Satellite channel interference suppression and 4K signal enhancement system and method based on AI prediction

Abstract

The invention relates to the technical field of satellite communication, in particular to a satellite channel interference suppression and 4K signal enhancement system and method based on AI prediction, and the system comprises a transmitting end unit which is used for transmitting an uplink signal and carrying out the self-receiving monitoring; the receiving end unit is used for receiving a downlink signal, monitoring a signal quality parameter and generating an abnormal report when detecting that the signal quality is abnormal; the satellite operation center unit is in communication connection with the transmitting end unit and the receiving end unit and is used for receiving the abnormal report, collecting satellite state information and environment data and generating an interference coping strategy; and the AI analysis unit is in communication connection with the satellite operation center unit and is used for analyzing the abnormal report, the satellite state information and the environment data, identifying interference characteristics, predicting an interference trend and providing strategy suggestions for the satellite operation center unit, so that the coping capacity and efficiency of the satellite communication system to 5G interference are improved, and the service life of the satellite communication system is prolonged. And the interference response time is shortened from the conventional hour level to the minute level.

Description

Technical Field

[0001] The present invention relates to the field of satellite communication technology, and particularly to a satellite channel interference suppression and 4K signal enhancement system and method based on AI prediction. Background Art

[0002] With the rapid development and wide deployment of 5G technology, C-band satellite communication systems are facing increasingly serious frequency band interference problems, especially in the downlink frequency range of 3 - 4 GHz. These interferences not only affect the quality of conventional satellite communication, but also pose challenges to the growing demand for 4K high-definition video transmission. Traditional satellite communication systems often adopt fixed solutions when dealing with such interferences, such as simple power adjustment or frequency fine-tuning, lacking flexibility and foresight.

[0003] In the prior art, satellite communication systems usually adopt independent transmitter and receiver designs, lacking an effective cooperation mechanism. When the receiver encounters interference, it often needs to communicate with the transmitter and satellite operators manually. This method is inefficient, has a long response time, and cannot quickly respond to the dynamically changing interference environment. At the same time, existing systems lack the ability to analyze and predict long-term interference trends, making it difficult to formulate effective long-term solutions. In addition, the identification of interference in traditional systems mainly relies on expert experience, and it is insufficiently adaptable to new interference patterns.

[0004] Therefore, there is an urgent need for a system that can intelligently identify interference, predict interference trends, and cooperate with multiple resources to suppress it, in order to ensure the quality of satellite communication, especially to meet the transmission requirements of high-bandwidth services such as 4K high-definition video. Summary of the Invention

[0005] In view of the above problems, the present invention proposes a satellite channel interference suppression and 4K signal enhancement system and method based on AI prediction. By introducing artificial intelligence technology, this system realizes accurate prediction of channel status and intelligent interference suppression. At the same time, it adopts advanced beamforming technology and signal enhancement algorithms, significantly improving the quality and reliability of 4K video transmission.

[0006] To solve the above technical problems, the present invention adopts the following technical solutions:

[0007] A satellite channel interference suppression and 4K signal enhancement system based on AI prediction, comprising:

[0008] A transmitter unit for transmitting an uplink signal and performing self-receiving monitoring;

[0009] A receiver unit for receiving a downlink signal, monitoring signal quality parameters, and generating an anomaly report when signal quality anomalies are detected;

[0010] The satellite operation center unit, communicatively connected to the transmitting end unit and the receiving end unit, is configured to receive the anomaly report, collect satellite status information and environmental data, and generate interference countermeasures;

[0011] The AI analysis unit, communicatively connected to the satellite operation center unit, is configured to analyze the anomaly report, the satellite status information and the environmental data, identify interference characteristics, predict interference trends, and provide strategic suggestions to the satellite operation center unit.

[0012] Preferably, the transmitting end unit includes:

[0013] A signal transmitting module, configured to transmit satellite uplink signals;

[0014] A self-receiving monitoring module, configured to receive and analyze the feedback signals of the uplink signals, and generate self-receiving status data;

[0015] A parameter adjustment module, configured to adjust transmission parameters according to the adjustment instructions sent by the satellite operation center unit;

[0016] A status reporting module, configured to send the self-receiving status data to the satellite operation center unit.

[0017] Preferably, the receiving end unit includes:

[0018] A signal receiving module, configured to receive satellite downlink signals;

[0019] A multi-dimensional monitoring module, configured to monitor parameters such as the signal-to-noise ratio, bit error rate, carrier-to-noise ratio, phase jitter, and spectral purity of the downlink signals;

[0020] An anomaly detection module, configured to compare the parameters with a dynamic baseline to identify minor anomalies, moderate anomalies, and severe anomalies;

[0021] An image recognition module, configured to obtain parameters that cannot be directly collected by identifying the device interface;

[0022] An anomaly reporting module, configured to generate an anomaly report including the anomaly level, anomaly parameters, and anomaly duration.

[0023] Preferably, the AI analysis unit includes:

[0024] A data fusion module, configured to perform spatio-temporal alignment and reliability assessment on the anomaly report, the satellite status information, and the environmental data;

[0025] An interference identification module, configured to extract interference characteristics, match them with an interference pattern library, and identify interference sources;

[0026] A trend prediction module for predicting interference trends based on a time series analysis model;

[0027] A strategy generation module for generating short-term, medium-term, and long-term countermeasure strategy suggestions based on the interference source and the interference trend;

[0028] A self-learning module for continuously updating the interference pattern library and the prediction model according to the strategy execution effect.

[0029] Preferably, the satellite operation center unit includes:

[0030] A communication coordination module for establishing a secure communication channel with the transmitting end unit and the receiving end unit;

[0031] A data collection module for collecting environmental data such as satellite status information, meteorological data, and the distribution of ground 5G base stations;

[0032] A decision matrix module for constructing a multi-dimensional decision matrix based on interference severity, duration, and trend;

[0033] A strategy formulation module for formulating a final strategy by combining the strategy suggestions provided by the AI analysis unit and the resource cost-effectiveness improvement analysis;

[0034] An execution monitoring module for decomposing the final strategy into specific tasks, assigning them to the transmitting end unit and the receiving end unit, and monitoring the execution progress.

[0035] Preferably, the decision matrix module includes:

[0036] A three-dimensional evaluation unit for comprehensively evaluating based on interference severity, duration, and trend;

[0037] A cost-benefit unit for calculating the resource cost and effectiveness improvement ratio of different strategies;

[0038] A priority stratification unit for determining the protection priority according to the business importance;

[0039] A technical feasibility unit for evaluating the technical implementation difficulty of various solutions;

[0040] A long-term impact unit for analyzing the long-term impact and sustainability of strategies.

[0041] Preferably, the strategy generation module includes:

[0042] A short-term strategy unit for generating short-term strategies including dynamic adjustment of transmission power, adaptive switching of coding methods, fine-tuning of receiver parameters, real-time adjustment of beam shape, and enhancement of signal processing;

[0043] A medium-term strategy unit for generating medium-term strategies including intelligent selection of backup frequency bands, design of frequency hopping patterns, temporary change of modulation methods, dynamic allocation of bandwidth, and time slot reorganization;

[0044] A long-term strategy unit for generating long-term strategies including re-planning of satellite resources, optimization of ground station locations, upgrade of system architectures, application for spectrum resources, and replacement of equipment.

[0045] Preferably, the trend prediction module includes:

[0046] A time series processing unit for performing time series decomposition on interference data;

[0047] A multi-model analysis unit for analyzing by combining multiple time series analysis models such as ARIMA and LSTM;

[0048] A period recognition unit for recognizing daily, weekly, monthly, and seasonal cycle patterns in interference;

[0049] An external factor analysis unit for quantifying the impact of external factors such as meteorological changes and ground activities on interference;

[0050] A prediction interval calculation unit for providing a confidence interval for the prediction result.

[0051] Preferably, the execution monitoring module includes:

[0052] A task decomposition unit for decomposing the strategy into specific execution steps;

[0053] An execution coordination unit for setting the execution order and time nodes;

[0054] A real-time monitoring unit for collecting execution progress and effect data;

[0055] A dynamic adjustment unit for adjusting the execution plan based on real-time feedback;

[0056] An effect evaluation unit for quantitatively evaluating the execution effect of the strategy.

[0057] A method for suppressing satellite channel interference and enhancing 4K signals based on AI prediction, including the following steps:

[0058] The receiving end unit monitors the quality of the downlink signal and generates an anomaly report when detecting signal quality anomalies and sends it to the satellite operation center unit;

[0059] The satellite operation center unit collects the anomaly report and collects satellite status information and environmental data;

[0060] The AI analysis unit analyzes the anomaly report, the satellite status information, and the environmental data to identify interference characteristics, predict interference trends, and generate policy recommendations;

[0061] The satellite operation center unit formulates interference response strategies based on the policy recommendations, including short-term strategies, medium-term strategies, and long-term strategies;

[0062] The satellite operation center unit issues the interference response strategies to the transmitting end unit and / or the receiving end unit for execution;

[0063] The satellite operation center unit monitors the execution effect of the interference response strategies and feeds back the execution results to the AI analysis unit for model optimization.

[0064] The beneficial effects of the present invention include:

[0065] 1. It improves the ability and efficiency of the satellite communication system to cope with 5G interference, shortens the interference response time from the traditional hour level to the minute level, and significantly enhances the system stability;

[0066] 2. It realizes the collaborative work of the transmitting end, the receiving end, and the satellite operation center, and improves the interference suppression effect through the coordination and optimization of multiple resources;

[0067] 3. It realizes the prediction of interference trends through AI technology, enables the system to take proactive countermeasures, and minimizes the interference impact;

[0068] 4. It establishes a three-level strategy system of short-term, medium-term, and long-term, and provides an all-round solution from parameter adjustment to frequency band switching to architecture reorganization;

[0069] 5. The system has self-learning ability, can continuously accumulate experience to optimize the interference recognition and processing ability, and adapt to new interference patterns. Description of the Drawings

[0070] Figure 1 is the overall architecture diagram of the satellite channel interference suppression and 4K signal enhancement system based on AI prediction of the present invention;

[0071] Figure 2 is the structural block diagram of the transmitting end unit of the present invention;

[0072] Figure 3 is the structural block diagram of the receiving end unit of the present invention;

[0073] Figure 4 is the structural block diagram of the AI analysis unit of the present invention;

[0074] Figure 5 is the structural block diagram of the satellite operation center unit of the present invention;

[0075] Figure 6 is the structural block diagram of the decision matrix module of the present invention;

[0076] Figure 7 is the structural block diagram of the strategy generation module of the present invention;

[0077] Figure 8 is the structural block diagram of the trend prediction module of the present invention;

[0078] Figure 9 is the structural block diagram of the execution monitoring module of the present invention;

[0079] Figure 10 is the flowchart of the method for suppressing satellite channel interference and enhancing 4K signals based on AI prediction of the present invention. Detailed implementation manners

[0080] Referring to Figure 1 , the system for suppressing satellite channel interference and enhancing 4K signals based on AI prediction provided by the present invention includes: a transmitting end unit 1, a receiving end unit 2, a satellite operation center unit 3, and an AI analysis unit 4. Among them, the transmitting end unit 1 and the receiving end unit 2 are respectively communicatively connected to the satellite operation center unit 3, and the AI analysis unit 4 is communicatively connected to the satellite operation center unit 3.

[0081] The transmitting end unit 1 is arranged on a satellite vehicle or a portable station, and is used for transmitting an uplink signal and performing self-receiving monitoring. The receiving end unit 2 is arranged at a fixed earth station, and is used for receiving a downlink signal, monitoring signal quality parameters, and generating an exception report when detecting signal quality anomalies. The satellite operation center unit 3, as the center of the system, is communicatively connected to the transmitting end unit 1 and the receiving end unit 2, and is used for receiving the exception report, collecting satellite status information and environmental data, and generating an interference countermeasure strategy. The AI analysis unit 4 is communicatively connected to the satellite operation center unit 3, and is used for analyzing the exception report, satellite status information and environmental data, identifying interference characteristics, predicting interference trends, and providing strategy suggestions to the satellite operation center unit 3.

[0082] Referring to Figure 2 , the transmitting end unit 1 of the present invention includes: a signal transmitting module 11, a self-receiving monitoring module 12, a parameter adjustment module 13, and a status reporting module 14.

[0083] The signal transmission module 11 is used to transmit satellite uplink signals. Preferably, the signal transmission module 11 includes a power adjustment sub-module, which can dynamically adjust the transmission power according to instructions within the range of 5W to 120W. In practical applications, for example, when encountering minor interference, the power can be finely adjusted from the standard 60W to 65W, increasing the signal-to-noise ratio by about 0.5dB; when encountering severe interference, it can be increased to 80W or higher, increasing the signal-to-noise ratio by more than 2 - 3dB, but not exceeding the upper limit of the equipment's rated power to avoid equipment damage.

[0084] The self-receiving monitoring module 12 is used to receive and analyze the feedback signals of the uplink signals and generate self-receiving status data. The parameters collected by the self-receiving monitoring module 12 include key indicators such as signal-to-noise ratio, bit error rate, and received power. Preferably, this module samples once every 5 seconds to obtain the real-time status; and records the complete parameters once every 30 minutes for trend analysis. The self-receiving monitoring module 12 also sets multi-level alarm thresholds. For example, when the signal-to-noise ratio is lower than 10dB, a minor alarm is triggered; when it is lower than 7dB, a moderate alarm is triggered; when it is lower than 5dB, a severe alarm is triggered. These thresholds are set based on the practical experience of the satellite communication industry and are applicable to most C-band satellite communication systems.

[0085] The parameter adjustment module 13 is used to adjust the transmission parameters according to the adjustment instructions sent by the satellite operation center unit 3. The adjustable parameters include transmission power, modulation mode, coding rate, bandwidth, etc. For example, in an interference environment, the modulation mode can be reduced from 8PSK to QPSK to improve the anti-interference ability; or the coding rate can be adjusted from 3 / 4 to 1 / 2 to enhance the error correction ability. Although the data throughput will be reduced, the link stability can be significantly improved.

[0086] The status reporting module 14 is used to send the self-receiving status data to the satellite operation center unit 3. Preferably, the status reporting module 14 adopts an encrypted transmission mechanism to ensure data security, and at the same time sets a hierarchical reporting strategy: the normal status is reported once every 10 minutes; the minor abnormal status is reported once every 2 minutes; the severe abnormal status is reported in real time. This hierarchical reporting mechanism not only ensures the timely handling of abnormal situations but also avoids the waste of network resources.

[0087] Refer to Figure 3 , the receiving end unit 2 of the present invention includes: a signal receiving module 21, a multi-dimensional monitoring module 22, an anomaly detection module 23, an image recognition module 24, and an anomaly reporting module 25.

[0088] The signal receiving module 21 is used to receive satellite downlink signals. Preferably, the signal receiving module 21 is equipped with a low-noise amplifier (LNA) with a noise figure lower than 0.7dB to improve the ability to receive weak signals.

[0089] The multi-dimensional monitoring module 22 is used to monitor parameters such as the signal-to-noise ratio, bit error rate, carrier-to-noise ratio, phase jitter, and spectral purity of the downlink signal. Preferably, the multi-dimensional monitoring module 22 adopts a parallel processing architecture to simultaneously monitor multiple parameters, and the sampling rate can reach 100 Hz to ensure that short interference pulses can be captured. For 4K video transmission, the key monitoring indicators include the signal-to-noise ratio (should be higher than 9 dB), bit error rate (should be lower than 1×10^-7), jitter (should be less than 50 ns), etc. These thresholds are determined based on the requirements of 4K video transmission for signal quality.

[0090] The anomaly detection module 23 is used to compare the above parameters with the dynamic baseline to identify minor anomalies, moderate anomalies, and severe anomalies. The dynamic baseline is the normal parameter range generated based on historical data statistical analysis and will be dynamically adjusted according to time, season, and working environment. Preferably, the anomaly detection adopts a three-level threshold mechanism: the minor anomaly threshold is 10% deviation from the baseline, the moderate anomaly threshold is 20% deviation from the baseline, and the severe anomaly threshold is 30% deviation from the baseline. For example, if the baseline value of the signal-to-noise ratio in a certain period is 12 dB, then a value lower than 10.8 dB triggers a minor anomaly, a value lower than 9.6 dB triggers a moderate anomaly, and a value lower than 8.4 dB triggers a severe anomaly.

[0091] The image recognition module 24 is used to obtain parameters that cannot be directly collected through the recognition of the device interface. This module solves the compatibility problem of data collection for multi-brand devices. Preferably, the image recognition module 24 adopts convolutional neural network (CNN) technology to capture the device display screen and identify parameter values, with an identification accuracy of over 95%. For devices of different brands, the system has pre-established an interface template library, including the interface layouts of common devices such as satellite receivers and decoders, supports the recognition of the display screen of the spectrum analyzer, and can extract key parameters such as the center frequency, spectral distribution, and signal intensity.

[0092] The anomaly reporting module 25 is used to generate an anomaly report containing the anomaly level, anomaly parameters, and anomaly duration. Preferably, the anomaly report adopts a structured data format, including information such as the anomaly occurrence time, duration, anomaly parameter name, anomaly value, normal baseline value, deviation percentage, and anomaly level, which is convenient for the AI analysis unit 4 for subsequent processing. For severe anomalies, a complete parameter record of 60 seconds before and after the anomaly will also be automatically attached for in-depth analysis.

[0093] Refer to Figure 4 , the AI analysis unit 4 of the present invention includes: a data fusion module 41, an interference recognition module 42, a trend prediction module 43, a strategy generation module 44, and a self-learning module 45.

[0094] The data fusion module 41 is used to perform spatio-temporal alignment and reliability assessment on anomaly reports, satellite status information, and environmental data. Spatio-temporal alignment refers to unifying data from different sources with different timestamps and spatial references into the same spatio-temporal framework. Preferably, interpolation algorithms are used for spatio-temporal alignment to process asynchronous data and ensure the temporal consistency of the data. Reliability assessment is an evaluation of data quality, including integrity, consistency, and accuracy scoring. Preferably, a quantization scale of 0-100 is used for reliability scoring, and data with a score below 60 will be marked as low reliability and given a lower weight in subsequent analyses.

[0095] The interference identification module 42 is used to extract interference features, match them with the interference pattern library, and identify the interference source. Interference features include spectral features (such as center frequency, bandwidth, power spectral density), time-domain features (such as duration, periodicity, amplitude variation), modulation features, etc. Preferably, a combination of wavelet transform and spectral analysis is used for feature extraction, which can effectively distinguish different types of interference. The interference pattern library stores the feature fingerprints of common interference sources, such as 5G base station interference, radar interference, solar noise, etc. A voting mechanism combining multiple classification algorithms such as support vector machine (SVM), K-nearest neighbor (KNN), and random forest is used for interference source identification to improve the identification accuracy. Experiments show that the identification accuracy of this mechanism can reach over 85% in complex interference environments, far exceeding traditional single-algorithm methods.

[0096] The trend prediction module 43 is used to predict the interference trend based on the time series analysis model. The detailed structure and function of this module will be described in detail later in combination with Figure 8 for a detailed description.

[0097] The strategy generation module 44 is used to generate short-term, medium-term, and long-term countermeasure suggestions based on the interference source and interference trend. The detailed structure and function of this module will be described in detail later in combination with Figure 7 for a detailed description.

[0098] The self-learning module 45 is used to continuously update the interference pattern library and prediction model according to the strategy execution effect. Preferably, the self-learning module 45 adopts a reinforcement learning framework, uses the strategy execution effect as a reward signal, and continuously optimizes the decision-making model. At the same time, for newly discovered interference patterns, the system will automatically extract features and add them to the interference pattern library to continuously expand the knowledge base. Experiments show that after 6 months of learning, the types of interference that the system can handle increase by about 40% compared to the initial stage, and the processing efficiency for known interference types increases by about 25%.

[0099] Referring to Figure 5 this invention's satellite operation center unit 3 includes: a communication coordination module 31, a data collection module 32, a decision matrix module 33, a strategy formulation module 34, and an execution monitoring module 35.

[0100] The communication coordination module 31 is used to establish a secure communication channel with the transmitting unit 1 and the receiving unit 2. Preferably, the communication uses an encryption protocol of TLS 1.3 or above to ensure the security of data transmission; at the same time, a redundant design of the communication link is implemented, including a primary link and a backup link, which automatically switches to the backup link when the primary link fails, ensuring the reliability of system communication. The link switching time is less than 3 seconds, meeting the requirements of real-time control.

[0101] The data collection module 32 is used to collect environmental data such as satellite status information, meteorological data, and the distribution of ground 5G base stations. The satellite status information includes orbital parameters, transponder status, signal coverage, etc.; the meteorological data includes rainfall, humidity, wind speed, etc.; the 5G base station data includes location coordinates, operating frequency, transmission power, etc. Preferably, the data collection module 32 adopts a distributed crawler architecture to obtain information from multiple authoritative data sources, such as obtaining real-time meteorological data from the meteorological department, obtaining base station distribution information from the communication management department, and obtaining satellite status from satellite operators. The data update frequency is set according to the speed of data change: the satellite status is updated every 5 minutes, the meteorological data is updated every 30 minutes, and the base station distribution data is updated every 24 hours.

[0102] The decision matrix module 33 is used to construct a multi-dimensional decision matrix based on interference severity, duration, and trend. The detailed structure and function of this module will be described in detail later in combination with Figure 6 for a detailed description.

[0103] The strategy formulation module 34 is used to formulate the final strategy by combining the strategy suggestions provided by the AI analysis unit 4 and the resource cost-effectiveness improvement analysis. Preferably, the strategy formulation adopts a weighted decision-making mechanism, with the weight of the AI suggestion being 70% and the weight of the manual review being 30%. The AI weight can be gradually increased after the system matures. The resource cost-effectiveness improvement analysis is a quantitative evaluation of the input-output ratio of different strategies. By comparing the ratio of the input costs of different strategies (such as equipment costs, frequency resource consumption, and manpower input) to the expected effect improvement (such as signal-to-noise ratio improvement and bit error rate reduction), the optimal solution is selected. For example, when facing medium-level interference, the system may compare the cost-benefit ratios of two strategies: "increasing the transmission power by 20%" and "switching to the backup frequency band". If the former has a cost of 2 units and a benefit of 3 units (ratio 1.5), and the latter has a cost of 5 units and a benefit of 10 units (ratio 2.0), then the latter is preferred.

[0104] The execution monitoring module 35 is used to decompose the final strategy into specific tasks, allocate them to the transmitting unit 1 and the receiving unit 2, and monitor the execution progress. The detailed structure and function of this module will be described in detail later in combination with Figure 9 for a detailed description.

[0105] Refer to Figure 6, the decision matrix module 33 of the present invention includes: a three-dimensional evaluation unit 331, a cost-benefit unit 332, a priority stratification unit 333, a technical feasibility unit 334, and a long-term impact unit 335.

[0106] The three-dimensional evaluation unit 331 is used for comprehensive evaluation based on interference severity, duration, and trend. The interference severity is quantified based on the degree of deviation of the signal quality parameter from the baseline, and is divided into slight (levels 1-3), moderate (levels 4-6), and severe (levels 7-9); the duration is divided into short (less than 1 hour), continuous (1-24 hours), and long-term (more than 24 hours); the trend is divided into improvement, stability, and deterioration. Preferably, the three-dimensional evaluation adopts a weighted scoring mechanism. For 4K video services, the weight of interference severity is 0.5, the weight of duration is 0.3, and the weight of trend is 0.2, and a comprehensive score is obtained for subsequent decision-making. For example, in a certain interference evaluation: severity level 7 (weight 0.5), duration 2 hours (continuous, weight 0.3), trend deterioration (weight 0.2), then the comprehensive score is 7×0.5 + 2×0.3 + 3×0.2 = 4.9, which belongs to a high-priority situation that requires immediate intervention.

[0107] The cost-benefit unit 332 is used to calculate the ratio of resource cost to effect improvement for different strategies. Preferably, the cost calculation considers factors such as equipment occupancy, bandwidth consumption, and manpower input; the effect evaluation considers indicators such as SNR improvement, BER reduction, and service recovery time. The cost-benefit ratio adopts a relative quantification method, with the cost-benefit ratio of the baseline strategy being 1.0, and other strategies are relatively compared. Practice has proved that in most interference cases, strategies with a cost-benefit ratio above 1.5 are worthy of implementation, and strategies below 1.0 are usually not economical.

[0108] The priority stratification unit 333 is used to determine the guarantee priority according to the business importance. Different business types are set with different priorities. For example, real-time news live broadcast has the highest priority (level 10), ordinary on-demand content has a lower priority (level 5), and background data transmission has the lowest priority (level 1). Preferably, in case of resource conflict, the system automatically guarantees high-priority services and sacrifices the quality of low-priority services if necessary. For example, in case of bandwidth limitation, the resolution of on-demand content may be reduced to ensure the smoothness of live broadcast content.

[0109] The technical feasibility unit 334 is used to evaluate the technical implementation difficulty of various solutions. The evaluation factors include technical maturity, implementation complexity, success rate, etc. Preferably, the technical feasibility adopts a quantitative score of 1-10. A score above 7 indicates that the technology is mature and reliable, a score of 4-6 indicates that the technology is feasible but there are certain risks, and a score below 3 indicates that the technology is difficult to implement. For example, the feasibility score of simply adjusting the transmission power is 9 points, while the feasibility score of developing a new spectrum sharing algorithm may be only 3 points.

[0110] The long-term impact unit 335 is used to analyze the long-term impact and sustainability of the strategy. The long-term impact analysis considers the long-term effects of the strategy on aspects such as system stability, equipment lifespan, and spectrum resources. Preferably, through methods such as historical data simulation and trend extrapolation, the system states 3 months, 6 months, and 12 months after the implementation of the strategy are predicted. For example, although frequent adjustment of the transmission power is effective in the short term, long-term analysis shows that it may accelerate equipment aging and reduce the overall system lifespan by about 5%, so it needs to be carefully considered when choosing long-term strategies.

[0111] Referring to Figure 7 , the strategy generation module 44 of the present invention includes: a short-term strategy unit 441, a medium-term strategy unit 442, and a long-term strategy unit 443.

[0112] The short-term strategy unit 441 is used to generate short-term strategies including dynamic adjustment of transmission power, adaptive switching of coding modes, fine-tuning of receiver parameters, real-time adjustment of beam shape, and enhancement of signal processing. The short-term strategies are aimed at immediate interference problems and usually take effect within minutes. Preferably, the dynamic adjustment of transmission power adopts a progressive incremental scheme, with each adjustment amplitude not exceeding 10% to avoid overcompensation; the adaptive switching of coding modes automatically selects the optimal forward error correction (FEC) code rate according to the bit error rate. For example, when interference intensifies, it drops from 3 / 4 to 1 / 2 or lower; the fine-tuning of receiver parameters includes LNA gain adjustment, filter bandwidth optimization, etc.; the real-time adjustment of beam shape enhances the signal strength in the target area and suppresses the signal reception in the interference direction by adjusting the beam direction and width of the phased array antenna; the enhancement of signal processing includes digital signal processing technologies such as adaptive filtering and interference cancellation. These short-term strategies usually take effect within 5 - 15 minutes and are suitable for dealing with sudden interference.

[0113] The medium-term strategy unit 442 is used to generate medium-term strategies including intelligent selection of backup frequency bands, design of frequency hopping patterns, temporary change of modulation modes, dynamic allocation of bandwidth, and time slot reorganization. The medium-term strategies are aimed at interference lasting for several hours to several days and require relatively large system adjustments. Preferably, the intelligent selection of backup frequency bands is based on spectrum occupancy analysis to automatically select the backup frequency band with the least interference; the design of frequency hopping patterns designs the optimal frequency hopping sequence according to the interference characteristics to improve the anti-interference ability; the temporary change of modulation modes can be reduced from high-order modulation (such as 8PSK) to low-order modulation (such as QPSK) under interference conditions to improve the link robustness; the dynamic allocation of bandwidth intelligently allocates limited bandwidth resources according to service priorities; the time slot reorganization optimizes resource utilization by adjusting the time division multiplexing (TDM) time slot allocation. These medium-term strategies usually take 1 - 4 hours to complete deployment and are suitable for dealing with persistent interference.

[0114] The long-term strategy unit 443 is used to generate long-term strategies including satellite resource re-planning, ground station location optimization, system architecture upgrade, spectrum resource application, and equipment replacement. The long-term strategies target interference or interference trends expected to last for several weeks to several months and involve major adjustments at the system level. Preferably, satellite resource re-planning includes transponder reallocation, satellite handover, etc.; ground station location optimization reduces the impact of interference through geographical location adjustment; system architecture upgrade may involve introducing new technologies such as adaptive beamforming, cognitive radio, etc.; spectrum resource application includes applying for the right to use a new frequency band from regulatory agencies; equipment replacement involves hardware upgrades to improve the overall performance of the system. These long-term strategies usually take several weeks to several months to implement and are suitable for coping with long-term interference trends and system performance improvement requirements.

[0115] Referring to Figure 8 , the trend prediction module 43 of the present invention includes: a time series processing unit 431, a multi-model analysis unit 432, a period identification unit 433, an external factor analysis unit 434, and a prediction interval calculation unit 435.

[0116] The time series processing unit 431 is used to perform time series decomposition on interference data. Preferably, time series decomposition uses classical time series analysis methods to decompose interference data into a trend component, a seasonal component, and a random component. The trend component reflects the long-term change trend of interference, the seasonal component reflects the periodic change of interference, and the random component reflects unpredictable fluctuations. The mathematical expression of time series decomposition is:

[0117] Y(t) = T(t) + S(t) + R(t),

[0118] where SY(t)S is the original time series, ST(t)S is the trend component, SS(t)$ is the seasonal component, and SR(t) is the random component. Practice shows that this decomposition method can effectively identify the regular changes in interference and improve the prediction accuracy. The multi-model analysis unit 432 is used to perform analysis by combining multiple time series analysis models such as ARIMA and LSTM. ARIMA (Autoregressive Integrated Moving Average Model) is suitable for linear time series prediction, and its mathematical expression is:

[0119] φ(B)(1 - B) d X t = θ(B)ε t ,

[0120] where φ(B) is the autoregressive polynomial, θ(B) is the moving average polynomial, (1 - B) d is the differencing operator, X is the time series, ε tis white noise. LSTM (Long Short-Term Memory) is a type of recurrent neural network, which is particularly suitable for capturing long-term dependencies in time series data. Preferably, the system performs weighted fusion on the prediction results of different models, and the weights are dynamically adjusted according to the historical prediction accuracy. Experiments show that the prediction accuracy of multi-model fusion is 10-15% higher than that of a single model, especially in a complex interference environment, the advantage is more obvious.

[0121] The period recognition unit 433 is used to recognize daily, weekly, monthly, and seasonal cycle patterns in the interference. Preferably, the period recognition adopts a method combining Fourier transform and autocorrelation analysis, which can extract periodicities of multiple time scales from the interference data. For example, the system can recognize the difference in interference patterns between weekdays and weekends, or recognize seasonal interference changes related to local meteorological conditions. Once these cycle patterns are recognized, it will greatly improve the prediction accuracy, especially important for medium- and long-term predictions.

[0122] The external factor analysis unit 434 is used to quantify the impact of external factors such as meteorological changes and ground activities on the interference. Preferably, a multivariate regression analysis method is adopted to establish a quantitative relationship model between external factors and interference parameters. For example, through the analysis of historical data, it is found that for every 1 mm increase in rainfall, the attenuation of C-band satellite signals increases by an average of 0.2 dB; for every 10% increase in humidity, the signal attenuation increases by 0.15 dB; for every 1 increase in the density of 5G base stations per square kilometer, the interference intensity increases by an average of 0.5 dB. These quantitative relationship models enable the system to predict interference changes based on external factor changes and improve the prediction accuracy.

[0123] The prediction interval calculation unit 435 is used to provide a confidence interval for the prediction result. Preferably, the confidence interval calculation is based on the historical error distribution of the prediction model, and usually provides prediction intervals at two confidence levels of 90% and 95%. For example, the system may predict that the signal-to-noise ratio after 24 hours is 8.5 dB, the 90% confidence interval is [7.8 dB, 9.2 dB], and the 95% confidence interval is [7.5 dB, 9.5 dB]. This interval prediction provides uncertainty information for decision-making, which helps risk assessment and strategy selection. In practical applications, for important services, usually the lower limit of the 95% confidence interval is used as the worst-case estimate for conservative decision-making; for general services, the mean of the 90% confidence interval can be used for decision-making.

[0124] Refer to Figure 9 , the execution monitoring module 35 of the present invention includes: a task decomposition unit 351, an execution coordination unit 352, a real-time monitoring unit 353, a dynamic adjustment unit 354, and an effect evaluation unit 355.

[0125] The task decomposition unit 351 is used to decompose the policy into specific execution steps. Preferably, the task decomposition adopts the Work Breakdown Structure (WBS) method to decompose complex policies into actionable specific tasks layer by layer. For example, the policy of "switching to the standby frequency band" may be decomposed into specific steps such as: preparing the standby equipment (1h) → testing the status of the standby frequency band (0.5h) → notifying the user of the upcoming switch (0.5h) → performing the frequency band switch (0.25h) → verifying the performance of the new frequency band (1h) → resuming services (0.5h), etc. Each step has a clear time estimate and completion standard.

[0126] The execution coordination unit 352 is used to set the execution order and time nodes. Preferably, the execution coordination adopts the Critical Path Method (CPM) to determine task dependencies and the optimal execution order, and at the same time, scheduling optimization is carried out considering resource constraints. For example, some tasks can be executed in parallel to save time, while some tasks must be executed sequentially to ensure safety. The execution coordination unit 352 is also responsible for generating execution checkpoints and rollback points to ensure safe rollback in case of problems during execution. Usually, the system will set checkpoints before each key step, record the system status before the operation, and prepare for possible rollback.

[0127] The real-time monitoring unit 353 is used to collect data on the execution progress and effects. Preferably, the real-time monitoring adopts a distributed monitoring architecture to collect key parameter data from the transmitter and receiver ends. The sampling frequency is dynamically adjusted according to the execution stage, reaching up to 10 samples per second during the critical transition period and dropping to 1 sample per minute during the stable period to balance monitoring accuracy and system load. The monitoring data is transmitted to the satellite operation center in real-time, and the delay is usually controlled within 3 seconds to ensure timely detection and response to abnormal situations.

[0128] The dynamic adjustment unit 354 is used to adjust the execution plan based on real-time feedback. Preferably, the dynamic adjustment adopts a preset threshold trigger mechanism. When the monitored parameter deviates from the expected value by more than the preset threshold (usually 15%), the execution plan adjustment is triggered. For example, if the signal-to-noise ratio improvement after the frequency band switch does not meet the expectation (expected to increase by 2dB but actually only increases by 0.5dB), the system will automatically trigger an adjustment, which may increase the transmission power or try other standby frequency bands. The adjustment decision adopts a decision tree algorithm to select the most suitable adjustment plan according to the type and degree of deviation.

[0129] The effect evaluation unit 355 is used to quantitatively evaluate the effect of policy execution. Preferably, the effect evaluation adopts two methods: before-and-after comparison and target achievement degree. The before-and-after comparison calculates the improvement amplitude of key parameters, such as how many dB the signal-to-noise ratio has increased and how many orders of magnitude the bit error rate has decreased; the target achievement degree calculates the ratio of the actual effect to the expected target. For example, if the expected signal-to-noise ratio is increased by 2 dB and the actual increase is 1.6 dB, the achievement degree is 80%. The evaluation results are quantified on a scale of 1-5 points. A score of 3 or above is considered that the policy is effective, and if it is lower than 2 points, the policy needs to be re-formulated. The evaluation report is automatically sent to the self-learning module 45 of the AI analysis unit 4 as the basis for experience accumulation and model optimization.

[0130] Referring to Figure 10 , the method for satellite channel interference suppression and 4K signal enhancement based on AI prediction provided by the present invention is applied to the system, and includes the following steps:

[0131] Step S1, the receiving end unit 2 monitors the quality of the downlink signal, and generates an anomaly report when the signal quality is detected to be abnormal and sends it to the satellite operation center unit 3.

[0132] Preferably, the signal quality monitoring adopts a multi-parameter and multi-threshold mechanism, covering key indicators such as signal-to-noise ratio, bit error rate, and carrier-to-noise ratio. The anomaly detection adopts a dynamic baseline comparison method, and defines three levels of anomalies: mild, moderate, and severe according to the deviation degree. For example, for 4K video transmission, when the signal-to-noise ratio is lower than 9 dB, the anomaly monitoring is triggered; when it is lower than 7 dB, a moderate anomaly report is generated; when it is lower than 5 dB, a severe anomaly report is generated. The anomaly report is transmitted to the satellite operation center through a secure encryption channel to ensure data security.

[0133] Step S2, the satellite operation center unit 3 collects the anomaly report, and collects satellite status information and environmental data.

[0134] Preferably, the satellite status information includes orbital parameters, transponder working status, signal coverage, etc.; the environmental data includes meteorological data, ground 5G base station distribution, terrain and landform, etc. The data collection adopts a distributed architecture, obtains information from multiple authoritative data sources, and ensures the comprehensiveness and reliability of the data. The data update frequency is set according to the data change speed. The update frequency of key parameters is higher, and the update frequency of general parameters is lower, balancing real-time performance and system load.

[0135] Step S3, the AI analysis unit 4 analyzes the anomaly report, the satellite status information, and the environmental data, identifies interference characteristics, predicts interference trends, and generates policy suggestions.

[0136] Preferably, for interference feature recognition, a multi-feature extraction and multi-algorithm fusion method is adopted. The features include spectrum features, time-domain features, modulation features, etc., and the algorithms include SVM, KNN, random forest, etc. Interference trend prediction combines multiple time series analysis models such as ARIMA and LSTM, and also considers the influence of environmental factors. Strategy recommendation generation is based on the interference recognition results and predicted trends, and combines with the historical case library to form multi-level strategy recommendations including short-term, medium-term, and long-term. For example, in the face of 5G base station interference with a frequency of 3.7 GHz, a duration of about 2 hours, and a stable trend, the system may recommend increasing the transmission power by 5 dB in the short term, switching to the 3.9 GHz backup frequency band in the medium term, and considering using the Ka band alternative in the long term.

[0137] Step S4, the satellite operation center unit 3 formulates interference response strategies based on the above-mentioned strategy recommendations, including short-term strategies, medium-term strategies, and long-term strategies.

[0138] Preferably, for strategy formulation, a multi-dimensional decision matrix evaluation method is adopted, comprehensively considering factors such as interference severity, duration, development trend, resource cost, technical feasibility, and long-term impact. In the decision-making process, the weight of AI recommendation is 70%, and the weight of manual review is 30% to ensure the scientificity and reliability of the decision. The strategy level is determined according to the nature and trend of the interference. A common combination is that only short-term strategies are activated for minor short-term interference, both short-term and medium-term strategies are activated for continuous moderate interference, and all three-level strategies are activated for severe long-term interference to form a three-dimensional defense system.

[0139] Step S5, the satellite operation center unit 3 distributes the interference response strategies to the transmitting end unit 1 and / or the receiving end unit 2 for execution.

[0140] Preferably, for strategy distribution, a task decomposition and execution coordination method is adopted to transform the strategy into specific executable tasks, and clarify the execution order, time nodes, and standards. The task instructions are transmitted to the execution unit through a secure channel, including complete execution parameters, expected effects, and checkpoints. For example, the transmission power adjustment task includes detailed information such as the target power value, adjustment step size, process monitoring parameters, and target confirmation criteria to ensure that the execution unit can complete the task accurately without error.

[0141] Step S6, the satellite operation center unit 3 monitors the execution effect of the interference response strategy and feeds back the execution result to the AI analysis unit 4 for model optimization.

[0142] Preferably, the execution monitoring adopts a real-time data acquisition and dynamic threshold detection mechanism to promptly detect execution deviations and trigger adjustments. The effect evaluation adopts a before-and-after comparison and a target achievement quantification method, and evaluates from multiple dimensions such as signal quality improvement, service recovery degree, and user experience. The execution results are fed back to the AI analysis unit 4 in the form of structured data, serving as the basis for model optimization and experience accumulation. Experience shows that through continuous optimization of the execution feedback, the system's interference response ability and efficiency will continue to improve. The response speed will increase by about 30% after 3 months, and the interference resolution efficiency will increase by about 25% after 6 months.

[0143] The above specific implementation manners are only preferred examples of the present invention and are not used to limit the protection scope of the present invention. The components of the technical solution described in the present invention can be implemented in a variety of different structures and forms. Any person skilled in the art can modify or replace the details and order of the technical solution of the present invention without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be subject to the content of the claims.

[0144] The above; only the preferred specific implementation manner of the present invention; but the protection scope of the present invention is not limited thereto; any person familiar with the art within the scope disclosed by the present invention; according to the solution of the present invention and its improved conceptions; making the same replacement or change; should be covered by the protection scope of the present invention.

Claims

1. Satellite channel interference suppression and 4K signal enhancement system based on AI prediction, characterized by: include: The transmitting end unit is used to transmit the uplink signal and perform self-receiving monitoring; A receiving end unit, for receiving downlink signals, monitoring signal quality parameters, and generating an abnormality report when abnormal signal quality is detected; A satellite operation center unit, which is in communication with the transmitting end unit and the receiving end unit, is used to receive the abnormality report, collect satellite status information and environmental data, and generate an interference response strategy; The AI ​​analysis unit is communicatively connected to the satellite operation center unit, and is used to analyze the abnormality report, the satellite status information and the environmental data, identify interference characteristics, predict interference trends, and provide strategic recommendations to the satellite operation center unit.

2. The system according to claim 1, characterized in that The transmitting end unit comprises: A signal transmission module, used for transmitting satellite uplink signals; A self-receiving monitoring module, used for receiving and analyzing the return signal of the uplink signal, and generating self-receiving status data; A parameter adjustment module, used to adjust the transmission parameters according to the adjustment instruction sent by the satellite operation center unit; A status reporting module is used to send the self-received status data to the satellite operation center unit.

3. The system according to claim 1, characterized in that The receiving end unit comprises: A signal receiving module, used for receiving satellite downlink signals; A multi-dimensional monitoring module, used to monitor parameters such as signal-to-noise ratio, bit error rate, carrier-to-noise ratio, phase jitter and spectrum purity of the downlink signal; an anomaly detection module, for comparing the parameters with a dynamic baseline to identify slight anomalies, moderate anomalies, and severe anomalies; Image recognition module, used to obtain parameters that cannot be directly collected by identifying the device interface; The exception report module is used to generate an exception report including the exception level, exception parameters and exception duration.

4. The system according to claim 1, characterized in that The AI ​​analysis unit includes: A data fusion module, used for performing spatiotemporal alignment and reliability assessment on the abnormality report, the satellite status information and the environmental data; Interference identification module, used to extract interference features, match them with the interference pattern library, and identify interference sources; Trend prediction module, used to predict interference trends based on the time series analysis model; A strategy generation module, used to generate short-term, medium-term and long-term response strategy suggestions based on the interference source and the interference trend; The self-learning module is used to continuously update the interference pattern library and prediction model according to the strategy execution effect.

5. The system according to claim 1, characterized in that The satellite operation center unit comprises: A communication coordination module, used to establish a secure communication channel with the transmitting end unit and the receiving end unit; Data collection module, used to collect satellite status information, meteorological data, ground 5G base station distribution and other environmental data; Decision Matrix Module, used to construct a multi-dimensional decision matrix based on interference severity, duration and trend; A strategy formulation module, used to formulate a final strategy based on the strategy suggestions and resource cost-effectiveness improvement analysis provided by the AI ​​analysis unit; The execution monitoring module is used to decompose the final strategy into specific tasks, assign them to the transmitting end unit and the receiving end unit, and monitor the execution progress.

6. The system according to claim 5, characterized in that The decision matrix module includes: A three-dimensional assessment unit for comprehensive assessment based on disturbance severity, duration and trend; Cost-effectiveness unit, used to calculate the resource cost and effect improvement ratio of different strategies; Priority stratification unit, used to determine the priority of protection according to the importance of the business; Technical feasibility unit, used to evaluate the technical implementation difficulty of various solutions; Long-term impact unit to analyse the long-term impact and sustainability of strategies.

7. The system according to claim 4, characterized in that The strategy generation module includes: Short-term strategy unit, used to generate short-term strategies including dynamic adjustment of transmit power, adaptive switching of coding methods, fine-tuning of receiving end parameters, real-time adjustment of beam shape and signal processing enhancement; Mid-term strategy unit, used to generate mid-term strategies including intelligent selection of spare frequency bands, design of frequency hopping patterns, temporary change of modulation mode, dynamic allocation of bandwidth and time slot reorganization; The long-term strategy unit is used to generate long-term strategies including satellite resource re-planning, ground station location optimization, system architecture upgrade, spectrum resource application and equipment replacement.

8. The system according to claim 4, characterized in that The trend prediction module includes: A time series processing unit, used for performing time series decomposition on interference data; Multi-model analysis unit, used to combine multiple time series analysis models such as ARIMA and LSTM for analysis; A cycle identification unit is used to identify daily, weekly, monthly, and seasonal cycle patterns in interference; External factor analysis unit, used to quantify the impact of external factors such as meteorological changes and ground activities on interference; The prediction interval calculation unit is used to provide confidence intervals for prediction results.

9. The system according to claim 5, characterized in that The execution monitoring module comprises: Task decomposition unit, used to decompose the strategy into specific execution steps; Execution coordination unit, used to set execution order and time nodes; Real-time monitoring unit, used to collect execution progress and effect data; A dynamic adjustment unit for adjusting the execution plan based on real-time feedback; The effect evaluation unit is used to quantitatively evaluate the effect of strategy execution.

10. A satellite channel interference suppression and 4K signal enhancement method based on AI prediction, applied to a system as claimed in any one of claims 1 to 9, characterized in that: The following steps are involved: The receiving end unit monitors the downlink signal quality, and generates an abnormality report when abnormal signal quality is detected and sends it to the satellite operation center unit; The satellite operation center unit collects the abnormality report, and collects satellite status information and environmental data; The AI ​​analysis unit analyzes the abnormality report, the satellite status information and the environmental data, identifies interference features, predicts interference trends, and generates strategy recommendations; The satellite operation center unit formulates interference response strategies based on the strategy suggestions, including short-term strategies, mid-term strategies and long-term strategies; The satellite operation center unit sends the interference response strategy to the transmitting end unit and / or the receiving end unit for execution; The satellite operation center unit monitors the execution effect of the interference response strategy and feeds back the execution result to the AI ​​analysis unit for model optimization.

Images