[0049] Next, calculate the received optical power P_rx = P_tx L_path.

[0050] Finally, based on P_rx and S_rx, and considering the characteristics of QPSK modulation, the theoretically achievable maximum communication throughput R_max under the current geometry is calculated. For example, the calculation result is 118 Mbps. This 118 Mbps is the "dynamic calibration reference value" under the current environment, which accurately reflects the ideal communication capability after track adjustment, rather than relying on outdated old data.

[0051] 3. Rapid probe measurement of a single indicator: After the dynamic reference value is generated, the system instructs satellite A to send a series of pre-set "probe-type" test data packets to satellite B. For example, satellite A will continuously send 1000 UDP test data packets of size 1KB, embedding a sending timestamp in each data packet. After receiving the data packets, satellite B records the receiving timestamp and immediately sends back an acknowledgment message.

[0052] The "actual data comparison unit" (e.g., a data processing module in a ground control center) collects these test results and calculates the actual transmission rate, packet loss rate, and end-to-end latency. Assume the test results show that the actual communication throughput R_actual is 105 Mbps.

[0053] The unit then compared R_actual (105Mbps) with the dynamic calibration reference value R_max (118Mbps) and calculated the deviation to be (118 - 105) / 118 ≈ 11%.

[0054] 4. Evaluation parameter self-calibration and output: The "behavioral intent recognition module" combines the initially identified task intent of "increasing communication throughput" with an 11% throughput deviation. Since the task intent is to increase throughput, but the measured value is lower than the theoretical value, the system determines that there may be link attenuation that has not been fully perceived.

[0055] The system automatically adjusts its internal evaluation parameters according to preset calibration rules (e.g., if the deviation is between 5% and 15%, the "effective link gain" parameter in the evaluation model will be reduced by 5%). Specifically, it corrects an internal parameter used to calculate link loss in the evaluation model to make it closer to the actual performance of 105Mbps. For example, it adjusts a parameter called "environmental attenuation factor" from the default value of 1.0 to 0.95 to reflect the slight attenuation that may exist in the actual link.

[0056] Ultimately, the system outputs a self-calibrated, calibrated assessment of the current communication throughput: 105 Mbps, along with a confidence level (e.g., 98%). This result is fast, reliable, and accurate because it is based on the latest physical realities and measured data, and eliminates errors from human intervention.

[0057] With this reliable assessment result, the ground control center can confidently decide to initiate the high-value Earth observation data transmission mission to ensure its smooth progress.

[0058] In one embodiment, the present invention proposes an internal command parsing unit whose built-in logic can identify specific information patterns contained in remote control command data packets received from the ground control center, such as "orbital maneuver command: aimed at optimizing link throughput" or "attitude adjustment: to enhance communication performance." Once a clear mission objective of "improving specific communication link throughput" is identified, an internal signal is immediately generated, prompting the satellite master control logic unit to switch the evaluation system to "communication performance self-calibration mode." This is used to achieve rapid and reliable evaluation of a single performance indicator in a space-based micro-nano satellite network through intelligent means. However, in the daily operation of the space-based micro-nano satellite network, in order to respond to the ground control center's need to improve the performance of specific inter-satellite communication links, predetermined orbit adjustment tasks are periodically executed. When the ground control center sends remote control command data packets to the micro-nano satellites participating in the adjustment, the module responsible for processing commands inside the satellite receives and analyzes these data packets in real time. This module's built-in logic can identify specific information patterns in the commands, such as "orbital maneuver command: aimed at optimizing link throughput" or "attitude adjustment: to enhance communication performance." Once this module recognizes that the instruction explicitly states the mission objective of "improving the throughput of a specific communication link," it immediately generates an internal signal. This signal is then transmitted to the satellite's main control logic unit, prompting it to switch the evaluation system from routine monitoring to a specialized "communication performance self-calibration mode." In this mode, the system stops using previous evaluation parameters and begins preparing to initiate a new communication throughput evaluation and calibration process tailored to the current environment, thus laying the foundation for subsequent performance verification. Unfortunately, during a planned orbit adjustment mission, the satellite's orbital region experienced anomalous solar activity. This anomalous activity caused a transient increase in the flux of high-energy particles in the space environment. Although the intensity of this particle event was insufficient to cause permanent damage to the satellite hardware or a complete disruption of the communication link, its continued radiation effects introduced minute and random transient bit flips during data transmission in the telemetry and remote control links. Specifically, when the remote control command data packet issued by the ground control center, intended to trigger the "communication performance self-calibration mode," was transmitted to the satellite through the affected link, some bits in its data frame were corrupted. These errors are not concentrated in checksums or frame headers, but are scattered throughout the semantic payload of the instruction—specifically, within the key strings or encoded segments that describe the concrete task intent. Due to these subtle bit flips, the instruction parsing unit encounters challenges in its built-in keyword matching and syntax analysis logic when analyzing received data packets. The originally clear semantic pattern of "improving the throughput of a specific communication link" becomes ambiguous due to the change of one or more key characters. This semantic uncertainty prevents the instruction parsing unit from accurately associating it with the explicit task intent of "improving the throughput of a specific communication link."Therefore, it cannot generate the expected specific internal event signal to trigger the "communication performance self-calibration mode." Instead, the system may simply recognize it as a routine orbit parameter update command, thus remaining in "routine monitoring mode," or trigger a generic, non-specific "link status check mode" without a clear matching intent. Ultimately, although the orbit adjustment task has been completed, because the system has failed to enter the dedicated "communication performance self-calibration mode," it cannot initiate the dynamic reference throughput generation logic or perform rapid probe measurements of a single metric. This means that the system will continue to use assessment parameters and reference benchmarks established before the orbit adjustment, which are no longer accurate, to evaluate communication throughput. In the new physical environment, this evaluation result based on old parameters will deviate significantly from reality, preventing ground control centers from quickly obtaining a reliable, effectively calibrated assessment result for the current communication throughput, thus affecting the timely deployment and decision-making of high-value services.

[0059] Further technical challenges: Under the complex operating conditions of space-based micro-nano satellite networking, where the space environment is abnormal and remote control commands are susceptible to minor bit flips, how can we ensure that the system can still accurately understand the true intentions of the ground control center and quickly and reliably initiate the calibration and evaluation process for a single communication throughput indicator when the semantic payload of the command, which aims to improve the throughput of a specific communication link, has a slight deviation, causing the onboard command parsing unit to be unable to accurately identify the mission intent and trigger the "communication performance self-calibration mode" in a timely manner?

[0060] This invention relates to a calibration performance evaluation method for a single performance indicator. In space-based micro-nano satellite networks, it aims to address the problem of intent recognition failure caused by semantic ambiguity in remote control commands by establishing a multi-source intent-assisted judgment mechanism based on operational situational awareness. This ensures that even with subtle semantic ambiguity in the commands, the system can still accurately understand and initiate a rapid and reliable calibration evaluation process for a single communication throughput indicator. The method mainly includes the following core logic: 1. Logic for detecting semantic ambiguity in instructions and requesting auxiliary information: Description: After receiving remote control command data packets from the ground control center, the command parsing unit inside the satellite first performs a preliminary integrity check (e.g., CRC check). If the check passes, but its built-in keyword matching and syntax analysis logic finds that the matching degree is lower than a certain preset precise matching threshold (e.g., less than 90% string similarity) when attempting to identify preset task intentions such as "improving the communication throughput of a specific link," then the command semantics are deemed ambiguous. In this case, the command parsing unit does not immediately give up, but instead sends an auxiliary information request signal to the "operational situation awareness module" inside the satellite.

[0061] Implementation: The instruction parsing unit can be a state machine running on an onboard FPGA (Field-Programmable Gate Array) for rapid protocol parsing and CRC verification. When semantic ambiguity is detected, for example, by calculating the Levenshtein distance (edit distance) between the received instruction string and the standard intent string (such as "orbital maneuver instruction: target link throughput optimization"), if the distance exceeds a certain threshold (e.g., 10% of the string length), it is considered semantically ambiguous. At this time, the FPGA sends an interrupt signal to the connected microcontroller (e.g., STM32H7 series) via an internal bus (e.g., AXI bus), notifying it to start the "running situational awareness module" for auxiliary information collection.

[0062] Rationale: This phased processing approach first utilizes hardware acceleration for preliminary verification to ensure the physical integrity of the instruction. When the semantic match is inaccurate, it does not immediately reject the instruction but proactively requests more contextual information, avoiding decision-making stagnation caused by a single compromised instruction, thus demonstrating initial tolerance for ambiguity.

[0063] 2. Multi-source operational status information collection and standardization logic: Description: Upon receiving a request for auxiliary information, the "Operational Situational Awareness Module" immediately activates, collecting real-time data from multiple sensors and system components within the satellite. This data includes: current orbital position and attitude information provided by the high-precision navigation and positioning system; real-time operational status of the onboard laser communication terminal (e.g., whether it is active, transmit power, and receive gain settings); and future mission plans stored in the mission management unit (e.g., whether there are scheduled orbit adjustments or communication link optimization missions within the next 24 hours). All collected information is converted into a unified internal data format for subsequent fusion processing.

[0064] Implementation: The "Operational Situational Awareness Module" can be a software service running on an onboard microcontroller. It acquires the position vector [X, Y, Z] and quaternions [q0, q1, q2, q3] in the J2000 coordinate system from a high-precision GPS receiver via an SPI or I2C interface. It obtains status words such as operating mode and power level from the laser communication terminal via a CAN bus or Ethernet interface. Simultaneously, it reads a pre-loaded task schedule table from non-volatile memory (e.g., NAND flash memory), stored in JSON format, containing fields such as task ID, type (e.g., "orbit optimization"), objective (e.g., "link A throughput"), and scheduled start time. All data is encapsulated into a unified structure, for example: struct SatelliteContext { float position[3]; / / X, Y, Z float attitude[4]; / / q0, q1, q2, q3 int laserCommStatus; / / 0: Off, 1: Standby, 2: Active float laserTxPower; char currentMissionType

[32] ; / / eg, "OrbitOptimization" char currentMissionTarget

[32] ; / / eg, "LinkAThroughput" long missionStartTime; / / Unix timestamp}; Rationale: Collecting information from multiple sources provides a more comprehensive context, compensating for the limitations of single instructions. Standardized data formats ensure compatibility between information from different sources, laying the foundation for subsequent fusion and judgment.

[0065] 3. Logic for Intention Inference and Contextual Consistency Judgment: Description: After collecting multi-source operational situation information, the "Operational Situation Awareness Module" cross-compares it with fuzzy instruction semantic fragments provided by the instruction parsing unit to infer the most likely mission intent. This process involves judging the consistency between the instruction fragment and the current operational context. For example, if the fuzzy instruction contains keywords such as "link" or "performance," and the satellite's real-time attitude data shows recent maneuvers related to link optimization, and the mission plan also includes an "optimization" task for a specific link within the current time period, then the system will comprehensively judge and infer the mission intent of "improving the communication throughput of a specific link."

[0066] Implementation: The intent inference logic can employ a rule-based reasoning engine. For example, a series of rules can be defined: Rule 1: If the instruction semantic fragment contains “link” AND “performance” AND (“optimize” or “enhance”), AND the satellite laser communication terminal status is “active” or “standby”, AND there is a task in the mission plan of type “orbit optimization” with the objective of “link throughput”, AND the recent attitude change of the satellite (e.g., Euler angle change of more than 0.1 degrees in the past 1 hour) is related to orbit adjustment, then the intent is inferred to be “to improve the specific link communication throughput”.

[0067] The attitude change can be obtained by calculating the angle difference between the current attitude quaternion and the attitude quaternion from 1 hour ago: AngleDiff = 2 acos(abs(q_current.q0 q_previous.q0 + q_current.q1 q_previous.q1 + q_current.q2 q_previous.q2 + q_current.q3 q_previous.q3)) If AngleDiff is greater than a certain threshold, then the pose is considered to have changed significantly.

[0068] The inference engine runs on a microcontroller and performs logical judgments on the collected context information by traversing the rule set.

[0069] Reason: This logic improves the reliability of intent recognition by cross-verifying information from multiple sources. Even if a single instruction is ambiguous, its credibility increases significantly when multiple independent information sources point to the same intent, thus enabling accurate inference of the true intent.

[0070] 4. Intent confirmation and pattern triggering logic: Description: When the "Operational Situation Awareness Module" infers the task intent of "improving the communication throughput of a specific link" through situational consistency judgment, it generates an internal "intent confirmation signal" and transmits it to the main control logic unit. Upon receiving this signal, the main control logic unit ultimately triggers the evaluation system to enter the "communication performance self-calibration mode." Activation of this mode is specifically manifested by setting a specific position in the system's internal status register to a high level, suspending the reference to old evaluation parameters, and preparing to start a new parameter calculation process.

[0071] Implementation: The "intent confirmation signal" can be a simple Boolean flag, such as `intentConfirmed = true`. The main control logic unit (e.g., an RTOS task scheduler) periodically checks this flag. Once `intentConfirmed` is detected as true, it performs the following actions: 1. Set the `SELF_CALIBRATION_MODE_ACTIVE` bit in the system status register to 1.

[0072] 2. Set the pointer to the old evaluation parameter table to NULL, or set the access permissions of the old parameter table to read-only.

[0073] 3. Start an RTOS task named `DynamicReferenceGenerator`, which is responsible for the subsequent dynamic reference throughput calculation.

[0074] 4. Clear the `intentConfirmed` flag to prevent repeated triggering.

[0075] Rationale: This is a clear decision point, ensuring that the resource-intensive self-calibration mode is only initiated after sufficient verification from multiple sources. This mechanism avoids resource waste due to misjudgment and ensures the accurate initiation of the evaluation process.

[0076] Working Principle: The working principle of this solution can be figuratively understood as the system possessing the ability to "perceive the context," rather than merely "reading text." In the unique operating environment of space-based micro-nano satellite networking, the satellite's orbit, attitude, and mission plan are constantly changing. Simultaneously, the transmission link of remote control commands is susceptible to transient interference from space environment factors such as high-energy particles, leading to slight deviations in the semantic payload of the commands. Conventional evaluation solutions are like executors who only read the instruction manual. Upon receiving a command, they strictly match and execute it according to the text in the manual. If the text in the manual is illegible or partially corrupted, it cannot understand the true meaning of the command, thus choosing to stagnate or only processing it according to a general, non-specific procedure. This solution performs highly efficiently and accurately in scenarios where the environment is relatively stable, command transmission is reliable, and the mission intent is clear and does not change frequently. For example, in routine server performance monitoring in a ground-based data center, commands are usually transmitted through a stable internal network, and the mission intent is clear; in this case, precise command matching is sufficient to meet the requirements.

[0077] However, in the unique operating environment of space-based micro-nano satellite networks, command transmission links are susceptible to transient interference from space environment factors such as high-energy particles, leading to slight deviations in the semantic payload of commands. Simultaneously, satellites require frequent orbit adjustments and mission switches, each accompanied by rapid changes in the physical environment. In this situation, if the system adheres to a literal interpretation of commands, once the commands become ambiguous due to interference, the system will be unable to determine the mission intent, thus failing to initiate timely evaluation processes for specific performance indicators. This can cause delays in the deployment of high-value services and even lead to serious operational risks.

[0078] This approach differs; it functions more like an experienced operator. When the text of an instruction is ambiguous, it doesn't simply rely on the manual. Instead, it looks up at the machine's operational status, recalls recent operational plans, and even combines its own experience to determine the next course of action. It expands the "understanding" of instructions from merely "reading text" to "perceiving the context." The system first detects whether the instruction's semantics are ambiguous. If so, it doesn't immediately abandon it but actively collects the satellite's real-time physical status (such as the latest orbital position and attitude), the operational status of onboard equipment (such as whether the laser communication terminal is activated), and pre-loaded mission plans. This information directly reflects the highly dynamic environment, and there are mutually corroborating relationships between them. For example, even if the phrase "throughput optimization" in the instruction becomes ambiguous due to bit flipping, if the satellite has just completed an orbital adjustment aimed at optimizing communication, and the laser communication equipment is active, these interconnected pieces of information will collectively point to the intention of "improving communication throughput." By fusing multi-source information and judging contextual consistency, the system can construct a more comprehensive "physical profile" of the mission intent. This allows it to reliably infer the true intent of the ground control center even when instructions are not entirely clear, and promptly activate a "communication performance self-calibration mode." This approach makes the system more flexible and resilient in the face of uncertainties in the space environment, enabling it to maintain mission progress and control even with incomplete information.

[0079] In another embodiment, the present invention proposes a dynamic reference throughput generation logic to ensure that the evaluation reference value accurately reflects the latest physical reality after frequent orbit adjustments in the space-based micro-nano satellite network. However, after a high-precision orbit adjustment mission, although the main structures of satellites A and B have been operating stably according to the preset trajectory, a series of micro-vibrations with extremely low amplitudes and frequencies in the tens of hertz range were induced inside the satellite platform during this maneuver due to the slight deviation between the thrust vector of the thruster and the center of mass of the satellite, as well as propellant sloshing and other factors. These micro-vibrations are not structural faults, but rather a natural physical response of the internal elastic structure of the satellite as it dissipates residual energy after completing a large maneuver. For example, the control loop of the reaction wheel, which is usually used for attitude stabilization, may produce continuous and difficult-to-suppress jitter at extremely low amplitudes when dealing with such micro-vibrations. Further observation revealed that these platform-level micro-vibrations induced by orbit adjustment are transmitted and integrated into the optical pointing mechanism of the onboard laser communication terminal through the satellite support structure. Specifically, the micrometer-level pointing accuracy of the piezoelectric ceramic driving mirror assembly used for fine aiming inside the terminal was affected. However, because these vibrations are high-frequency and extremely small in amplitude (e.g., the jitter angle on optical components is on the order of microradians), they are far below the sampling rate and measurement accuracy limits of the satellite's primary attitude determination sensors (such as star sensors and inertial measurement units). This means that the overall satellite attitude data reported by the primary attitude sensors still appears "stable" and as expected, failing to capture these instantaneous, high-frequency micro-deviations on the optical pointing axis. As a result, when the physical link characteristic calculation unit receives the seemingly "normal" orbital position and attitude data reported by the satellite, it uses this data, combined with pre-stored physical design parameters such as laser emission power, receiver sensitivity, and beam divergence angle, to perform theoretical link loss calculations. When calculating pointing loss, this unit assumes that the laser beam has ideal pointing stability based on the reported attitude information. However, due to the aforementioned undetected optical pointing micro-vibrations, the actual laser beam does not point perfectly and stably to the center of the target receiver in space, but rather exhibits a high-frequency, periodic micro-"oscillation." This actual, uncaptured instantaneous pointing deviation causes the instantaneous effective received optical power in the link to be lower than the theoretically calculated value. For example, if theoretical calculations assume the beam center is precisely aligned with the receiver, but the actual beam spot jitters at the receiver edge in micro-radians, the average received power will decrease significantly. Ultimately, combining these fixed parameters with real-time geometric configuration data reported by the satellite, the computing unit, through a series of precise physical-level calculations, arrives at the theoretically maximum achievable communication throughput under current conditions. However, because its calculations fail to incorporate the actual instantaneous optical pointing deviations caused by micro-vibrations, this real-time calculated dynamic calibration reference value will systematically exceed the true maximum throughput achievable by the actual link.For example, theoretical calculations might yield 120Mbps, but due to actual jitter, the link cannot consistently reach this throughput at any given moment, and its effective average throughput might only be 110Mbps. This discrepancy means that the system will always face an overly optimistic benchmark in subsequent evaluations and comparisons, thus failing to accurately reflect the true performance of the current network and providing a reliable basis for subsequent service deployments.

[0080] Further technical challenges exist in space-based micro-nano satellite networking. When a predetermined orbit adjustment induces a small amount of high-frequency vibration on the satellite platform, and these vibrations further cause an instantaneous deviation in the optical pointing axis of the inter-satellite laser communication terminal that is not captured by the main attitude sensor, how can we ensure that the dynamic reference throughput generation logic can accurately perceive and quantify this subtle, non-explicit physical change, so as to generate a reliable dynamic calibration reference value for communication throughput that truly reflects the actual link performance?

[0081] This invention provides a calibration performance evaluation method applicable to a single performance index. In space-based micro / nano satellite networks, it aims to assess pointing stability based on the instantaneous fluctuation characteristics of received laser signals, addressing the problem of undetected optical pointing deviations caused by minute high-frequency vibrations of the satellite platform. This ensures that the generated dynamic calibration reference value accurately reflects the actual link performance. The method mainly includes the following core logic: 1. Logic for monitoring and extracting high-frequency received signal fluctuations: Description: A dedicated "high-frequency signal analysis module" is deployed on the receiving satellite (e.g., Satellite B). This module continuously monitors the instantaneous fluctuations in the power of the optical signal entering the laser communication receiver, or its electrical conversion signal, at a frequency much higher than the sampling rate of the satellite's main attitude sensor. It focuses on capturing rapid and periodic changes in signal intensity caused by minute instantaneous deviations in the optical pointing axis of the transmitter. By processing these high-frequency signal samples in real time, the module extracts feature data that can quantify instantaneous pointing stability.

[0082] Implementation: This module can consist of a high-sampling-rate analog-to-digital converter (ADC, e.g., sampling frequency of 100kHz or higher) and a digital signal processor (DSP, e.g., based on an FPGA or a high-performance embedded microcontroller such as the STM32H7 series). The ADC converts the analog signal output from the photodetector into a digital signal. The DSP performs real-time analysis on these digital samples, for example, calculating the root mean square (RMS) fluctuation of the instantaneous received power P(t) within a short time window T (e.g., 10 milliseconds). The calculation formula is: RMS_P = sqrt((1 / T) integral((P(t) - P_avg)^2 dt) from 0 to T) Where P_avg is the average received power within this time window. This RMS_P value is used as the "instantaneous pointing jitter index," which directly reflects the degree of instantaneous jitter of the laser beam on the receiver.

[0083] Reason: This method directly perceives pointing jitter at the "result" level of the signal, avoiding the complexity of directly measuring micro-vibrations at the transmitting end, and its high sampling rate can capture subtle, high-frequency changes that cannot be detected by the main attitude sensor.

[0084] 2. Spaceborne micro-pointing jitter feature transmission and ground coordination logic: Description: The "High-Frequency Signal Analysis Module" on the receiving satellite periodically transmits the extracted "Instantaneous Pointing Jitter Index" to the "Physical Link Characteristic Calculation Unit" at the ground control center via inter-satellite or satellite-to-ground telemetry links. This transmission contains only compressed feature data, not the original high-frequency signal samples, to maximize the saving of telemetry link bandwidth.

[0085] Implementation: The DSP encapsulates the calculated "instantaneous pointer jitter index" (e.g., a floating-point or fixed-point number) into a small telemetry data packet. This data packet is transmitted to the telemetry subsystem via the satellite's internal bus (e.g., CAN bus or Ethernet) and downlinked to the ground via an S-band or Ka-band link according to the CCSDS protocol standard. The "physical link characteristic calculation unit" (e.g., a server running data processing services) at the ground control center receives and parses these telemetry data packets, extracting the latest "instantaneous pointer jitter index".

[0086] Reason: By transmitting simplified feature data instead of the original data stream, the burden on the telemetry link is effectively reduced, ensuring the real-time nature and transmission efficiency of key information, which is crucial for resource-constrained space-based micro-nano satellite networking.

[0087] 3. Dynamic pointing loss correction and reference throughput recalculation logic: Description: After receiving the latest orbital position and attitude information reported by the satellite, as well as the "instantaneous pointing jitter index" from the receiving satellite, the "Physical Link Characteristic Calculation Unit" at the ground control center integrates these into the dynamic reference throughput calculation process. This unit uses the jitter index to correct the pointing loss portion in the theoretical link loss calculation in real time, thereby obtaining a more accurate dynamic calibration reference value for communication throughput that reflects the actual link performance.

[0088] Implementation method: The calculation unit first calculates the real-time distance d between the two satellites based on the orbital position and attitude information reported by the satellite, as well as the preset physical parameters of the laser communication link (such as laser wavelength λ, transmitting and receiving antenna gains G_tx, G_rx, and beam divergence angle θ_div), and preliminarily calculates the pointing loss L_pointing_nominal based on the main attitude information.

[0089] The unit then uses the received instantaneous pointing jitter index (RMS_P) to correct L_pointing_nominal. The correction method can be based on a pre-calibrated mapping or physical model. For example, a pointing loss correction factor, Correction_Factor = 1 - k RMS_P, can be defined, where k is a positive coefficient determined through ground experiments or initial on-orbit calibration, reflecting the degree of jitter's impact on loss.

[0090] The corrected pointing loss is L_pointing_corrected = L_pointing_nominal Correction_Factor.

[0091] Next, the corrected total link loss L_path_corrected = (λ / (4πd))^2 G_tx G_rxL_pointing_corrected is calculated.

[0092] Then, calculate the corrected received optical power P_rx_corrected = P_tx L_path_corrected.

[0093] Finally, based on P_rx_corrected and the receiver sensitivity S_rx, combined with the modulation and coding scheme, the theoretical maximum signal-to-noise ratio SNR_max_corrected and the corresponding theoretical maximum bit rate R_bit_max_corrected are calculated. For example, R_bit_max_corrected = B log2(1 + SNR_max_corrected), where B is the bandwidth.

[0094] This R_max_corrected is the corrected "dynamic calibration reference value", which accurately reflects the actual link communication capability considering the influence of micro-vibrations.

[0095] Reason: By directly incorporating the instantaneous pointing jitter information sensed by the receiver into the theoretical link loss calculation, this logic can compensate for subtle physical changes not captured by the main attitude sensor in real time, thereby ensuring that the generated dynamic calibration reference value is highly close to the actual instantaneous transmission capability of the link and avoiding systematic deviations in the evaluation results.

[0096] The working principle of this solution can be figuratively understood as the system possessing the ability to "sensitize minute vibrations." In space-based micro-nano satellite networks, even after a satellite completes its orbital adjustments and its overall attitude appears stable, there may still be minute vibrations within it that are imperceptible to the naked eye. This is similar to a person who, after strenuous exercise, may have slightly trembling fingers even after their body is stable. These minute vibrations are transmitted to the laser communication equipment, causing extremely subtle oscillations in the laser beam in space. Traditional assessment methods, like an observer only seeing the big picture, judge the laser beam's direction based on the satellite's overall attitude report, assuming it is completely stable. This method can provide sufficiently accurate assessments in relatively calm environments, when the equipment itself is very stable, or when the requirements for pointing precision are not so stringent. For example, when using large radio telescopes for observation on the ground, their massive structure itself has high stability, and the atmospheric environment is relatively controllable. In this case, assessments based on the overall attitude are reliable.

[0097] However, in the unique operating environment of space-based micro-nano satellite networking, the satellite platform frequently undergoes precise orbit adjustments, which can easily induce minute internal vibrations. Laser communication links have extremely high requirements for pointing accuracy; even micro-radian-level oscillations can significantly affect communication quality. In this situation, if the system adheres to a literal interpretation of the overall attitude, these subtle oscillations will be ignored, leading to an overly optimistic assessment of communication capabilities that fails to reflect the true situation.

[0098] This approach differs; it acts more like an experienced engineer. Once the laser communication link is established, it doesn't rely solely on the satellite's overall attitude report. Instead, it deploys a highly sensitive "stethoscope" on the receiving satellite. This "stethoscope" doesn't focus on the data transmitted by the laser beam, but rather on listening to the "pulse" of the laser signal itself—whether its instantaneous intensity is fluctuating rapidly. If the transmitting laser beam experiences subtle oscillations due to minute vibrations, the received signal strength will change rapidly and periodically. This "stethoscope" can detect these minute signal fluctuations and convert them into a simple "jitter index." This index acts as real-time feedback, telling us how stable the laser beam is during actual transmission. Subsequently, this "jitter index" is quickly transmitted to the computing unit at the ground control center. After integrating the satellite's overall position, attitude, and the laser's basic performance parameters, the computing unit uses this "jitter index" to correct its estimation of potential energy loss during signal transmission. In this way, the system can transform previously imperceptible minute physical vibrations into precise corrections to the communication capability assessment, thereby generating a reference value that truly reflects the actual transmission capability of the current link. This approach enables the system to be more flexible and resilient in the face of uncertainties in the space environment, allowing it to maintain mission progress and control even with incomplete information.

Claims

1. A calibration performance evaluation method applicable to a single performance indicator, characterized in that, Includes the following steps: (1) Receive remote control command, identify the task intent contained in the remote control command, and trigger evaluation mode according to the task intent; (2) In the evaluation mode, real-time physical state information of the satellite is obtained, and dynamic calibration reference values ​​of a single performance index under the current environment are calculated based on the physical model. (3) Perform probe-type measurement on the single performance index to obtain the measured results, and calculate the deviation between the measured results and the dynamic calibration reference value; (3) Based on the task intent and the deviation, adjust the evaluation parameters and output the calibration evaluation result of the single performance index after self-calibration.

2. The calibration performance evaluation method applicable to a single performance indicator according to claim 1, characterized in that, The specific operation of step (1) is as follows: The instruction parsing unit receives remote control instruction data packets and performs an integrity check on the remote control instruction data packets; When the instruction parsing unit detects that the semantic matching degree of the remote control instruction data packet is lower than a preset threshold, the instruction parsing unit sends an auxiliary information request to the operation situation awareness module. The operational situation awareness module collects real-time satellite orbital position information, attitude information, onboard laser communication terminal working status information, and future mission plan information. The intent inference logic cross-compares the instruction semantic fragments of the remote control instruction data packet with the multi-source operational status information to determine the consistency of the context and infer the task intent. When the intent inference logic infers that the task intent is to improve the communication throughput of a specific link, the intent inference logic generates an intent confirmation signal. The main control logic unit receives the intent confirmation signal and triggers the evaluation system to enter the evaluation mode.

3. The calibration performance evaluation method applicable to a single performance indicator according to claim 1, characterized in that, Step (2) involves the following operations: In the evaluation mode, real-time orbital position and attitude information of the satellite are acquired; The receiving satellite monitors the instantaneous fluctuations in the power of the optical signal entering the laser communication receiver or its electrical conversion signal; Extract the instantaneous pointer jitter index; The instantaneous jitter index is transmitted to the physical link characteristic calculation unit; The physical link characteristic calculation unit uses the satellite's real-time orbital position information, the attitude information, and the preset inter-satellite laser communication link physical design parameters, combined with the instantaneous pointing jitter index, to correct the pointing loss part in the theoretical link loss calculation; A dynamic calibration reference value for communication throughput in the current environment is calculated based on the corrected link loss.

4. The calibration performance evaluation method applicable to a single performance indicator according to claim 1, characterized in that, The specific operation of step (3) is as follows: Monitor the real-time bit error rate, signal strength fluctuations, and instantaneous changes in voltage and current of the satellite's internal power bus of the inter-satellite link; Identify whether there is instantaneous spatial environmental interference or internal electromagnetic interference during the measurement; When a transient interference event is detected, pause the current probe measurement; A new round of probe measurements is restarted to perform probe-based measurements on the single performance index and obtain the measured results. Calculate the deviation between the measured result and the dynamic calibration reference value; Record the time, type, and duration of the interference; Interference information is used as an auxiliary basis for judgment during the evaluation parameter self-calibration process.

5. The calibration performance evaluation method applicable to a single performance indicator according to claim 1, characterized in that, The specific operation of step (4) is as follows: S1. Identify satellite subsystems or link performance indicators that have a nonlinear coupling relationship with the evaluation parameters to be adjusted, and obtain a list of potential impacts; S2. Obtain real-time operational constraint data from the satellite health and safety management unit to obtain multi-dimensional constraint conditions; S3. Combining the mission intent, the deviation, the potential impact list, and the multi-dimensional constraints, simulate the impact of different adjustment strategies on the entire satellite system and select the optimal adjustment strategy. S4. Adjust the evaluation parameters according to the optimal adjustment strategy, and output the calibration evaluation result of the single performance index after self-calibration, and output a comprehensive report on the system health status.

6. The calibration performance evaluation method applicable to a single performance indicator according to claim 5, characterized in that, The step of identifying satellite subsystems or link performance indicators that have a nonlinear coupling relationship with the evaluation parameters to be adjusted, and obtaining a list of potential impacts, includes: In the evaluation mode, the system actively injects a preset micro-amplitude perturbation signal into the evaluation parameter to be adjusted; Simultaneously, the system monitors in real time the response data of multiple satellite subsystems or link performance indicators related to the evaluation parameters to be adjusted; Based on the response data, the system identifies nonlinear correlation patterns between the evaluation parameters to be adjusted and the performance indicators of the multiple satellite subsystems or links; The system dynamically updates the coupling relationship between evaluation parameters and satellite subsystem or link performance indicators based on the identified nonlinear correlation patterns; Based on the updated coupling relationships, the system generates a list of potential impacts that reflects the current actual operating status.

7. The calibration performance evaluation method applicable to a single performance indicator according to claim 6, characterized in that, The step of identifying the nonlinear correlation pattern between the evaluation parameter to be adjusted and the performance indicators of the multiple satellite subsystems or links includes: Multi-scale time-frequency analysis was performed on the response data to identify correlation patterns within different time scales and frequency ranges; Feature extraction is performed on the identified association patterns; Based on the differences in feature vectors, the association patterns are clustered to decouple complex patterns into independent association patterns. Based on the independent association pattern, the coupling relationship between the evaluation parameters and the performance indicators of the satellite subsystem or link is dynamically updated; Based on the updated coupling relationships, a list of potential impacts reflecting the current actual operating status is generated.

8. The calibration performance evaluation method applicable to a single performance indicator according to claim 5, characterized in that, The step of obtaining real-time operational constraint data from the satellite health and safety management unit to obtain multi-dimensional constraint conditions includes: Obtain real-time operational constraint data from the satellite health and safety management unit; Perform data integrity verification and data range validity checks on the real-time operational constraint data; When the data integrity verification or the data range validity check fails, a data retransmission request is sent to the satellite health and safety management unit. When the satellite health and safety management unit is unable to provide effective data retransmission, it retrieves backup constraint data that matches the current satellite operation mode from the preset backup constraint database. The backup constraint data is fused with the acquired valid real-time operational constraint data to obtain multi-dimensional constraint conditions.

9. The calibration performance evaluation method applicable to a single performance indicator according to claim 8, characterized in that, The steps for obtaining real-time operational constraint data from the satellite health and safety management unit include: In the evaluation mode, real-time operational constraint data is obtained from the satellite health and safety management unit; The real-time operational constraint data is sampled over multiple cycles to obtain a set of sampled data; Perform statistical analysis on the sampled data to determine whether there are any abnormal fluctuations in the sampled data that exceed a preset statistical threshold; When abnormal fluctuations are found in the sampled data, the constraint data item corresponding to the abnormal fluctuation is identified. Based on the constraint data items corresponding to the abnormal fluctuations, the abnormal data points are removed from the sampled data; The sampled data after removing the abnormal data points is smoothed to obtain corrected real-time running constraint data. The corrected real-time running constraint data is used as a multi-dimensional constraint condition.

10. A calibration performance evaluation system applicable to a single performance indicator, characterized in that, The system includes: The instruction receiving module is used to receive remote control instructions; The intent recognition module is used to identify the task intent contained in the remote control command; The mode triggering module is used to trigger the evaluation mode according to the task intent; The status acquisition module is used to acquire real-time physical status information of the satellite in the evaluation mode. The reference value calculation module is used to calculate the dynamic calibration reference value of a single performance index under the current environment based on the physical model. The measurement module is used to perform probe-type measurements on the single performance index to obtain the actual measurement results; The deviation calculation module is used to calculate the deviation between the measured result and the dynamic calibration reference value; The parameter adjustment module is used to adjust the evaluation parameters based on the task intent and the deviation. The results output module is used to output the calibration evaluation results of the single performance index after self-calibration.