Spacecraft attitude control decision-making method, system, equipment and medium based on LDPC confidence.

Abstract

This invention relates to the field of aerospace telemetry, tracking, and command (TT&C) communication technology, specifically to a spacecraft attitude control decision-making method, system, device, and medium based on LDPC confidence level. The method includes: acquiring and processing the intermediate frequency signal of the target spacecraft, extracting attitude control command data frames; performing LDPC iterative decoding on the attitude control command data frames, extracting reliability measurement information during the decoding process, and calculating the confidence feature value of the current command data frame; constructing a confidence time series sequence by combining the confidence feature value of the current attitude control command data frame with historical confidence feature values, analyzing and processing the confidence time series sequence to identify channel state risk types; determining an adaptive dynamic decision threshold based on the channel state risk type, and making a confidence decision on the current attitude control command data frame in conjunction with the confidence feature value; and executing the corresponding target spacecraft attitude control strategy based on the confidence decision result. The aim is to improve the reliability and security of spacecraft attitude control command execution.

Description

Technical Field

[0001] This invention relates to the field of aerospace telemetry, tracking, and command (TT&C) communication technology, specifically to a spacecraft attitude control decision-making method, system, equipment, and medium based on LDPC confidence. Background Technology

[0002] The core objective of spacecraft attitude control is to maintain preset attitude parameters during orbital operation to ensure mission performance. For example, communication satellites need attitude control to align their antennas with Earth for data transmission and remote sensing, while deep space probes rely on precise attitude pointing to observe target celestial bodies. If the attitude deviates from the preset range or if there are abnormal attitude phenomena in the controlled target, the mission may be interrupted. To address this, spacecraft are typically equipped with an autonomous attitude monitoring and correction system consisting of attitude sensors, an attitude control computer, and actuators. Meanwhile, ground stations send attitude control commands to the spacecraft via a space-to-ground link to collaboratively complete attitude adjustments.

[0003] Currently, in the space-to-ground transmission and validity verification of attitude control commands, existing technologies typically employ Cyclic Redundancy Check (CRC) or post-decoding verification mechanisms to determine command validity. If the verification passes, the command is executed directly, and attitude anomalies are corrected by adjusting thruster operating status, restarting attitude control equipment, or adjusting momentum wheel speed. This spacecraft attitude control decision-making approach has certain limitations. Firstly, the CRC check or post-decoding verification mechanisms relied upon by existing technologies are essentially binary judgments, only achieving a qualitative judgment of "valid / invalid" commands. In scenarios with low signal-to-noise ratios in the space-to-ground link, erroneous command bitstreams may pass the verification without being detected. Execution of such erroneous commands will lead to attitude control deviations and exacerbate the risk of attitude anomalies. Secondly, the lack of quantitative assessment of the reliability of attitude control commands makes it difficult to provide the attitude control system with a forward-looking decision-making basis based on command reliability. The control system can only detect anomalies through attitude sensor feedback after command execution, making it difficult to avoid attitude control problems caused by erroneous commands in advance, thus restricting the safety and reliability of spacecraft attitude control. Summary of the Invention

[0004] To improve the reliability and safety of spacecraft attitude control command execution, this invention provides a spacecraft attitude control decision-making method, system, device, and medium based on LDPC confidence. The specific technical solution adopted is as follows:

[0005] The first aspect of the present invention provides a spacecraft attitude control decision-making method based on LDPC confidence, the method comprising:

[0006] Acquire and process the intermediate frequency signal of the target spacecraft to extract attitude control command data frames;

[0007] The attitude control command data frame is subjected to LDPC iterative decoding to extract reliability measurement information during the decoding process and calculate the confidence feature value of the current command data frame.

[0008] The confidence feature value of the current attitude control command data frame is constructed into a confidence time series with the historical confidence feature value. The confidence time series is then analyzed and processed to identify the channel state risk type.

[0009] The adaptive dynamic decision threshold is determined based on the channel state risk type, and the confidence decision is made on the current attitude control command data frame in combination with the confidence feature value.

[0010] The corresponding target spacecraft attitude control strategy is executed based on the credibility decision results.

[0011] Further, the intermediate frequency signal of the target spacecraft is acquired and processed to extract attitude control command data frames, including:

[0012] Carrier synchronization and symbol synchronization processing are performed on the intermediate frequency signal, and the baseband information frame code stream is obtained by demodulation.

[0013] The frame header field is parsed from the baseband information frame bitstream, and the frame header field contains at least an instruction type identifier and an LDPC bitrate identifier;

[0014] The attitude control command data frame is selected from the baseband information frame bitstream based on the command type identifier in the frame header field, and the LDPC bit rate corresponding to the attitude control command data frame is determined based on the bit rate identifier in the frame header field.

[0015] Further, the attitude control command data frame is subjected to LDPC iterative decoding to extract reliability measurement information during the decoding process and calculate the confidence feature value of the current command data frame, including:

[0016] Configure the parity check matrix parameters and iteration number of the LDPC decoder according to the bit rate of the attitude control command data frame;

[0017] The normalized minimum sum algorithm is used to iteratively decode the attitude control command data frame, and the posterior probability log-likelihood ratio of each data bit is extracted as a reliability metric after decoding.

[0018] Feature extraction is performed on the absolute value of the posterior probability log-likelihood ratio of all data bits in the frame, and the result is used as the confidence feature value characterizing the overall reliability of the frame.

[0019] Furthermore, the confidence feature values ​​of the current attitude control command data frame and historical confidence feature values ​​are used to construct a confidence time series. This confidence time series is then analyzed to identify channel state risk types, including:

[0020] A sliding window mechanism is used to store the confidence feature values ​​of multiple historical frames, and a confidence time series sequence is constructed in chronological order.

[0021] Missing data in the confidence time series is detected, and the missing data is divided into continuous missing data and non-continuous missing data according to the continuity of the missing data;

[0022] For continuously missing data, a linear interpolation algorithm with SNR correction is used to fill the missing data based on the effective confidence feature values ​​before and after the missing segment and the trend of channel SNR changes. For non-continuously missing data, a curve fitting algorithm is used to fill the missing data based on the correlation between historical confidence feature values, frame number and channel SNR.

[0023] The confidence time series after data filling is smoothed, and the channel state risk type is identified by combining abrupt change detection and trend prediction. The risk types include random noise, continuous interference and sudden interference.

[0024] Furthermore, an adaptive dynamic decision threshold is determined based on the channel state risk type, including:

[0025] The signal-to-noise ratio of the link between the target spacecraft and the ground station is acquired in real time, and the basic threshold value is calculated based on the Shannon channel capacity.

[0026] The basic threshold value is adjusted according to the channel state risk type to obtain the adaptive dynamic decision threshold.

[0027] Furthermore, a confidence decision is made on the current attitude control command data frame by combining confidence feature values, including:

[0028] If the confidence feature value of the current attitude control command data frame is not lower than the adaptive dynamic decision threshold and the channel state risk type is random noise, it is determined to be a trustworthy command.

[0029] If the confidence feature value of the current attitude control command data frame is not lower than the adaptive dynamic decision threshold, and the channel quality is judged to be deteriorating by trend prediction, it is determined to be a conditionally reliable command.

[0030] If the confidence feature value of the current attitude control command data frame is lower than the adaptive dynamic decision threshold, or if the channel state risk type is abnormal, it is determined to be an abnormal command.

[0031] Furthermore, the execution of the corresponding target spacecraft attitude control strategy based on the confidence decision result includes:

[0032] If it is a trusted instruction, the attitude control instruction data frame is sent to the attitude control system for execution;

[0033] If it is a conditionally trusted instruction, send an instruction to the ground station while executing the current instruction to trigger the attitude control instruction in the next frame to adjust the LDPC code rate;

[0034] If it is an abnormal command, the attitude control command data frame is intercepted and a retransmission request is sent to the ground station. The request contains the position information of the low confidence bits in the frame. After receiving the retransmission data returned by the ground station, the corresponding bits in the original command frame are replaced and local iterative decoding verification is performed. If the confidence feature value is still lower than the adaptive dynamic decision threshold after verification, the onboard preset conservative safety command is triggered.

[0035] The second aspect of the present invention provides a spacecraft attitude control decision system based on LDPC confidence level, used to execute the spacecraft attitude control decision method based on LDPC confidence level described in the first aspect of the present invention, the system comprising:

[0036] The signal acquisition and processing module is configured to acquire and process the intermediate frequency signal of the target spacecraft and extract attitude control command data frames.

[0037] The LDPC decoding and feature extraction module is configured to perform LDPC iterative decoding on the attitude control command data frame, extract reliability measurement information during the decoding process, and calculate the confidence feature value of the current command data frame.

[0038] The channel state analysis module is configured to construct a confidence time series sequence from the confidence feature value of the current attitude control command data frame and the historical confidence feature value, and to analyze and process the confidence time series sequence to identify the channel state risk type.

[0039] The credibility decision module is configured to determine an adaptive dynamic decision threshold based on the channel state risk type, and combine the confidence feature value to make a credibility decision on the current attitude control command data frame.

[0040] The attitude control execution module is configured to execute the corresponding target spacecraft attitude control strategy based on the confidence decision results.

[0041] The third aspect of the present invention provides an electronic device, the electronic device comprising: a processor and a memory communicatively connected to the processor; wherein the memory stores instructions executable by the processor, the instructions being executed by the processor to enable the processor to perform the steps of the spacecraft attitude control decision method based on LDPC confidence as described in the first aspect of the present invention.

[0042] The fourth aspect of the present invention provides a computer-readable storage medium storing a program for implementing a spacecraft attitude control decision method based on LDPC confidence. The program for implementing the spacecraft attitude control decision method based on LDPC confidence is executed by a processor to implement the steps of the spacecraft attitude control decision method based on LDPC confidence as described in the first aspect of the present invention.

[0043] The present invention has the following beneficial effects:

[0044] This invention provides a spacecraft attitude control decision-making method based on LDPC confidence levels. In the spacecraft telemetry and control communication link, the soft information generated during LDPC decoding is transformed into real-time, quantitative decision-making criteria, thus constructing a secure decision-making closed loop preceding traditional CRC checksums. First, by extracting confidence feature values ​​from the command data frames from the LDPC decoder and constructing a confidence time-series sequence with historical data for analysis, the reliability of uplink commands is quantitatively assessed. This allows for capturing the changing trends of command reliability from multiple frame dimensions, avoiding the one-sidedness of single-frame data judgment. Then, the confidence time-series sequence is combined with an adaptive dynamic decision threshold for reliability decision-making. This decision-making process can identify potential risks based on time-series trends and dynamically adjust the judgment criteria according to channel conditions, reducing the probability of erroneous commands passing the judgment under harsh channel conditions such as low signal-to-noise ratio. It can effectively identify potential erroneous commands that traditional CRC checksums cannot detect before command delivery and execution, thereby improving the reliability and security of spacecraft attitude control command execution. This method transforms the reliability information of the physical layer of the communication link into control decision-making criteria, providing technical assurance for the safe operation of spacecraft in harsh channel environments. Attached Figure Description

[0045] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart of a spacecraft attitude control decision-making method based on LDPC confidence, provided in one embodiment of the present invention.

[0047] Figure 2 This is a schematic diagram of the structure of a spacecraft attitude control decision system based on LDPC confidence, provided in an embodiment of the present invention. Detailed Implementation

[0048] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a spacecraft attitude control decision-making method, system, device, and medium based on LDPC confidence level proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0050] The following description, in conjunction with the accompanying drawings, details a specific scheme for a spacecraft attitude control decision-making method, system, device, and medium based on LDPC confidence provided by this invention.

[0051] Please see Figure 1 This document illustrates a flowchart of a spacecraft attitude control decision-making method based on LDPC confidence, according to an embodiment of the present invention. The method includes:

[0052] Step S100: Acquire and process the intermediate frequency signal of the target spacecraft and extract attitude control command data frames;

[0053] In some embodiments, an intermediate frequency receiver adapted to aerospace telemetry and control scenarios is used to receive the 70MHz intermediate frequency signal transmitted downlink from the target spacecraft. This signal is a BPSK signal that has been spread spectrum or high code rate modulated, conforming to the signal specifications of the integrated telemetry and control data transmission system, and its expression is as follows:

[0054]

[0055] In the formula, express The amplitude of the intermediate frequency signal received at any given time; A is the amplitude of the intermediate frequency signal, the value of which is determined by the spacecraft's transmit power and the transmission loss of the space-to-ground link; , is the carrier frequency of the intermediate frequency signal; The phase of the signal is modulated, including attitude control command data; This refers to the initial phase deviation of the carrier wave. Additive white Gaussian noise is used to simulate random interference in a channel.

[0056] Step S100 specifically includes:

[0057] Step S110: Perform carrier synchronization and symbol synchronization processing on the intermediate frequency signal, and demodulate to obtain the baseband information frame code stream; specifically, the intermediate frequency receiver eliminates the initial phase deviation and frequency deviation of the carrier through the carrier acquisition module to achieve carrier synchronization;

[0058] In some embodiments, a Costas ring-based acquisition algorithm is employed. First, the frequency offset is estimated using a frequency search module, and the correlation value between the signal and the local carrier at different frequencies is calculated. When the correlation value exceeds a preset threshold, frequency acquisition is considered successful. After frequency locking, the phase difference between the received signal and the local regenerated carrier is calculated using the phase detector of the Costas ring, expressed as:

[0059]

[0060] In the formula, The locally regenerated carrier phase is generated by a local voltage-controlled oscillator and is used to synchronize with the received signal carrier phase. The phase difference is input to the loop filter to smooth noise, and then the voltage-controlled oscillator (VCO) is used to adjust the local carrier phase, gradually adjusting until... Approaching To achieve carrier phase synchronization;

[0061] In some embodiments, to eliminate timing deviations in symbol transmission and ensure that the sampling time is aligned with the optimal symbol decision point, symbol synchronization employs a synchronization algorithm based on Gardner timing error detection: the signal after carrier synchronization... Oversampling was performed, with the sampling rate configured to be 2 or 4 times the symbol rate, ultimately resulting in a sampled sequence. , The sampling point number;

[0062] Calculate timing error:

[0063]

[0064] in, This is the value of the current sampling point. , For adjacent sampling point values, timing error This reflects the deviation between the current sampling time and the optimal decision point of the symbol; The input loop filter and output control signal adjust the sampling clock phase to ensure that subsequent sampling points are accurately aligned with the symbol center, thereby achieving symbol synchronization.

[0065] In some embodiments, binary phase-shift keying (BPSK) demodulation is performed on the synchronized signal to recover the baseband symbol sequence. Specifically, the demodulation logic involves multiplying the symbol-synchronized signal with a local regenerated carrier, low-pass filtering the result, and then outputting the baseband symbol sequence from a decision unit. Finally, the sequence is concatenated into a baseband information frame stream according to the frame structure. Each information frame contains a fixed length of code elements, providing a unified format for parsing subsequent frames.

[0066] Step S120: Parse the frame header field from the baseband information frame bitstream. The frame header field includes at least an instruction type identifier and an LDPC bitrate identifier. Specifically, parse the baseband information frame bitstream output in step S110. Divide into independent information frames according to the preset frame length. , representing the Each information frame Includes frame header and data segment;

[0067] In some embodiments, two key fields are parsed:

[0068] The bitrate identifier field is a binary encoded field, occupying 2 to 4 code elements. The specific length is determined by the number of compatible bitrates and is used to identify the LDPC bitrate corresponding to the current frame. The encoding of this field corresponds to the four bitrates specified in the CCSDS standard. After parsing this field, the LDPC parity check matrix parameters corresponding to the current bitrate can be obtained by calling the preset "bitrate-submatrix parameter mapping table".

[0069] The instruction type identifier field is a binary encoded field, occupying 1 to 2 code elements, used to distinguish the type of frame. "1" indicates that the frame is an "attitude control instruction data frame", such as control instructions for spacecraft attitude adjustment and orbit correction, while "0" indicates that it is an "ordinary data transmission frame", such as non-control data such as remote sensing data and status monitoring data.

[0070] Step S130: Select attitude control command data frames from the baseband information frame bitstream based on the command type identifier in the frame header field, and determine the LDPC bit rate corresponding to the attitude control command data frames based on the bit rate identifier in the frame header field; specifically, select frames with the identifier "1" based on the command type identifier field, which are the attitude control command data frames. Ordinary data transmission frames marked "0" are removed to avoid them occupying computing resources for subsequent LDPC decoding, confidence analysis, etc., thus reducing the onboard computing load.

[0071] In some embodiments, the LDPC bitrate and decoding parameters corresponding to the frame are determined by querying the preset "bitrate-submatrix parameter mapping table" based on the bitrate identifier field, including the number of rows and columns of the parity check matrix and the submatrix dimension. The filtered instruction frames are then temporarily stored in the cache unit to facilitate the quick retrieval of the decoding parameters of the corresponding frame during subsequent multi-frame timing analysis, ensuring the consistency of multi-frame data processing.

[0072] In summary, step S100, through intermediate frequency signal demodulation and frame parsing, realizes the transformation from the spacecraft's downlink raw intermediate frequency signal to attitude control decision data. On the one hand, carrier synchronization and symbol synchronization eliminate link distortion, ensuring the integrity and accuracy of the baseband information frame code stream, providing high-quality data input for subsequent confidence analysis. On the other hand, frame header field parsing accurately filters attitude control command frames, avoiding invalid data from consuming resources, while pre-locking the LDPC code rate and parity check matrix parameters, improving the efficiency of the decoding process. This forms the preconditions for attitude control decisions, providing a reliable data foundation for subsequent attitude control decisions based on LDPC confidence.

[0073] Step S200: Perform LDPC iterative decoding on the attitude control command data frame, extract the reliability measurement information in the decoding process, and calculate the confidence feature value of the current command data frame;

[0074] Step S200 specifically includes:

[0075] Step S210: Configure the parity check matrix parameters and iteration count of the LDPC decoder according to the bit rate of the attitude control command data frame; specifically, this step, based on the bit rate of the attitude control command data frame determined in step S130, completes the hardware parameters and iteration strategy configuration of the LDPC decoder, laying the foundation for subsequent accurate decoding. The specific implementation process is as follows:

[0076] Based on the LDPC code rate parsed in step S130, the preset "code rate-submatrix parameter mapping table" is called to configure the core parameters of the quasi-cyclic LDPC parity check matrix: the parity check matrix consists of multiple submatrices, each of which is a cyclic shift matrix or a zero matrix; different code rates correspond to different parameter combinations. The parameter configuration is based on ensuring that the row weight and column weight of the parity check matrix meet the decoding performance requirements, and can be selected according to actual needs.

[0077] In some embodiments, the number of iterations is dynamically adjusted based on the channel signal-to-noise ratio (SNR) synchronously estimated during demodulation in step S110. The value range is 8 to 12. For example, when the SNR is low and the channel quality is poor, the number of iterations is set to 10 to 12 to improve decoding accuracy and reduce bit errors by increasing the number of iterations. When the SNR is high and the channel quality is good, the number of iterations is set to 8 to 9 to reduce on-board computing power consumption while ensuring decoding performance. The number of iterations can be written to the decoder's control register and updated before each attitude control command data frame is decoded to adapt to the time-varying characteristics of the channel.

[0078] Step S220: The attitude control command data frame is iteratively decoded using the normalized minimum sum algorithm, and the posterior probability log-likelihood ratio of each data bit is extracted as a reliability metric after decoding. Specifically, the variable node message and the channel likelihood ratio are initialized, and the channel likelihood ratio is calculated from the baseband symbol signal demodulated in step S110. For each verification node, the likelihood ratio message transmitted by its associated variable node is collected, and the feedback message is updated using the normalized MSA algorithm, the expression of which is:

[0079]

[0080] In the formula, This is a normalization factor used to correct the approximation error of the traditional minimum sum algorithm; Indicates and verifies nodes The set of all associated variable nodes is determined by the positions of the non-zero elements in the LDPC check matrix, where the matrix is... The column number corresponding to the non-zero element in a row is the variable node number; Indicates excluding the current variable node The set after; This is a symbolic function that outputs the product of all associated message symbols. This is a minimum value function that outputs the minimum absolute value of all associated messages. To verify the feedback message from the node to the variable node, the subscript... Representing the The nth check node, corresponding to the nth check matrix of the LDPC check matrix. Okay, subscript Representing the The nth variable node, corresponding to the nth variable node in the check matrix. The column, with its numerical value reflecting the check node's relationship to the variable node. The degree of confidence in the judgment; For set Any variable node in the set, that is, a node in the set where... External variable node index; For variable nodes To the verification node The input message, which is calculated by the variable node based on the initial likelihood ratio of the channel and feedback messages from other check nodes, reflects the variable node's... The initial confidence level of its own judgment result; by integrating the input messages of the associated variable nodes, this formula generates a feedback message that is both reliable and low-complexity, balancing the accuracy of LDPC decoding with the difficulty of on-board hardware implementation.

[0081] For each variable node n, collect the feedback messages transmitted by its associated verification node m. Update the message output to the verification node. Its expression is:

[0082]

[0083] In the formula, For variable nodes The set of all associated verification nodes, This indicates that the current verification node should be removed. The set after; In order to be with the first The set of all check nodes with connections to each variable node is determined by the positions of the non-zero elements in the LDPC check matrix. The row number corresponding to the non-zero element in the column is the check node number; summation operation The summation of messages from all associated verification nodes, combined with the initial likelihood ratio. Implement message optimization; variable node updates and check node updates are executed alternately until the preset number of iterations is reached or decoding converges. For variable nodes The initial likelihood ratio of the channel is calculated and generated from the baseband symbol signal obtained in step S110; For set Any one of the verification nodes in the process; For verification nodes To the variable node The feedback message, the magnitude of which reflects the verification node's response to the variable node. The strength of the correction to the judgment;

[0084] After decoding converges, access the variable node computation unit and extract each data bit. The posterior probability log-likelihood ratio This refers to reliability metrics, and the formula is:

[0085]

[0086] In the formula, For bits The logarithm ratio of the posterior probability of a decision of 0 to the logarithm of the posterior probability of a decision of 1: If The decision for bit n is 0. The larger the value, the higher the confidence level of the judgment being 0; if Bit n is judged to be 1. The larger the absolute value, the higher the confidence level of the decision being 1; extract all bits. An LLR array is formed and temporarily stored in the decoder output buffer.

[0087] Step S230: Perform feature extraction operation on the absolute value of the log-likelihood ratio of the posterior probability of all data bits in the frame, and use the operation result as the confidence feature value characterizing the overall reliability of the frame; in this embodiment, the feature extraction operation includes taking the minimum value, taking the arithmetic mean, and calculating the information entropy;

[0088] In some embodiments, the minimum absolute value of all elements in the LLR array is taken as the confidence feature value, which reflects the confidence of the least reliable bit in the frame, focuses on the weak links in the frame, captures the risk of single bit errors, such as low confidence bits that cannot be detected by CRC check, and adapts to high security requirements such as attitude roll emergency correction.

[0089] In some embodiments, the arithmetic mean of the absolute values ​​of all elements in the LLR array is taken as the confidence feature value, which reflects the average reliability level of the whole frame, adapts to common attitude control scenarios such as track fine-tuning, and avoids misjudgment caused by individual random noise bits.

[0090] In some embodiments, for each bit, by The posterior probability of a decision of 0 is calculated as an exponential function. The larger the value, the closer the posterior probability is to 1, and the higher the bit certainty. Then, the single-bit confidence entropy is calculated, and the sum of the confidence entropies of all bits is obtained to get the feature value of the whole frame. This method takes into account the bit confidence distribution, avoids the oversensitivity of the extreme value method and the local masking of the mean method in the above embodiments, and is suitable for complex time-varying channel scenarios such as deep space exploration.

[0091] Finally, based on the spacecraft's current mission scenario, the corresponding feature value calculation method is automatically selected, and the selected confidence feature value is temporarily stored in the historical confidence time series storage unit, providing data support for the time series analysis in step S300. Step S200 realizes the transformation from attitude control command data frames to quantized reliability indicators; based on the code rate determined in step S100, the LDPC decoder parameters are dynamically configured, and then the normalized MSA algorithm is used to ensure decoding accuracy under the constraints of on-board computing power. The extracted LLR array provides a direct basis for bit-level reliability assessment; the entire process is compatible with the integrated telemetry, tracking, and command architecture, improving the accuracy of attitude control command reliability assessment while ensuring compatibility with the spacecraft's existing hardware resources.

[0092] Step S300: Construct a confidence time series from the confidence feature value of the current attitude control command data frame and the historical confidence feature value, and analyze and process the confidence time series to identify the channel state risk type;

[0093] Step S300 specifically includes:

[0094] Step S310: Use a sliding window mechanism to store the confidence feature values ​​of multiple historical frames and construct a confidence time series sequence in chronological order;

[0095] In some embodiments, the system first presets a fixed length of The historical confidence sequence cache unit is used to store the target spacecraft's most recent... Confidence feature value of frame attitude control command data frame; The value can be set based on the spacecraft's storage resources and the real-time requirements of attitude control. This buffer unit employs a first-in, first-out (FIFO) sliding window mechanism; that is, when a new frame's confidence feature value is stored, if the buffer is already full... If a frame is not stored, the earliest stored frame will be automatically deleted to ensure that the timing sequence always reflects the latest changes in channel and command reliability.

[0096] In some embodiments, the confidence feature value of the current attitude control command data frame calculated in step S200 is defined as subscript Indicates the time sequence number of the current frame; The corresponding frame number, bitrate, and acquisition time are written into the historical confidence sequence buffer unit; all stored confidence feature values ​​are extracted from the buffer unit, arranged in ascending order of frame number, forming a sequence of fixed length. Confidence time series: ,in, The earliest frame in the buffer unit, i.e., the first frame. Frame confidence feature value; The confidence feature value of the previous frame is used for the current frame.

[0097] Step S320: Detect missing data in the confidence time series and divide it into continuous missing data and non-continuous missing data according to the continuity of the missing data; specifically, traverse the confidence time series generated in step S310. For each element in the array, check if there is a null or invalid value;

[0098] In some embodiments, invalid values ​​can be determined based on the feature value calculation logic in step S200. The confidence feature values ​​derived from the absolute value of LLR are all non-negative numbers. If a negative number appears, it is determined to be invalid. If a frame is empty or invalid, the frame is marked as a data missing frame and its frame number is recorded.

[0099] In some embodiments, if the marked missing frame number satisfies the following condition: the next frame number = the previous frame number + 1, and the number of consecutive missing frames is greater than or equal to a preset threshold, then it is determined to be consecutive missing data, and the start frame number and end frame number of the missing window are recorded; if the missing frame number does not satisfy the condition of consecutive increment or the number of consecutive missing frames is less than the preset threshold, then it is determined to be non-consecutive missing data, and the discrete sequence number set of all missing frames is recorded.

[0100] Step S330: For continuously missing data, a linear interpolation algorithm with signal-to-noise ratio correction is used to fill the missing data based on the effective confidence feature values ​​before and after the missing segment and the trend of channel signal-to-noise ratio changes; for non-continuously missing data, a curve fitting algorithm is used to fill the missing data based on the correlation between historical confidence feature values, frame number and channel signal-to-noise ratio.

[0101] In some embodiments, for consecutive missing data, linear interpolation is used to fill in the missing data by combining the effective feature values ​​before and after the missing segment with the channel signal-to-noise ratio (SNR) change trend, thus avoiding the filled value deviating from the actual channel state. First, the effective frame data before and after the consecutive missing window is extracted: the confidence feature value of the frame before the missing window. and the real-time SNR value of the corresponding frame The SNR is synchronously estimated by the demodulation process in step S100, reflecting the channel quality during the transmission of this frame; the confidence feature value of the frame after the missing window. , The number of consecutively missing frames and the real-time SNR value of the corresponding frames. ;

[0102] Define SNR correction factor Adaptive channel SNR variation trend:

[0103] like Then the channel quality will deteriorate. Use values ​​less than 1 to mitigate the increasing trend of the fill value and avoid overestimating the channel quality;

[0104] like Then the channel quality is stable. A value of 1 requires no correction;

[0105] like Then the channel quality shows an improving trend. Use values ​​greater than 1 to reinforce the growth trend of the filler value and better reflect the actual channel conditions;

[0106] In some embodiments, the formula for calculating the fill value in the missing window is:

[0107]

[0108] In the formula, For the missing window Frame confidence feature value padding. The frame offset index within the missing window, i.e., the first to the last consecutively missing frames. frame; The change in feature value per unit frame interval reflects the overall slope of change of valid frames before and after the loss. The number of frames between the current filling frame and the frame preceding the missing window;

[0109] In some embodiments, for non-continuous missing data, high-precision filling can be achieved by fitting cubic spline curves based on the correlation between confidence feature values, frame numbers, and SNR in historical data. First, complete confidence time-series data stored during the historical operation of the target spacecraft is extracted. Each frame of data contains three dimensions: frame number, SNR value, and confidence feature value, thus constructing a training set. Then, based on the training set, a cubic spline interpolation algorithm is used to fit the function. The function uses the frame number Taking the channel SNR as input, the output is the predicted confidence feature value. The fitting logic is as follows: divide the frame number and SNR into segments, and construct a cubic polynomial local fitting function within each segment, ensuring the continuity of the function values ​​and first derivatives of adjacent segments, so that... To accurately reflect the inherent correlation, for non-contiguous missing frames, the SNR values ​​of adjacent valid frames before and after the missing frame are extracted, and the SNR estimate of the missing frame can be obtained through linear interpolation; finally, the missing frame number and the SNR estimate are substituted into the fitting function. This allows us to obtain the confidence feature values ​​for missing frames.

[0110] In some embodiments, for the aforementioned continuous or non-continuous missing values, the following verification can be performed after the filling is completed: Deviation verification: Calculate the relative deviation between the filled value and the feature values ​​of adjacent valid frames. If the relative deviation is less than or equal to a preset deviation threshold, the filling is determined to be valid; if the relative deviation is greater than the threshold, the minimum value of the feature values ​​of adjacent valid frames is taken as the filling value; Temporal smoothness verification: Input the filled temporal sequence into the smoothness detection module, calculate the difference between the feature values ​​of adjacent frames, and if all differences are less than or equal to a preset jump threshold, output to step S340; if there is a difference greater than the preset jump threshold, re-execute the filling algorithm and adjust the training parameters.

[0111] Step S340: Smooth the confidence time series after data padding, and identify the channel state risk type by combining abrupt change detection and trend prediction. The risk types include random noise, persistent interference, and sudden interference. Specifically, the confidence time series output in step S330 is defined as: And input the multi-frame trend analysis module.

[0112] In some embodiments, a moving average algorithm with a fixed window size is used to process the sequence. The smoothed sequence is calculated, and the smoothing process eliminates the eigenvalue fluctuations caused by random noise in a single frame, reflecting the short-term average state of the channel.

[0113] In some embodiments, the Cumulative Sum (CUSUM) algorithm is used to identify sudden disturbances by the deviation between the accumulated eigenvalues ​​and the moving average, and a cumulative deviation sequence is defined:

[0114]

[0115] in, This is the cumulative deviation value from the previous frame, with an initial value of 0; This represents the deviation between the current frame's feature value and the moving average. The moving average of the current frame is obtained by averaging the built-in confidence feature values ​​of the sliding window; The preset deviation threshold can be set based on historical normal fluctuation deviation statistics to distinguish between normal fluctuations and sudden disturbances; if If the value exceeds the preset alarm threshold, then the first alarm is determined to be... A sudden change occurs in the frame, corresponding to a sudden channel interference.

[0116] In some embodiments, a small-scale recurrent neural network is used to fill the time sequence. Using the input, predict the confidence feature value for the next frame. After training with historical time-series data, if If the predicted values ​​show a downward trend for multiple consecutive frames, it is determined that the channel has a long-term risk of deterioration.

[0117] In some embodiments, based on the above trend analysis results, channel risks are divided into: random noise: single frame That is, the feature value of the current frame is lower than the moving average. And predicted value Interference is determined to be occasional noise caused by instantaneous channel fluctuations, which has little impact on command reliability; continuous interference: 3 or more consecutive frames. or moving average Continuous decline, and predicted value The interference is determined to be persistent, caused by factors such as slow channel fading, which will gradually reduce command reliability; sudden interference: a sudden change point is detected. If the value exceeds the alarm threshold and the current frame decreases by more than a preset percentage (e.g., 30%) compared to the previous frame, and there is no continuous decrease with a moving average, it is determined to be a momentary strong interference.

[0118] This embodiment first addresses the issue of time-series data breaks caused by satellite-to-ground link interruptions and transient interference by using a sliding window and categorized missing data filling, ensuring that the sequence of input trend analysis is complete and closely matches the actual channel conditions. Secondly, through multi-dimensional analysis using moving averages, abrupt change detection, and trend prediction, it classifies risks into three categories: accidental noise, persistent interference, and sudden interference, overcoming the limitations of traditional single-frame static judgment and providing a dual basis for subsequent reliability decisions in terms of time dimension and risk type. Simultaneously, the entire process employs lightweight algorithms, such as linear interpolation with SNR correction and small-scale RNNs, to adapt to the spacecraft's onboard computing power and storage resource constraints, improving the sensitivity of attitude control decisions to the time-varying characteristics of the channel while ensuring real-time performance.

[0119] Step S400: Determine the adaptive dynamic decision threshold based on the channel state risk type, and make a confidence decision on the current attitude control command data frame in combination with the confidence feature value;

[0120] In some embodiments, determining an adaptive dynamic decision threshold based on the channel state risk type includes: acquiring the signal-to-noise ratio (SNR) of the link between the target spacecraft and the ground station in real time, and calculating a basic threshold value based on the Shannon channel capacity; adjusting the basic threshold value according to the channel state risk type to obtain the adaptive dynamic decision threshold; firstly, extracting the SNR of the link between the target spacecraft and the ground station synchronously estimated during demodulation in step S100, which can be calculated by the ratio of signal power to noise power, i.e. ,in To demodulate the baseband signal power, The noise power during periods without signal; according to Shannon's theorem, the channel capacity of a binary input AWGN channel (a typical model of aerospace-to-ground link). The unit is bit / s / Hz; where: It is a logarithmic function to the base 2, reflecting the maximum information carrying capacity of the channel; coefficients Derived from BPSK modulation; SNR is the real-time channel signal-to-noise ratio, the worse the channel quality, the smaller the SNR. The closer the threshold is to 0, the weaker the channel's ability to transmit information. Based on the correlation between channel capacity and command reliability, a basic threshold function can be designed as follows: And configured so that the larger the channel capacity, the higher the basic threshold. The higher the threshold, the stricter the requirements for instruction reliability; conversely, the lower the threshold, the less likely normal instructions will be misjudged as abnormal.

[0121] In some embodiments, the basic threshold value can also be queried based on a preset code rate-signal-noise ratio-threshold mapping table. That is, through extensive aerospace channel simulations, the LDPC decoding performance at different code rates and SNRs is simulated to establish a three-dimensional mapping table. The parameters in the table include the code rate, signal-to-noise ratio (SNR), and basic threshold as resolved in step S100. The mapping rule is configured as follows: for the same SNR, the lower the code rate, the lower the basic threshold; for the same code rate, the higher the SNR, the higher the basic threshold. The basic threshold is queried by extracting the code rate and real-time SNR of the current frame and directly querying the corresponding basic threshold from the mapping table. This method eliminates the need for real-time calculation, reducing on-board computing power consumption.

[0122] In some embodiments, the basic threshold is adjusted based on the channel state risk type identified in step S340. By making adaptation adjustments, an adaptive dynamic decision threshold is obtained. If the risk type is continuous interference, the threshold will be increased by 10% of the preset percentage to avoid erroneous command execution due to continuously deteriorating channel conditions; if the risk type is occasional noise, the threshold will be decreased by 5% of the preset percentage to avoid excessive interception of normal commands; if the risk type is sudden interference, the basic threshold will remain unchanged to prioritize the security of command decision-making.

[0123] Step S400 specifically includes:

[0124] Step S410: If the confidence feature value of the current attitude control command data frame is not lower than the adaptive dynamic decision threshold and the channel state risk type is random noise, it is determined to be a reliable command; firstly, extract the confidence feature value of the current attitude control command data frame calculated in step S200. (Including LLR minimum, arithmetic mean, or information entropy), and adaptive dynamic decision thresholds. In comparison, if If the reliability quantification criteria are met, the risk identification results from step S340 are invoked to confirm that the current channel risk type is random noise. The criteria for this determination are: the moving average sequence does not show a continuous decrease, the cumulative deviation sequence is less than or equal to the alarm threshold, and the RNN trend prediction value... If both of the above conditions are met, the current attitude control command data frame is determined to be a trusted command, marked as allowed, and the command is temporarily stored in the decision output buffer, waiting to proceed to step S500 for execution.

[0125] Step S420: If the confidence feature value of the current attitude control command data frame is not lower than the adaptive dynamic decision threshold, and the channel quality is judged to be deteriorating by trend prediction, it is determined to be a conditionally reliable command; similar to step S410, first confirm that the confidence feature value of the current frame is greater than or equal to the adaptive dynamic decision threshold. If the current instruction itself meets the reliability standard, then the trend analysis results from step S340 are invoked to confirm that the channel exhibits a deterioration trend but has not reached the sustained interference standard. The determination is based on the RNN prediction value. Furthermore, the predicted values ​​for two consecutive frames show a downward trend, and the moving average sequence However, if the continuous interference standard of three consecutive frame decreases is met and the channel risk type is still random noise, but the short-term trend shows a decline in quality; if both of the above conditions are met at the same time, the current attitude control command data frame is determined to be a conditionally reliable command, and the command is temporarily stored in the decision output buffer for execution; at this time, channel degradation warning information can also be generated and sent to the code rate adjustment module and the ground station.

[0126] Step S430: If the confidence feature value of the current attitude control command data frame is lower than the adaptive dynamic decision threshold, or the channel state risk type is abnormal, it is determined to be an abnormal command. Specifically, the core of step S430 is to determine that the reliability of the current command is insufficient or the channel is in a severe abnormal state, and it needs to be intercepted to avoid erroneous execution. The determination is made if any of the following conditions are met: The confidence feature value is below the threshold, specifically: the current frame confidence feature value... Less than the adaptive dynamic decision threshold At this point, the reliability of the instruction itself does not meet the standards for safe execution;

[0127] The abnormal channel state risk type includes continuous interference, i.e., the result of step S340 is continuous interference; it also includes sudden interference, i.e., the result of step S340 is sudden interference. If either of the above conditions is met, the current attitude control command data frame is determined to be an abnormal command and marked as intercepted: the transmission of the command to the attitude control system is terminated to avoid erroneous execution.

[0128] In summary, step S400 calculates the basic threshold using Shannon channel capacity theory or a pre-set mapping table in dual modes, and dynamically adjusts it based on the channel risk type. This solves the problem of misjudgment of fixed thresholds in complex time-varying channels, ensuring that the threshold is both compatible with the LDPC code rate and the channel SNR, and can flexibly change with the risk status. On the other hand, through a multi-dimensional judgment logic based on confidence feature value, channel trend, and risk type, it distinguishes between reliable, conditionally reliable, and abnormal commands. This ensures the efficient execution of high-reliability commands, provides an early warning mechanism for conditionally reliable commands to address channel degradation in advance, and avoids security risks by intercepting abnormal commands.

[0129] Step S500: Execute the corresponding target spacecraft attitude control strategy based on the confidence level decision result;

[0130] Step S500 specifically includes:

[0131] Step S510: If it is a trusted instruction, the attitude control instruction data frame is sent to the attitude control system for execution. In this embodiment, to ensure that trusted instructions are transmitted to the attitude control system for execution efficiently and without risk, the specific process is based on the determination result of step S410:

[0132] First, the attitude control command data frames filtered in step S100 are extracted, including attitude adjustment parameters such as thruster magnitude, momentum wheel target speed, and attitude angle correction values. The command format is then converted according to the interface protocol of the attitude control system. For example, the angle deviation correction values ​​in the commands are converted from binary symbols to decimal values ​​that the attitude control system can directly parse, ensuring that the command parameters conform to the data parsing specifications of the attitude control system.

[0133] Then, using the LLR array (log-likelihood ratio of posterior probability per bit) extracted in step S200, key bit verification is performed on the instruction frame after format conversion: the bits related to the core parameters of attitude control in the instruction are filtered out, and their absolute LLR values ​​are verified to be ≥ the preset safety threshold to avoid accidental errors introduced by format conversion.

[0134] Finally, the verified instructions are sent to the attitude control system via the onboard bus. After receiving the instructions, the attitude control system drives actuators such as thrusters, momentum wheels, and solar panel drive motors to perform attitude adjustment actions according to the instructions. At the same time, it feeds back the execution status to the onboard main control unit in real time, forming a full-process traceability record of instruction execution.

[0135] Step S520: If it is a conditionally reliable instruction, send an instruction to the ground station while executing the current instruction to trigger the adjustment of the LDPC code rate for the next frame attitude control instruction. The goal of this embodiment is to execute the current instruction and optimize the reliability of subsequent instructions in advance. Similar to step S510, after performing format adaptation and key bit verification on the current attitude control instruction data frame, it is sent to the attitude control system for execution via the on-board bus to ensure that the current attitude adjustment task is not affected by the channel degradation trend. Then, based on the channel degradation trend information output in step S340, such as the RNN predicting a decrease in the confidence feature value of the next frame and a decrease in the moving average sequence, the following operation is performed: extract the current frame LDPC code rate parsed in step S100, denoted as... Query the preset bitrate-error correction capability mapping table and select the one with stronger error correction capability. The target bitrate is used as the target bitrate, and the mapping rule follows that the lower the bitrate, the stronger the error correction capability. For example, a bitrate of 7 / 8 corresponds to a switch to a bitrate of 4 / 5, a bitrate of 4 / 5 corresponds to a switch to a bitrate of 2 / 3, and a bitrate of 2 / 3 corresponds to a switch to a bitrate of 1 / 2. Based on the target bitrate, the corresponding LDPC parity-check matrix parameters are retrieved from the bitrate-submatrix parameter mapping table and pre-written into the parameter register of the LDPC decoder. When the next attitude control command data frame is received, the configuration parameters are directly called to complete the decoding adaptation, avoiding parameter switching delay. Finally, a bitrate switching notification frame is generated, containing the current bitrate, the target bitrate, and the reason for switching, and sent to the ground station via the downlink data transmission link. After receiving the notification, the ground station completes the command encoding and modulation according to the target bitrate before sending the next attitude control command frame, realizing satellite-ground bitrate coordinated switching and ensuring the reliability of subsequent command transmission.

[0136] Step S530: If it is an abnormal command, intercept the attitude control command data frame and send a retransmission request to the ground station. The request contains the location information of the low-confidence bits in the frame. After receiving the retransmission data returned by the ground station, replace the corresponding bits in the original command frame and perform local iterative decoding verification. If the confidence feature value is still lower than the adaptive dynamic decision threshold after verification, trigger the execution of the onboard preset conservative safety command. Specifically, for scenarios where the command is unreliable or the channel is in a bad state as determined in step S430, the specific process is as follows: Command interception: Terminate the current attitude control command data frame from being sent to the attitude control center. The transmission of the control system is marked as to be retransmitted and temporarily stored in the on-board temporary buffer area, which is independent of the historical confidence sequence buffer unit; low confidence bit position information encoding: the vulnerable bit position map generated in step S200 is called. The vulnerable bit position map is a binary mask with the same bit length as the instruction frame, representing that the bit at the position is a low confidence bit; in this embodiment, in order to compress the amount of transmitted data, run-length encoding is performed on the FBM: the length of consecutive "0" or "1" is represented by binary, the purpose of which is to reduce the amount of encoded data to 10%-20% of the original FBM, adapting to the bandwidth constraints of the satellite-to-ground link;

[0137] Retransmission request frame: includes frame identifier, original frame sequence number, i.e. frame sequence number in step S310, used by the ground station to locate the command frame to be retransmitted and the encoded FBM; embedding the retransmission request frame into the remote control feedback field of the downlink data transmission frame and sending it to the ground station through the 70MHz intermediate frequency link can effectively avoid occupying link resources alone.

[0138] In some embodiments, after receiving a retransmission request frame, the ground station parses the original frame sequence number and the encoded FBM to recover the complete FBM; it extracts the original attitude control command data frame under that sequence number, and only re-executes LDPC encoding on the low-confidence bits marked "1" in the FBM to generate a fragile bit retransmission subframe; then the spacecraft receives the retransmission subframe sent by the ground station, matches the temporarily stored command frame to be retransmitted with the frame sequence number; according to the bit position identifier in the retransmission subframe, it replaces the low-confidence bits at the corresponding positions in the original frame with the new bit values ​​of the retransmission subframe to form a repaired command frame.

[0139] In some embodiments, utilizing the partially parallel architecture of the LDPC decoder, local iterative decoding is performed only on the replaced bits and associated nodes in the repaired instruction frame: Associated node location: For each replaced low-confidence bit, based on the variable node and check node association set defined in step S200, all check nodes associated with it in the check matrix are found; for these variable nodes and check nodes, 3-5 iterative decodings are performed to update the LLR values ​​of the associated nodes; the iterative process also uses the normalized minimum sum algorithm of step S200 to ensure decoding accuracy. Based on the updated LLR array, the confidence feature values ​​of the repaired instruction frame are recalculated. ;Will With the adaptive dynamic decision threshold of step S400 Comparison: If If the repair is successful, the repaired command frame will be sent to the attitude control system for execution according to step S510. The repair was deemed unsuccessful, triggering an onboard safety precaution command.

[0140] In some embodiments, the conservative safety command execution process involves calling a safety command library stored in the onboard radiation-hardened Flash memory. Based on the attitude state fed back by the current attitude sensor, an appropriate conservative command is selected, including but not limited to: if the attitude sensor detects a small attitude deviation, such as a pitch angle deviation <0.5°, a momentum wheel zeroing command is executed: controlling the momentum wheel speed to reduce to 0 to prevent the deviation from being exacerbated by the momentum wheel's inertia; if an attitude roll trend is detected, a solar panel alignment command is executed, adjusting the solar panel angle via the solar panel drive motor, and using the torque generated by solar radiation pressure to stabilize the attitude. After the safety command is executed, a safety command activation notification is sent to the ground station, including the current attitude state and the reason for activation, awaiting subsequent intervention commands from the ground station.

[0141] This embodiment ensures the real-time performance and accuracy of attitude adjustment through the release of trusted commands and verification of key bits. The code rate switching of conditionally trusted commands proactively mitigates the risk of channel degradation, and the retransmission of abnormal commands effectively reduces the consumption of space-to-ground link resources, making it suitable for resource-constrained scenarios such as small satellites and deep space probes. Furthermore, local iterative decoding reduces onboard computing power consumption (reducing the number of iterations by 50%), and conservative safety commands provide attitude safety for extreme channel scenarios. Finally, the confidence update data generated throughout the process can be fed back to the timing analysis stage in step S300 to continuously optimize the accuracy of subsequent decisions. This step transforms the quantitative decisions based on LDPC confidence into actually executable attitude control actions, effectively avoiding attitude anomalies caused by erroneous commands, thereby improving the safety, mission reliability, and resource utilization efficiency of spacecraft on-orbit attitude control.

[0142] In summary, the spacecraft attitude anomaly control decision-making method based on LDPC confidence provided by this invention transforms the soft information generated during LDPC decoding in the spacecraft telemetry and control communication link into real-time, quantitative decision-making criteria, thereby constructing a secure decision-making closed loop that precedes traditional CRC verification. First, by extracting the confidence feature values ​​of the command data frames from the LDPC decoder and constructing a confidence time-series sequence with historical data for analysis, quantitative evaluation of the credibility of uplink commands is achieved. This allows for capturing the changing trends of command credibility from multiple frame dimensions, avoiding the one-sidedness of single-frame data judgment. Then, the confidence time-series sequence is combined with an adaptive dynamic decision threshold for credibility decision-making. This decision-making process can both identify potential risks based on time-series trends and dynamically adjust the judgment criteria according to channel conditions, reducing the probability of erroneous commands passing the judgment under harsh channel conditions such as low signal-to-noise ratio. It can effectively identify potential erroneous commands that traditional CRC verification cannot detect before command delivery and execution, thereby improving the reliability and security of spacecraft attitude control command execution. This method transforms the reliability information of the physical layer of the communication link into control decision-making criteria, providing technical assurance for the safe operation of spacecraft in harsh channel environments.

[0143] Please see Figure 2 This diagram illustrates the structure of a spacecraft attitude control decision system based on LDPC confidence, according to an embodiment of the present invention. The system includes:

[0144] The signal acquisition and processing module is configured to acquire and process the intermediate frequency signal of the target spacecraft and extract attitude control command data frames.

[0145] The LDPC decoding and feature extraction module is configured to perform LDPC iterative decoding on the attitude control command data frame, extract reliability measurement information during the decoding process, and calculate the confidence feature value of the current command data frame.

[0146] The channel state analysis module is configured to construct a confidence time series sequence from the confidence feature value of the current attitude control command data frame and the historical confidence feature value, and to analyze and process the confidence time series sequence to identify the channel state risk type.

[0147] The credibility decision module is configured to determine an adaptive dynamic decision threshold based on the channel state risk type, and combine the confidence feature value to make a credibility decision on the current attitude control command data frame.

[0148] The attitude control execution module is configured to execute the corresponding target spacecraft attitude control strategy based on the confidence decision results.

[0149] The third aspect of the present invention provides an electronic device, the electronic device comprising: a processor and a memory communicatively connected to the processor; wherein the memory stores instructions executable by the processor, the instructions being executed by the processor to enable the processor to perform the steps of the spacecraft attitude control decision method based on LDPC confidence as described in the first aspect of the present invention.

[0150] The fourth aspect of the present invention provides a computer-readable storage medium storing a program for implementing a spacecraft attitude control decision method based on LDPC confidence. The program for implementing the spacecraft attitude control decision method based on LDPC confidence is executed by a processor to implement the steps of the spacecraft attitude control decision method based on LDPC confidence as described in the first aspect of the present invention.

[0151] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0152] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for spacecraft attitude control decision making based on LDPC belief, characterized in that, The method comprises: Acquiring and processing the target spacecraft intermediate frequency signal, extracting the attitude control instruction data frame; LDPC iterative decoding is performed on the attitude control instruction data frame, and reliability measurement information in the decoding process is extracted and the confidence characteristic value of the current instruction data frame is calculated; The confidence characteristic value of the current attitude control instruction data frame and the historical confidence characteristic value are constructed into a confidence time sequence, and the confidence time sequence is analyzed and processed to identify the channel state risk type; An adaptive dynamic decision threshold is determined according to the channel state risk type, and the confidence characteristic value is combined to make a reliability decision on the current attitude control instruction data frame; According to the reliability decision result, the corresponding target spacecraft attitude control strategy is executed.

2. The LDPC confidence based spacecraft attitude control decision method of claim 1, wherein, Acquiring and processing the target spacecraft intermediate frequency signal, extracting the attitude control instruction data frame, comprises: Carrier synchronization and symbol synchronization processing is performed on the intermediate frequency signal to obtain a baseband information frame code stream through demodulation; The frame header field is parsed from the baseband information frame code stream, and the frame header field at least contains instruction type identification and LDPC code rate identification; According to the instruction type identification in the frame header field, the attitude control instruction data frame is screened out from the baseband information frame code stream, and the LDPC code rate corresponding to the attitude control instruction data frame is determined according to the code rate identification in the frame header field.

3. The LDPC belief-based spacecraft attitude control decision method of claim 1, wherein, LDPC iterative decoding is performed on the attitude control instruction data frame, and reliability measurement information in the decoding process is extracted and the confidence characteristic value of the current instruction data frame is calculated, comprising: The check matrix parameters and the number of iterations of the LDPC decoder are configured according to the code rate of the attitude control instruction data frame; Iterative decoding is performed on the attitude control instruction data frame by using the normalized minimum sum algorithm, and the log-likelihood ratio of the posterior probability of each data bit is extracted as the reliability measurement information after decoding is completed; The absolute value of the log-likelihood ratio of the posterior probability of all data bits in the frame is subjected to feature extraction operation, and the operation result is taken as the confidence characteristic value representing the overall reliability of the frame.

4. The LDPC belief-based spacecraft attitude control decision method of claim 1, wherein, The confidence characteristic value of the current attitude control instruction data frame and the historical confidence characteristic value are constructed into a confidence time sequence, and the confidence time sequence is analyzed and processed to identify the channel state risk type, comprising: The historical multiple-frame confidence characteristic values are stored by using a sliding window mechanism, and the confidence time sequence is constructed in time sequence; Missing data in the confidence time sequence is detected, and the missing data is divided into continuous missing data and non-continuous missing data according to the continuity of the missing data; For the continuous missing data, a linear interpolation algorithm with signal-to-noise ratio correction is used to fill in based on the effective confidence characteristic values before and after the missing section and the signal-to-noise ratio change trend; for the non-continuous missing data, a curve fitting algorithm is used to fill in based on the association relationship among the historical confidence characteristic values, the frame number and the channel signal-to-noise ratio; The confidence time sequence after data filling is smoothed, and the channel state risk type is identified by combining the mutation point detection and trend prediction, and the risk types include accidental noise, continuous interference and burst interference.

5. The LDPC belief-based spacecraft attitude control decision method according to any one of claims 1 to 4, characterized in that, An adaptive dynamic decision threshold is determined according to the channel state risk type, comprising: Real-time acquisition of the signal-to-noise ratio of a link between a target spacecraft and a ground station, and calculation of a basic threshold value based on Shannon channel capacity; Adjustment of the basic threshold value according to a channel state risk type to obtain an adaptive dynamic decision threshold.

6. The LDPC belief-based spacecraft attitude control decision method of claim 5, wherein, Conducting a credibility decision on a current attitude control instruction data frame in combination with a confidence characteristic value, including: If the confidence characteristic value of the current attitude control instruction data frame is not lower than the adaptive dynamic decision threshold and the channel state risk type is accidental noise, determining that the instruction is credible; If the confidence characteristic value of the current attitude control instruction data frame is not lower than the adaptive dynamic decision threshold, determining that the instruction is conditionally credible by judging a deterioration trend of channel quality through trend prediction; If the confidence characteristic value of the current attitude control instruction data frame is lower than the adaptive dynamic decision threshold or the channel state risk type is abnormal, determining that the instruction is abnormal.

7. The LDPC belief-based spacecraft attitude control decision method of claim 6, wherein, Executing a corresponding target spacecraft attitude control strategy according to a credibility decision result, including: If the instruction is credible, sending the attitude control instruction data frame to an attitude control system for execution; If the instruction is conditionally credible, sending an instruction to the ground station to trigger adjustment of an LDPC code rate of a next frame of attitude control instruction while executing the current instruction; If the instruction is abnormal, intercepting the attitude control instruction data frame and sending a retransmission request to the ground station, the request including position information of low-confidence bits in the frame; after receiving retransmission data returned by the ground station, replacing corresponding bits in the original instruction frame and performing local iterative decoding verification; if the confidence characteristic value after verification is still lower than the adaptive dynamic decision threshold, triggering execution of a preset conservative safety instruction on the satellite.

8. A spacecraft attitude control decision system based on LDPC belief, characterized in that, A system for executing the LDPC confidence-based spacecraft attitude control decision method of any one of claims 1 to 7, the system comprising: A signal acquisition and processing module configured to acquire and process intermediate frequency signals in a target spacecraft, and extract a frame of attitude control instruction data; An LDPC decoding and feature extraction module configured to perform LDPC iterative decoding on the frame of attitude control instruction data, extract reliability measurement information in the decoding process, and calculate a confidence characteristic value of the current instruction data frame; A channel state analysis module configured to construct a confidence timing sequence from the confidence characteristic value of the current frame of attitude control instruction data and historical confidence characteristic values, and analyze and process the confidence timing sequence to identify a channel state risk type; A credibility decision module configured to determine an adaptive dynamic decision threshold according to the channel state risk type, and conduct a credibility decision on the current frame of attitude control instruction data in combination with the confidence characteristic value; An attitude control execution module configured to execute a corresponding target spacecraft attitude control strategy according to a credibility decision result.

9. An electronic device, comprising: The electronic device comprises a processor and a memory connected in communication with the processor; the memory stores instructions executable by the processor, and the instructions are executed by the processor to enable the processor to execute the steps of the LDPC confidence-based spacecraft attitude control decision method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a program for implementing the LDPC confidence-based spacecraft attitude control decision method, and the program is executed by the processor to implement the steps of the LDPC confidence-based spacecraft attitude control decision method in any one of claims 1 to 7.

Images