Multi-band collaborative satellite communication remote data transmission system

Abstract

The invention relates to the technical field of satellite communication transmission optimization, and discloses a multi-band collaborative satellite communication remote data transmission system. According to the system, terminal historical data are processed through waveform decomposition, steady-state and transient characteristics are extracted, and a terminal-level dynamic transmission characteristic spectrum is constructed in combination with real-time parameters. And carrying out feature level comparison on the feature spectrum and ground station measured data, fusing time delay, frequency offset and coding distortion differences to generate a difference fusion spectrum, mapping the difference fusion spectrum to a satellite network topology, and identifying an abnormal propagation path and a corresponding physical region. And then activating multi-frequency-band cooperative monitoring, and completing accurate positioning and self-adaptive response of an abnormal region by adjusting satellite node parameters and applying controllable disturbance. According to the scheme, terminal-level behavior characteristics and network topology mapping are utilized, accurate diagnosis from signal tracing to a space path is achieved, and the accuracy of satellite network transmission anomaly management and the autonomous response capability of a system are improved.

Description

Technical Field

[0001] This invention relates to the field of satellite communication transmission optimization technology, specifically to a multi-band collaborative satellite communication remote data transmission system. Background Technology

[0002] Current satellite communication systems, especially when handling data transmission from remote and mobile terminals, generally employ transmission strategies based on link budget and universal channel models. These strategies primarily focus on overall signal quality parameters and use this as a basis for coarse-grained adjustments to power control or modulation and coding schemes. The system treats signals generated by different user terminals as homogeneous communication streams, lacking a fine-grained characterization of the unique signal transmission characteristics of individual terminals. When transmission link anomalies occur, such as increased latency or signal distortion, existing technologies can typically only determine the existence of the anomaly, but struggle to quickly pinpoint the root cause to specific terminal behavior patterns or a particular segment of the physical propagation path within the network. This results in delayed and insufficiently targeted response measures.

[0003] The shortcomings of existing technologies lie in their passivity and ambiguity. The system relies on ground stations for macroscopic quality assessment of received signals, making it impossible to correlate signal degradation with the specific historical characteristics of the terminal that generated the signal. This makes it difficult for the system to distinguish between signal characteristics changes caused by variations in the terminal's hardware state or software anomalies, and signal degradation caused by external environmental interference with the propagation path. Anomaly diagnosis remains at the symptom level, unable to trace back to the terminal-level cause or the path-level spatial location. Fault localization often requires manual intervention and lengthy investigations, failing to meet the real-time and autonomous requirements of highly dynamic satellite networks. Multi-band coordination mechanisms are typically only used for load balancing or capacity enhancement, without being combined with refined anomaly tracing and localization, resulting in a lack of precise decision-making basis for system resource adjustments. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-band coordinated satellite communication remote data transmission system to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a multi-band coordinated satellite communication remote data transmission system, the system comprising: Transmission feature spectrum generation stage: The waveform decomposition process is performed on the historical transmission data of the user terminal to extract the steady-state and transient features of the signal waveform. Combined with the real-time acquired Doppler frequency shift parameters and coding sequence structure, a terminal-level dynamic transmission feature spectrum is constructed. Difference fusion and positioning stage: The dynamic transmission feature spectrum is compared with the ground station measured data stream at the feature level, and the differences in three dimensions, namely time domain delay, frequency domain offset and coding distortion, are fused to generate a difference fusion spectrum. The difference fusion spectrum is then mapped to the satellite network topology to identify the abnormal signal propagation path and the corresponding physical area. Collaborative monitoring phase: Based on the identified physical area, the multi-band collaborative monitoring mechanism is activated, the transmission parameter configuration of the corresponding satellite node is adjusted, and controllable signal disturbances are applied to adjacent nodes to complete the accurate location of abnormal transmission areas and system response.

[0006] Preferably, the transmission feature spectrum generation stage specifically includes: The waveform decomposition process is performed on the historical transmission data of the user terminal. The waveform decomposition process includes using an adaptive filter bank to separate the steady-state components and transient components of the signal waveform, and calculating the energy ratio and zero-crossing point distribution of each component. Synchronously collect the Doppler frequency shift parameters in the current transmission link. The Doppler frequency shift parameters include the frequency offset value and its rate of change, and obtain its phase noise characteristics through a spectrum analyzer. Analyze the structure of the real-time transmission encoding sequence to obtain its frame length, synchronization header mode, and error correction coding redundancy; The processed steady-state features, transient features, Doppler frequency shift parameters, and coding sequence structure are input into the feature fusion network. The feature fusion network outputs a dynamic transmission feature spectrum that characterizes the transmission state of the terminal at a specific time through weighted splicing and dimensionality reduction operations.

[0007] Preferably, the operation of inputting the processed steady-state features, transient features, Doppler frequency shift parameters, and coding sequence structure into the feature fusion network includes: Time alignment processing is performed on steady-state and transient features to eliminate feature misalignment caused by asynchronous sampling times; The Doppler frequency shift parameter is normalized to make it have the same numerical dimensions as the signal characteristics; The encoded sequence structure is vectorized and embedded to transform it into a high-dimensional feature vector; The weight of each feature is dynamically calculated based on its impact on transmission quality in historical transmission data. All processed features are then weighted and concatenated, with the weights being dynamically calculated. Principal component analysis is performed on the spliced ​​high-dimensional feature vectors to reduce their dimensionality, resulting in a low-dimensional dynamic transmission feature spectrum that retains complete information.

[0008] Preferably, the difference fusion and positioning stage specifically includes: Receive dynamic transmission feature spectrum from the transmission feature spectrum generation stage; Receive the measured data stream reported in real time by the ground station. The measured data stream includes a signal strength sequence, a bit error rate sequence, and a transmission delay sequence. The dynamic transmission feature spectrum is compared point by point with the measured data stream at the feature level to calculate the time domain delay difference, frequency domain offset difference, and coding distortion difference. A Kalman filter is used to smooth the three differences and eliminate measurement noise. The smoothed differences are superimposed according to a preset fusion rule to generate a comprehensive difference fusion spectrum; The difference fusion spectrum is input into a preset satellite network topology model, which includes the location of all satellite nodes and the status information of inter-satellite links; The propagation of abnormal signals is simulated in the satellite network topology model by using a graph traversal algorithm. Links where the signal strength attenuation exceeds a threshold are marked, thereby determining the propagation path of abnormal signals and the corresponding physical area coordinates.

[0009] Preferably, the step of superimposing the smoothed differences according to a preset fusion rule to generate a comprehensive difference fusion spectrum is as follows: Assign basic weights to the time-domain delay differences; Assign basic weights to the frequency domain offset difference; Assign basic weights to the amount of coding distortion difference; The three basic weights are dynamically adjusted based on the current network load status. When the network load is high, the weight of the time domain delay difference is increased, and when the network load is low, the weight of the frequency domain offset difference is increased. The three weighted difference quantities are linearly superimposed to output the difference fusion spectrum.

[0010] Preferably, the graph traversal algorithm adopts a breadth-first search strategy, starting from the initial abnormal node indicated in the difference fusion spectrum, traversing all its connected links, calculating the signal attenuation on each link, and recording the link sequence where the attenuation value exceeds a preset threshold. The link sequence is the abnormal signal propagation path.

[0011] Preferably, the collaborative monitoring phase specifically includes: Receive the abnormal signal propagation path and physical region coordinates output from the difference fusion and localization stage; Based on the coordinates of the physical region, a command is sent to the satellite node covering the physical region to activate its multi-band collaborative monitoring mechanism. The multi-band collaborative monitoring mechanism includes extending the working frequency band to multiple preset frequency points and increasing the signal sampling rate. Send control commands to adjacent nodes that have direct link connections with satellite nodes in the anomalous area, instructing them to transmit a set of controllable signal disturbances with specific waveforms and power; The changes in signal parameters of satellite nodes in the monitoring abnormal area before and after the application of the controllable signal disturbance include received signal strength indication and bit error rate.

[0012] Preferably, the process of instructing the transmission of a set of controllable signals with specific waveforms and power includes: Generate a set of swept frequency signals whose frequency varies linearly; Control adjacent nodes to transmit the frequency sweep signal at a preset transmission power; Record the launch start time, launch duration, and launch frequency sequence.

[0013] Preferably, the process of signal parameter changes of the satellite node monitoring the abnormal area before and after applying the controllable signal disturbance includes: During the controlled signal disturbance transmission, the received signal strength indication data of satellite nodes in the abnormal area are continuously collected at a high sampling rate to form a curve of intensity change over time; Synchronously collect the bit error rate data of satellite nodes in the abnormal area to form a bit error rate curve over time; Extract the intensity variation curve and bit error rate variation curve during the transmission period of the controllable signal disturbance; The extracted curve is compared with the baseline curve under undisturbed conditions by differential calculation to calculate the magnitude of the signal parameter change.

[0014] Preferably, the system further includes: The nature of the anomaly is determined based on the magnitude of the change in the signal parameters. If the received signal strength indicator drops significantly and the bit error rate rises significantly at the same time, it is determined to be an abnormal channel attenuation. If the received signal strength indicator does not change much but the bit error rate rises abnormally, it is determined to be an abnormal coding or interference. The judgment result is associated with and stored along with the corresponding abnormal signal propagation path and physical area coordinates, and the system anomaly knowledge base is updated.

[0015] Compared with the prior art, the beneficial effects of the present invention are: By decomposing the waveforms of historical transmission data from user terminals and extracting their steady-state and transient features, and combining real-time Doppler frequency shift and coding structure to construct a dynamic transmission feature spectrum, deep modeling and digital representation of terminal signal characteristics are achieved. This feature spectrum, as a unique fingerprint of the terminal, provides a high-precision comparison benchmark for signal anomaly detection. This enables the system to perceive subtle deviations in signal waveform characteristics at the microscopic level, thereby identifying signal distortions caused by abnormal terminal conditions or potential interference before macroscopic performance indicators deteriorate significantly, thus elevating the granularity of fault diagnosis from the link level to the terminal level.

[0016] By comparing dynamic feature spectra with ground station measured data streams at the feature level, and fusing differences in the time, frequency, and coding domains to generate a difference fusion spectrum, this spectrum is mapped onto the satellite network topology, enabling spatial localization of anomalies. This method transforms abstract communication quality differences into a visual map associated with the network's physical structure, clearly indicating the abnormal propagation paths leading to signal degradation and their corresponding geographical areas. Based on this localization result, multi-band collaborative monitoring is activated. By adjusting the parameters of satellite nodes in suspicious areas and applying controllable signal perturbations, the system shifts from passive observation to active detection. This proactive interactive verification mechanism significantly reduces the probability of false positives, accurately locates anomaly areas, and makes subsequent system responses more targeted and efficient, improving the self-healing capability and transmission reliability of the entire network. Attached Figure Description

[0017] Figure 1 This is a schematic diagram illustrating the working principle of the multi-band coordinated satellite communication remote data transmission system described in this invention. Figure 2 A flowchart for the transmission feature spectrum generation stage; Figure 3 A flowchart for input processing of the feature fusion network; Figure 4 A comparison chart of key performance indicators of satellite nodes during the collaborative monitoring phase; Figure 5 A trade-off curve between information retention rate and processing speed improvement. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1This invention provides a multi-band collaborative satellite communication remote data transmission system. The system includes: in the transmission feature spectrum generation stage, waveform decomposition processing is performed on historical transmission data of user terminals to extract steady-state and transient features of the signal waveform. Combined with real-time acquired Doppler frequency shift parameters and coding sequence structure, a terminal-level dynamic transmission feature spectrum is constructed; in the difference fusion and positioning stage, the dynamic transmission feature spectrum is compared with the ground station measured data stream at the feature level, and the differences in three dimensions—time domain delay, frequency domain offset, and coding distortion—are fused to generate a difference fusion spectrum. The difference fusion spectrum is then mapped to the satellite network topology to identify abnormal signal propagation paths and corresponding physical areas; in the collaborative monitoring stage, a multi-band collaborative monitoring mechanism is activated based on the identified physical areas, the transmission parameter configuration of the corresponding satellite nodes is adjusted, and controllable signal disturbances are applied to adjacent nodes to complete the accurate positioning and system response of abnormal transmission areas.

[0020] Example 1: See Figure 2 In a specific implementation, the process begins with waveform decomposition processing of historical transmission data from the user terminal. This waveform decomposition includes using an adaptive filter bank to separate the steady-state and transient components of the signal waveform, and calculating the energy ratio and zero-crossing distribution of the steady-state and transient components. Simultaneously, Doppler frequency shift parameters in the current transmission link are acquired, including frequency offset values ​​and their rate of change. The phase noise characteristics of the Doppler frequency shift parameters are obtained using a spectrum analyzer. The coded sequence structure of the real-time transmission is analyzed to obtain the frame length, synchronization header mode, and error correction coding redundancy. The processed steady-state features, transient features, Doppler frequency shift parameters, and coded sequence structure are input into a feature fusion network. The feature fusion network, through weighted concatenation and dimensionality reduction operations, outputs a dynamic transmission feature spectrum characterizing the terminal's transmission state at a specific moment. In practical implementation, waveform decomposition processing employs an adaptive filter bank to decompose the input signal. The adaptive filter bank dynamically adjusts the passband range based on the signal's spectral characteristics to separate steady-state and transient components. The energy proportion of the steady-state component is obtained by calculating its proportion in the total signal energy, and the zero-crossing distribution of the transient component is obtained by statistically analyzing the number of times the signal crosses zero points per unit time and its distribution pattern. In some embodiments, the acquisition of Doppler frequency shift parameters is achieved through a frequency tracking module integrated into the satellite modem. The frequency tracking module outputs the frequency offset value and its rate of change in real time, while an external spectrum analyzer performs quantitative analysis of the signal phase noise.

[0021] In specific implementation, the operation of inputting the processed steady-state features, transient features, Doppler frequency shift parameters, and coded sequence structure into the feature fusion network includes: performing time alignment processing on the steady-state and transient features, which uses an interpolation algorithm to adjust the feature timestamps to eliminate feature misalignment caused by asynchronous sampling times; normalizing the Doppler frequency shift parameters, which uses a minimum-maximum scaling method to map the Doppler frequency shift parameters to the [0,1] interval, making them have the same numerical dimensions as the signal features; and performing vectorized embedding representation on the coded sequence structure, which uses a lookup table to transform the discrete symbols of the coded sequence into high-dimensional feature vectors. In some embodiments, the weight of each feature is dynamically calculated based on the correlation coefficient between the feature and the transmission quality index in historical transmission data, and the weight calculation formula is expressed as: in: It is the weight of the i-th feature. It is the Pearson correlation coefficient between the i-th feature and the transmission quality index. This represents the total number of features. All processed features are weighted and concatenated, linearly combining the normalized features into a high-dimensional vector according to their weights. Principal component analysis (PCA) is then performed on the concatenated high-dimensional feature vector to reduce its dimensionality. PCA retains principal components whose cumulative variance contribution exceeds a preset threshold, resulting in a low-dimensional dynamic transmission feature spectrum with complete information retention. It's understood that time alignment relies on a high-precision clock synchronization signal to ensure the accuracy of feature alignment. Optionally, the dimension of the vectorized embedding representation is fixed at 128 dimensions to maintain processing consistency. It's understood that the PCA dimensionality reduction threshold is set to 95% to balance dimensionality compression and information preservation.

[0022] Example 2: See Figure 3In a specific implementation, the difference fusion and positioning stage specifically includes receiving a dynamic transmission feature spectrum from the transmission feature spectrum generation stage. The dynamic transmission feature spectrum is transmitted in vector form through the system's internal data bus. Simultaneously, it receives the measured data stream reported in real time by the ground station. The measured data stream includes a signal strength sequence, a bit error rate sequence, and a transmission delay sequence. The measured data stream is transmitted to the data processing center via the telemetry channel. The dynamic transmission feature spectrum and the measured data stream are compared point by point at the feature level to calculate the time-domain delay difference, frequency-domain offset difference, and coding distortion difference. A Kalman filter is used to smooth and filter the three differences to eliminate measurement noise. The smoothed differences are superimposed according to a preset fusion rule to generate a comprehensive difference fusion spectrum. The difference fusion spectrum is input into a preset satellite network topology model. The satellite network topology model contains the positions of all satellite nodes and the status information of inter-satellite links. A graph traversal algorithm is used to simulate the propagation of abnormal signals in the satellite network topology model, marking the links where the signal strength attenuation exceeds a threshold, thereby determining the propagation path of the abnormal signal and the corresponding physical area coordinates. In practice, the point-by-point comparison between the dynamic transmission feature spectrum and the measured data stream is performed in a dedicated feature comparison engine. The feature comparison engine first timestamps the feature vectors of the dynamic transmission feature spectrum and the data points of the measured data stream. The time-domain delay difference is obtained by calculating the absolute difference between the predicted transmission delay in the dynamic transmission feature spectrum and the corresponding values ​​of the transmission delay sequence in the measured data stream. The frequency-domain offset difference is obtained by calculating the difference between the signal center frequency represented in the dynamic transmission feature spectrum and the measured instantaneous frequency of the signal. The coding distortion difference is quantified by calculating the deviation between the bit error rate of the received data frame and the benchmark bit error rate given in the dynamic transmission feature spectrum based on historical data.

[0023] In practical implementation, the Kalman filter is designed independently for each of the three difference channels. Each Kalman filter is configured as a univariate system, with the state variable being the optimal estimate of the corresponding difference and the observed value being the original value of the difference calculated in real time. The process noise covariance and the observation noise covariance are initialized based on the historical fluctuation statistics of each difference. The smoothing filtering process is recursively performed through two steps: prediction and update. The final output is a smoothed difference sequence that has filtered out high-frequency measurement noise. In some embodiments, the process noise covariance matrix is ​​set as a scalar value that is dynamically adjusted according to link stability. In scenarios with drastic changes in channel conditions, the process noise covariance is appropriately increased to enhance the tracking capability of the Kalman filter.

[0024] In practice, preset fusion rules normalize and weight the smoothed differences to generate a difference fusion spectrum. The formula is expressed as: in: This represents the dimensionless difference fusion spectrum value. This represents the difference in time-domain delay after smoothing. This represents the historical mean of the time-domain delay difference. The historical standard deviation of the time-domain delay variation. The weighting coefficients represent the differences in time-domain delay. This represents the difference in frequency domain offset after smoothing. This represents the historical mean of the frequency domain offset difference. This represents the historical standard deviation of the frequency domain offset difference. The weighting coefficients represent the frequency domain offset differences. This represents the difference in encoding distortion after smoothing. This represents the historical mean of the amount of coding distortion difference. This represents the historical standard deviation of the coding distortion variation. These represent the weighting coefficients for the coding distortion variance. It can be understood that through z-score normalization, three variance quantities that originally had different dimensions are transformed into dimensionless standard scores, thus enabling linear superposition. In some embodiments, the historical mean and historical standard deviation are obtained from a long-term, maintained statistical database of variance quantities. , , The policy module dynamically allocates resources based on the current communication service type; for example, it assigns a higher weight to latency-sensitive services based on their time-domain latency differences. Optionally, the satellite network topology model is stored in a graph database as a directed graph. Vertex attributes include satellite identifiers and latitude / longitude coordinates, while edge attributes include link type, propagation loss model, and current state flag. The graph traversal algorithm employs a breadth-first search strategy, starting from an initial node with an abnormally high differential fusion spectrum value. It traverses all its connected links, calculating the signal attenuation on each link. The attenuation level is estimated by querying the link propagation loss model and combining it with the current transmit power and differential fusion spectrum value. The link sequence where the estimated attenuation value exceeds a preset threshold is recorded. This link sequence is the identified abnormal signal propagation path. The physical region coordinates are determined by calculating the geographic coordinates of the intersection center point of the coverage areas of all satellite nodes along this path.

[0025] Example 3: In a specific implementation, the process of superimposing the smoothed differences according to a preset fusion rule to generate a comprehensive difference fusion spectrum includes assigning basic weights to the time-domain delay difference, the frequency-domain offset difference, and the coding distortion difference; dynamically fine-tuning the three basic weights according to the current network load status, increasing the weight of the time-domain delay difference when the network load is high, and increasing the weight of the frequency-domain offset difference when the network load is low; and linearly superimposing the weighted three differences to output the difference fusion spectrum. In the specific implementation, the initial values ​​of the basic weights are determined based on historical data analysis: the initial value of the basic weight of the time-domain delay difference is set to 0.5, the initial value of the basic weight of the frequency-domain offset difference is set to 0.3, and the initial value of the basic weight of the coding distortion difference is set to 0.2. The dynamic fine-tuning operation is implemented through a weight function, the input of which is the real-time monitored network load parameter, defined as the ratio of the current number of active connections to the maximum supported number of connections. It can be understood that the weight of the time-domain delay difference is increased when the network load is high because the time uncertainty caused by the increased channel competition under high load is more significant. Optionally, the linear superposition operation is performed in an adder circuit, which receives the time-domain delay difference signal, frequency-domain offset difference signal, and coding distortion difference signal after weight scaling, and outputs their sum as the difference fusion spectrum.

[0026] In its implementation, the graph traversal algorithm employs a breadth-first search strategy. Starting from the initial anomalous node indicated in the difference fusion spectrum, it traverses all its connected links, calculates the signal attenuation on each link, and records the link sequence where the attenuation value exceeds a preset threshold. This link sequence represents the propagation path of the anomalous signal. The breadth-first search strategy uses a first-in-first-out (FIFO) queue data structure to manage the nodes to be visited. The search process begins with the initial anomalous node being enqueued. When the queue is not empty, the head node is removed, and all its unvisited adjacent nodes are visited. The attenuation of the signal along the links from the head node to each adjacent node is calculated. The signal attenuation on a single link is also considered. Calculated using the following model: in: This represents the link attenuation value from node i to node j. This represents the transmitted signal power at node i. This represents the received signal power at node j. This represents the baseline value of the free space path loss for the link between node i and node j. In some embodiments, the baseline value of the free space path loss is... The satellite network topology model is pre-calculated and stored based on satellite node location information and a known space propagation model. In specific implementations, the known space propagation model employs the free-space path loss principle. This model calculates the geometric distance of the signal propagation path based on satellite node location information, including satellite orbital parameters, real-time latitude and longitude coordinates, and relative position data of inter-satellite links. By analyzing this location information, the model calculates the path length of the signal from the transmitting node to the receiving node and, combined with commonly used frequency band parameters for satellite communication, such as the center frequency of L-band, C-band, or Ku-band, estimates the signal attenuation value under ideal free-space conditions. It is understood that determining whether the attenuation value exceeds a preset threshold is crucial for identifying abnormal links. Optionally, the preset threshold is dynamically set based on the theoretical signal-to-noise ratio tolerance of the link. When recording link sequences, links that meet the attenuation conditions are stored in a list in the order of access. When the breadth-first search is completed, this list contains the complete abnormal signal propagation path. In some embodiments, the physical area coordinates are determined by querying the geometric center geographic coordinates of the coverage areas of all satellite nodes along the abnormal signal propagation path.

[0027] Example 4: In a specific implementation, this stage begins by receiving the abnormal signal propagation path and physical area coordinates output from the difference fusion and positioning stage. Based on the physical area coordinates, instructions are sent to the satellite nodes covering the physical area to activate a multi-band collaborative monitoring mechanism. This mechanism includes extending the operating frequency band to multiple preset frequency points and increasing the signal sampling rate. Simultaneously, control instructions are sent to adjacent nodes with direct link connections to the abnormal area satellite nodes to instruct them to transmit a set of controllable signal disturbances with specific waveforms and power. Finally, changes in signal parameters of the abnormal area satellite nodes before and after the application of the controllable signal disturbances are monitored, including received signal strength indication and bit error rate. In this implementation, the abnormal signal propagation path and physical area coordinates are received through a message queue, which ensures the reliability and order of data transmission. The physical area coordinates are used to query the satellite network topology database to determine the identifiers of the satellite nodes covering the area. The activation instruction, containing configuration parameters for the multi-band collaborative monitoring mechanism, is encrypted and sent to the target satellite node via the satellite control channel. In some embodiments, the activation process of the multi-band collaborative monitoring mechanism involves reconfiguring the software-defined radio module on the satellite node. The software-defined radio module switches its operating frequency band from a single frequency point to multiple preset frequency points according to instructions. For example, the frequency point list includes specific center frequencies of the L-band, C-band, and Ku-band, while the signal sampling rate is increased from the default value of 1MHz to 5MHz to capture more detailed signal characteristics. Optionally, the selection of preset frequency points is based on a historical interference pattern database, prioritizing frequency bands with fewer recent interference events for monitoring.

[0028] In practice, the process of instructing adjacent nodes to transmit controllable signal disturbances includes generating a set of frequency sweep signals with linearly varying frequencies. The generation of these sweep signals is accomplished by a signal generator at the ground control center. The signal generator produces waveform data based on preset parameters and uploads it to adjacent nodes via the uplink. The adjacent nodes are then controlled to transmit the sweep signals at a preset transmit power. The transmit power is dynamically adjusted based on the link budget to avoid excessive interference with normal communication. The transmission start time, transmission duration, and transmission frequency sequence are recorded. The frequency variation of the sweep signal follows a linear formula: in: This represents the frequency value of the swept signal at position s in the sequence. Indicates the start frequency of the frequency sweep. Indicates the frequency at which the sweep ends. This represents the total number of points in the frequency sweep sequence. This indicates the current sequence index. In some embodiments, the total number of points S in the frequency sweep sequence is fixed at 1000 points, and the frequency sweep starting frequency is... Set to 1GHz, sweep termination frequency The frequency band is set to 2 GHz, the transmission duration is set to 100 milliseconds, and the transmission power is calculated based on the distance between adjacent nodes and satellite nodes in the anomaly area to ensure that the signal strength at the receiving end is within a detectable range. Optionally, the transmission frequency sequence is stored in the node's memory in tabular form for easy backtracking during subsequent analysis. Refer to Table 1, which shows the frequency configuration for multi-band collaborative monitoring.

[0029] Table 1: Frequency Point Configuration Table for Multi-Band Collaborative Monitoring Frequency Index Center frequency (MHz) Bandwidth (MHz) Sampling rate (MHz) 1 1500 10 5 2 4000 20 5 3 12000 40 5 The process of monitoring signal parameter changes of satellite nodes in an abnormal region before and after the application of a controllable signal disturbance includes continuously acquiring received signal strength indication data from the satellite nodes in the abnormal region at a high sampling rate during the controlled signal disturbance transmission period to form a strength-time curve, and simultaneously acquiring bit error rate data from the satellite nodes in the abnormal region to form a bit error rate-time curve. The strength and bit error rate curves during the controlled signal disturbance transmission period are extracted, and the extracted curves are differentially compared with the baseline curve under undisturbed conditions to calculate the magnitude of signal parameter changes. In specific implementation, signal parameter acquisition is achieved through a monitoring module on the satellite nodes in the abnormal region. The monitoring module samples the received signal strength indication value at a rate of 10,000 times per second, and the bit error rate is calculated every millisecond based on statistics from the most recent 1,000 symbols. The baseline curve is generated by averaging historical data within 10 seconds before the disturbance transmission. The differential comparison operation calculates the absolute value of the difference between the curves at corresponding time points and integrates to obtain the total magnitude of change.

[0030] See Figure 4In the satellite node performance evaluation during the collaborative monitoring phase, this figure quantifies the operational performance of nodes A through E using three core indicators: coverage success rate (%), response time (ms), and detection accuracy (%). Specifically, node A has the best coverage success rate (nearly 99%) and detection accuracy (approximately 97%), with the shortest response time (approximately 12ms). Node B maintains a high coverage success rate (82%) and detection accuracy (approximately 90%), with a slightly higher response time (19ms) than node A. The coverage success rates of nodes C, D, and E decrease sequentially (79%, 65%, 62%), while the detection accuracy shows a corresponding downward trend (approximately 87%, 82%, 79%), and the response times gradually increase (21ms, 25ms, 29ms). The correlation between these indicators reflects the performance trade-offs of the collaborative monitoring mechanism: advantages in coverage and detection accuracy usually correspond to better response efficiency, while node performance degradation may be related to resource load after multi-band expansion and parameter fluctuations under signal disturbances. This type of quantitative comparison provides data support for the optimization of satellite node parameters and the scheduling of nodes in abnormal areas.

[0031] Example 5: In a specific implementation, the process of monitoring the changes in signal parameters of satellite nodes in an abnormal area before and after the application of controllable signal disturbance includes continuously collecting received signal strength indication data of satellite nodes in the abnormal area at a high sampling rate during the controllable signal disturbance transmission period to form a strength-time variation curve, and simultaneously collecting bit error rate data of satellite nodes in the abnormal area to form a bit error rate-time variation curve. The strength variation curve and bit error rate variation curve during the controllable signal disturbance transmission period are extracted, and the extracted curves are differentially compared with the baseline curve under no disturbance conditions to calculate the change amplitude of signal parameters. In practice, high sampling rate continuous acquisition is achieved through analog-to-digital converters on satellite nodes in abnormal areas. The analog-to-digital converter samples the received signal at a rate of not less than 10 MHz. The received signal strength indication data is calculated and output in real time by the power detection circuit inside the satellite node. The bit error rate data is obtained by the forward error correction decoding module after decoding each data block by calculating the ratio of the number of error bits to the total number of bits. The extraction operation extracts the data segment within the corresponding time window from the continuous data stream based on the recorded start time and duration of the disturbance transmission. The baseline curve is obtained by calculating the moving average of the received signal strength indication data and the bit error rate data for a period of time before the disturbance transmission. The differential comparison operation is performed in the data processing unit, which subtracts the value of each sampling point on the baseline curve from the extracted curve and takes the absolute value.

[0032] In practice, the variation amplitude of signal parameters is calculated by integrating and normalizing the difference, and the calculation formula is as follows: in: Indicates the overall range of change. Indicates the start time of the disturbance. Indicates the end time of the disturbance. This represents the received signal strength indication value at time t during the disturbance period. This represents the baseline received signal strength indication value at time t. This represents the bit error rate at time t during the disturbance period. This represents the baseline bit error rate at time t. It is a weighting coefficient used to balance the relative weights of changes in received signal strength indication and changes in bit error rate. In some embodiments, the weighting coefficient... The value is adjusted based on the impact of changes in received signal strength indication and bit error rate on the accuracy of anomaly classification in historical data.

[0033] In practical implementation, the nature of the anomaly is determined based on the magnitude of the signal parameter changes. If the received signal strength indicator (RSS) drops significantly and the bit error rate (BER) rises significantly simultaneously, it is determined to be a channel attenuation anomaly. If the RSS is relatively stable but the BER rises abnormally, it is determined to be a coding or interference anomaly. This determination is performed by a classification logic module, which compares the integral terms of the relative changes in the RSS and BER with their respective preset thresholds. In some embodiments, a significant drop in the RSS is determined when the integral term of the relative changes in the RSS exceeds a threshold. The criterion for a synchronous and significant increase in the bit error rate is that the integral term of the relative change in the bit error rate exceeds a threshold. Furthermore, the two changes are highly correlated in time. The criterion for judging that the change in the received signal strength indication is not significant is that the integral term of the relative change in the received signal strength indication is lower than a threshold. The criterion for judging an abnormally high bit error rate is that the integral term of the relative change in bit error rate exceeds a threshold. Optionally, temporal correlation is quantified by calculating the Pearson correlation coefficient between the intensity change curve and the bit error rate change curve. The judgment result is associated with the corresponding abnormal signal propagation path and physical area coordinates and stored, updating the system anomaly knowledge base. The association storage operation packages the anomaly nature, the link sequence of the abnormal signal propagation path, the latitude and longitude of the physical area coordinates, and the change amplitude into a single record entry and writes it into the database table of the system anomaly knowledge base.

[0034] See Figure 5In the dynamic transmission feature spectrum generation process of the feature fusion network, the performance of PCA dimensionality reduction is reflected in the trade-off between information retention rate and processing speed improvement. Specifically, the reduced dimension is used as the independent variable, and the information retention rate (green curve) and processing speed improvement (orange curve) constitute a dual-indicator evaluation system: as the reduced dimension increases from 0 to 70, the information retention rate shows a rapid initial increase followed by a gradual plateauing (from approximately 78% to 99%), while the processing speed improvement shows a continuous decreasing trend (from approximately 99 times to approximately 71 times). This pattern corresponds to the dimensionality reduction process of high-dimensional vectors after feature weighted concatenation: while low-dimensional dimensionality reduction can significantly improve processing speed, it results in significant information loss; as the dimension increases, the information retention rate gradually approaches saturation, while the loss in processing speed gradually narrows. In terms of parameter configuration, the optimal balance between the reduced dimension and the dual indicators needs to be selected based on the performance requirements of the actual transmission scenario (such as high real-time performance or high information integrity).

[0035] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0036] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A multi-band coordinated satellite communication remote data transmission system, characterized in that, The operation process is divided into three stages: Transmission feature spectrum generation stage: The waveform decomposition process is performed on the historical transmission data of the user terminal to extract the steady-state and transient features of the signal waveform. Combined with the real-time acquired Doppler frequency shift parameters and coding sequence structure, a terminal-level dynamic transmission feature spectrum is constructed. Difference fusion and positioning stage: The dynamic transmission feature spectrum is compared with the ground station measured data stream at the feature level, and the differences in three dimensions, namely time domain delay, frequency domain offset and coding distortion, are fused to generate a difference fusion spectrum. The difference fusion spectrum is then mapped to the satellite network topology to identify the abnormal signal propagation path and the corresponding physical area. Collaborative monitoring phase: Based on the identified physical area, the multi-band collaborative monitoring mechanism is activated, the transmission parameter configuration of the corresponding satellite node is adjusted, and controllable signal disturbances are applied to adjacent nodes to complete the accurate location of abnormal transmission areas and system response.

2. The multi-band coordinated satellite communication remote data transmission system according to claim 1, characterized in that, The transmission feature spectrum generation stage specifically includes: The waveform decomposition process is performed on the historical transmission data of the user terminal. The waveform decomposition process includes using an adaptive filter bank to separate the steady-state components and transient components of the signal waveform, and calculating the energy ratio and zero-crossing point distribution of each component. Synchronously collect the Doppler frequency shift parameters in the current transmission link. The Doppler frequency shift parameters include the frequency offset value and its rate of change, and obtain its phase noise characteristics through a spectrum analyzer. Analyze the structure of the real-time transmission encoding sequence to obtain its frame length, synchronization header mode, and error correction coding redundancy; The processed steady-state features, transient features, Doppler frequency shift parameters, and coding sequence structure are input into the feature fusion network. The feature fusion network outputs a dynamic transmission feature spectrum that characterizes the transmission state of the terminal at a specific time through weighted splicing and dimensionality reduction operations.

3. The multi-band coordinated satellite communication remote data transmission system according to claim 2, characterized in that, The operation of inputting the processed steady-state features, transient features, Doppler frequency shift parameters, and coding sequence structure into the feature fusion network includes: Time alignment processing is performed on steady-state and transient features to eliminate feature misalignment caused by asynchronous sampling times; The Doppler frequency shift parameter is normalized to make it have the same numerical dimensions as the signal characteristics; The encoded sequence structure is vectorized and embedded to transform it into a high-dimensional feature vector; The weight of each feature is dynamically calculated based on its impact on transmission quality in historical transmission data. All processed features are then weighted and concatenated, with the weights being dynamically calculated. Principal component analysis is performed on the spliced ​​high-dimensional feature vectors to reduce their dimensionality, resulting in a low-dimensional dynamic transmission feature spectrum that retains complete information.

4. The multi-band coordinated satellite communication remote data transmission system according to claim 1, characterized in that, The difference fusion and positioning stage specifically includes: Receive dynamic transmission feature spectrum from the transmission feature spectrum generation stage; Receive the measured data stream reported in real time by the ground station. The measured data stream includes a signal strength sequence, a bit error rate sequence, and a transmission delay sequence. The dynamic transmission feature spectrum is compared point by point with the measured data stream at the feature level to calculate the time domain delay difference, frequency domain offset difference, and coding distortion difference. A Kalman filter is used to smooth the three differences and eliminate measurement noise. The smoothed differences are superimposed according to a preset fusion rule to generate a comprehensive difference fusion spectrum; The difference fusion spectrum is input into a preset satellite network topology model, which includes the location of all satellite nodes and the status information of inter-satellite links; The propagation of abnormal signals is simulated in the satellite network topology model by using a graph traversal algorithm. Links where the signal strength attenuation exceeds a threshold are marked, thereby determining the propagation path of abnormal signals and the corresponding physical area coordinates.

5. A multi-band coordinated satellite communication remote data transmission system according to claim 4, characterized in that, The smoothed differences are superimposed according to a preset fusion rule to generate a comprehensive difference fusion spectrum: Assign basic weights to the time-domain delay differences; Assign basic weights to the frequency domain offset difference; Assign basic weights to the amount of coding distortion difference; The three basic weights are dynamically adjusted based on the current network load status. When the network load is high, the weight of the time domain delay difference is increased, and when the network load is low, the weight of the frequency domain offset difference is increased. The three weighted difference quantities are linearly superimposed to output the difference fusion spectrum.

6. A multi-band coordinated satellite communication remote data transmission system according to claim 4, characterized in that, The graph traversal algorithm employs a breadth-first search strategy. Starting from the initial abnormal node indicated in the difference fusion spectrum, it traverses all its connected links, calculates the signal attenuation on each link, and records the link sequence where the attenuation value exceeds a preset threshold. The link sequence is the abnormal signal propagation path.

7. A multi-band coordinated satellite communication remote data transmission system according to claim 1, characterized in that, The collaborative monitoring phase specifically includes: Receive the abnormal signal propagation path and physical region coordinates output from the difference fusion and localization stage; Based on the coordinates of the physical region, a command is sent to the satellite node covering the physical region to activate its multi-band collaborative monitoring mechanism. The multi-band collaborative monitoring mechanism includes extending the working frequency band to multiple preset frequency points and increasing the signal sampling rate. Send control commands to adjacent nodes that have direct link connections with satellite nodes in the anomalous area, instructing them to transmit a set of controllable signal disturbances with specific waveforms and power; The changes in signal parameters of satellite nodes in the abnormal area before and after the application of the controllable signal disturbance are monitored, including the received signal strength indication and bit error rate.

8. A multi-band coordinated satellite communication remote data transmission system according to claim 7, characterized in that, The process of instructing the transmission of a controllable signal with a specific waveform and power includes: Generate a set of swept frequency signals whose frequency varies linearly; Control adjacent nodes to transmit the frequency sweep signal at a preset transmission power; Record the launch start time, launch duration, and launch frequency sequence.

9. A multi-band coordinated satellite communication remote data transmission system according to claim 8, characterized in that, The process of signal parameter changes of the satellite nodes monitoring the abnormal area before and after applying the controllable signal disturbance includes: During the controlled signal disturbance transmission, the received signal strength indication data of satellite nodes in the abnormal area are continuously collected at a high sampling rate to form a strength change curve over time; Synchronously collect the bit error rate data of satellite nodes in the abnormal area to form a bit error rate curve over time; Extract the intensity variation curve and bit error rate variation curve during the transmission period of the controllable signal disturbance; The extracted curve is compared with the baseline curve under undisturbed conditions by differential calculation to calculate the magnitude of the signal parameter change.

10. A multi-band coordinated satellite communication remote data transmission system according to claim 9, characterized in that, Also includes: The nature of the anomaly is determined based on the magnitude of the change in the signal parameters. If the received signal strength indicator drops significantly and the bit error rate rises significantly at the same time, it is determined to be an abnormal channel attenuation. If the received signal strength indication does not change significantly but the bit error rate increases abnormally, it is determined to be an encoding or interference anomaly. The judgment result is associated with and stored along with the corresponding abnormal signal propagation path and physical area coordinates, and the system anomaly knowledge base is updated.

Images