Satellite-borne AIS receiving method and system based on channel state prediction

Abstract

The application provides a kind of satellite-borne AIS receiving method and system based on channel state prediction, it is related to satellite communication technical field, its method includes receiving AIS analog radio frequency signal by radio frequency front end and sequentially carrying out gain linear amplification, frequency mixing filtering and analog-digital conversion, generates AIS digital receiving signal;Based on AIS digital receiving signal, dynamic channel state vector is constructed and input to channel state prediction model to carry out Doppler prediction, and output channel state prediction result;According to channel state prediction result, determine Doppler two-dimensional joint search window and carry out frequency offset compensation to AIS digital receiving signal, generate AIS digital baseband signal;Based on AIS digital baseband signal, signal quality evaluation function is constructed and radio frequency front end is dynamically adjusted, so that adaptive receiving control facing dynamic channel state can be realized, improve the success rate of acquisition, demodulation accuracy and receiving stability of AIS analog radio frequency signal under dynamic satellite link.

Description

Technical Field

[0001] This invention relates to the field of satellite communication technology, and in particular to a satellite-borne AIS receiving method and system based on channel state prediction. Background Technology

[0002] With the increasing demands for maritime transport, ship supervision, and maritime safety, Automatic Identification Systems (AIS) have become an important technological means to obtain dynamic information about ships. AIS enables large-scale and continuous monitoring of ship information in global waters, especially in the open sea, and has significant application value in areas such as maritime traffic safety, fishery resource management, and marine environmental protection.

[0003] Traditional AIS systems primarily rely on ground base stations or shipborne terminals for data exchange. Their communication coverage is limited by geographical location, making it difficult to meet the ship information acquisition needs in open ocean areas, polar seas, and large-scale marine monitoring scenarios. Furthermore, in practical applications, existing spaceborne AIS receivers are prone to significant Doppler frequency shifts due to the large relative speed between the satellite and the ship. This causes a shift in the carrier frequency of the AIS analog radio frequency signal, leading to problems such as frequency synchronization difficulties, increased signal acquisition time, and degraded demodulation performance. Simultaneously, the AIS analog radio frequency signal attenuates significantly after long-distance propagation, resulting in lower effective signal strength received by the receiver. This makes it susceptible to space noise and other wireless communication signals in the VHF band, reducing the signal-to-noise ratio and lowering the reliability of data reception in weak signal environments.

[0004] Therefore, it is necessary to provide a satellite-borne AIS reception method and system based on channel state prediction to solve the above-mentioned technical problems. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a satellite-borne AIS receiving method and system based on channel state prediction. This method solves the problems of insufficient Doppler frequency shift compensation, difficulty in accurately predicting dynamic changes in channel state, signal attenuation caused by long-distance spatial propagation, poor anti-interference performance, and deteriorated demodulation performance of existing satellite-borne AIS receivers in high-dynamic, strong interference, and long-distance transmission scenarios.

[0006] This invention provides a satellite-borne AIS reception method based on channel state prediction, the method comprising: The AIS analog radio frequency signal is received by the radio frequency front-end of the onboard AIS receiver and then sequentially subjected to gain linear amplification, mixing filtering and analog-to-digital conversion to generate the AIS digital received signal. A dynamic channel state vector is constructed based on the AIS digital received signal. The dynamic channel state vector is input into the channel state prediction model for Doppler prediction, and the channel state prediction result is output. Based on the channel state prediction results, a two-dimensional Doppler joint search window is determined and frequency offset compensation is performed on the AIS digital received signal to generate an AIS digital baseband signal. A signal quality evaluation function is constructed based on the AIS digital baseband signal, and the radio frequency front end is dynamically adjusted.

[0007] Preferably, the step of receiving the AIS analog radio frequency signal through the radio frequency front-end of the spaceborne AIS receiver and sequentially performing gain linear amplification, mixing filtering, and analog-to-digital conversion to generate the AIS digital received signal specifically includes: The AIS analog radio frequency signal is received through the radio frequency front-end of the onboard AIS receiver; The AIS analog RF signal is linearly amplified by a low-noise amplifier to output an AIS RF amplified signal. The AIS RF amplified signal is then input to a phase-locked loop frequency synthesizer for mixing and down-conversion, and finally bandpass filtered by an adaptive RF filter network to obtain the AIS analog received signal. The AIS analog received signal is bandpass sampled using a high-speed analog-to-digital converter to generate the AIS digital received signal.

[0008] Preferably, the step of constructing a dynamic channel state vector based on the AIS digital received signal specifically includes: A preset time sliding window is used to segment the AIS digital received signal based on the time sliding window to obtain an AIS digital signal sequence. The AIS digital signal sequence is subjected to spectral transformation using a fast Fourier transform algorithm to extract Doppler frequency shift features; The AIS digital signal sequence is subjected to quadrature demodulation to obtain in-phase component data and quadrature component data, and phase change analysis is performed to extract phase drift trend characteristics; The sampling amplitude data corresponding to the AIS digital signal sequence is obtained and the signal amplitude change is calculated. The trend analysis of the signal amplitude change is performed to obtain the signal attenuation trend. The signal amplitude change and the signal attenuation trend are normalized respectively to generate signal amplitude fluctuation characteristics and signal attenuation rate characteristics, which are then combined to obtain signal amplitude attenuation characteristics. Retrieve the local reference sequence, perform time-domain sliding correlation detection on the AIS digital signal sequence and the local reference sequence, and extract the propagation delay variation characteristics; The dynamic channel state vector is constructed by integrating the Doppler frequency shift characteristics, the phase drift trend characteristics, the signal amplitude attenuation characteristics, and the propagation delay variation characteristics.

[0009] Preferably, the step of inputting the dynamic channel state vector into the channel state prediction model for Doppler prediction and outputting the channel state prediction result specifically includes: The preamble field in the AIS digital signal sequence is located, and the preamble field is equally divided along the time axis into a preamble segment and a preamble segment. Fast Fourier Transform spectral analysis is then performed on both the preamble segment and the preamble segment to extract the estimated Doppler frequency shift of the preamble segment. and the estimated Doppler frequency shift of the preceding sub-segment And calculate the measured Doppler rate of change. In the formula, This indicates the time interval between the preamble's preceding sub-segment and the preamble's following sub-segment; Historical channel state sequences are retrieved, and the historical channel state sequences are time-aligned with the dynamic channel state vector to construct a multi-scale channel state evolution sequence, which is then input into the channel state prediction model. The dependencies of the multi-scale channel state evolution sequence are mined through a time-series feature extraction layer to generate channel state evolution trend features. The measured Doppler change rate is concatenated with the channel state evolution trend feature to obtain the fused prediction feature F. This fused prediction feature is then input into the state transition equation, and joint predictions are performed on the Doppler frequency shift feature, the phase drift trend feature, the signal amplitude attenuation feature, and the propagation delay change feature for the future prediction window. The channel state prediction result is then generated by summarizing these predictions. The corresponding calculation formula is as follows: In the formula, These represent the predicted Doppler frequency shift characteristics, predicted Doppler rate of change, predicted phase drift trend characteristics, predicted signal amplitude attenuation characteristics, and predicted propagation delay variation characteristics of the future prediction window, respectively. This represents the Doppler frequency shift characteristic in the dynamic channel state vector; This represents the Doppler acceleration obtained based on the fitting of fused prediction features; This represents the phase drift trend characteristics in the dynamic channel state vector; This represents the phase drift rate obtained based on the fitting of fused prediction features; represent the amplitude attenuation mapping function and the propagation delay mapping function, respectively; t represents the reception time of the AIS digital received signal; This indicates the step length of the future forecast window.

[0010] Preferably, determining the Doppler two-dimensional joint search window based on the channel state prediction result specifically includes: Preset Doppler frequency shift error Doppler rate of change error The search range for predicted Doppler frequency shift features is limited based on the channel state prediction results. and the search range for predicting Doppler rate of change The corresponding formula is as follows: In the formula, These represent the lower limit and upper limit of the search for predicting Doppler frequency shift features, respectively. These represent the lower limit of the search for predicting the Doppler rate of change and the upper limit of the search for predicting the Doppler rate of change, respectively. Using the predicted Doppler frequency shift feature as the horizontal axis and the predicted Doppler rate of change as the vertical axis, the search range of the predicted Doppler frequency shift feature is determined. Search range of the predicted Doppler rate of change Perform orthogonal cross-multiplication to generate the Doppler two-dimensional joint search window.

[0011] Preferably, frequency offset compensation is performed on the AIS digital received signal based on the Doppler two-dimensional joint search window to generate the AIS digital baseband signal, specifically including: Traverse the two-dimensional grid search points in the two-dimensional joint Doppler search window, construct Doppler compensation parameters based on the predicted Doppler frequency shift characteristics and the predicted Doppler change rate corresponding to the two-dimensional grid search points, and use the Doppler compensation parameters to perform frequency offset compensation on the AIS digital received signal to obtain an intermediate compensated digital signal; The local reference sequence is invoked, and the intermediate compensated digital signal corresponding to each two-dimensional grid search point is subjected to conjugate sliding correlation operation with the local reference sequence to generate a correlation peak sequence corresponding to each two-dimensional grid search point. The corresponding calculation formula for the correlation peak is as follows: In the formula, This represents the correlation peak value corresponding to the number of time-domain sliding sampling offset points m; This represents the intermediate compensation digital signal corresponding to the search point in the two-dimensional grid; The local reference sequence is represented by ; m represents the number of time-domain sliding sampling offset points of the local reference sequence relative to the intermediate compensated digital signal; N represents the number of sampling points of the local reference sequence; n represents the sampling point index of the local reference sequence; * represents the complex conjugate operator. The relevant peak sequence corresponding to the search points in the two-dimensional grid is traversed, and the Doppler compensation parameter corresponding to the largest relevant peak is selected as the optimal compensation parameter. The optimal compensation parameter is then used to perform frequency offset compensation on the AIS digital received signal, and the AIS digital baseband signal is output.

[0012] Preferably, the step of constructing a signal quality evaluation function based on the AIS digital baseband signal and dynamically adjusting the RF front-end specifically includes: The AIS digital baseband signal is demodulated to generate the baseband real-time signal-to-noise ratio (SNR) and the mean symbol phase error. Inter-symbol interference amplitude The bit error rate (BER) is verified, and combined with the predicted phase drift trend characteristics. The predicted signal amplitude attenuation characteristics The predicted propagation delay variation characteristics Construct the signal quality evaluation function and output the comprehensive quality score Q of the AIS digital baseband signal; A preset high-quality reception threshold and a low-quality reception threshold are defined. When the overall quality score Q is greater than or equal to the high-quality reception threshold, the RF parameters of the RF front-end remain unchanged. When the overall quality score Q is greater than the low-quality reception threshold but less than the high-quality reception threshold, a small adjustment command is output and the RF parameters are updated. When the overall quality score Q is less than or equal to the low-quality reception threshold, a large adjustment command is output and the RF parameters are updated.

[0013] Preferably, updating the radio frequency parameters via the small adjustment command and updating the radio frequency parameters via the large adjustment command specifically includes: Based on the predicted phase drift trend characteristics The predicted signal amplitude attenuation characteristics The predicted propagation delay variation characteristics Construct channel degradation correction coefficients; Read the basic small-amplitude adjustment step size from the small-amplitude adjustment instruction and multiply it by the channel degradation correction coefficient to calculate the actual small-amplitude adjustment step size of the radio frequency parameters and update the radio frequency parameters; Read the basic large adjustment step size in the large adjustment command and multiply it by the channel degradation correction coefficient to calculate the actual large adjustment step size of the radio frequency parameters and update the radio frequency parameters.

[0014] A spaceborne AIS receiving system based on channel state prediction, the system comprising: The radio frequency acquisition module is used to receive AIS analog radio frequency signals through the radio frequency front end of the onboard AIS receiver and sequentially perform gain linear amplification, mixing filtering and analog-to-digital conversion to generate AIS digital received signals; The channel prediction module is used to construct a dynamic channel state vector based on the AIS digital received signal, input the dynamic channel state vector into the channel state prediction model for Doppler prediction, and output the channel state prediction result. The offset compensation module is used to determine the Doppler two-dimensional joint search window based on the channel state prediction result and to perform frequency offset compensation on the AIS digital received signal to generate an AIS digital baseband signal. The radio frequency control module is used to construct a signal quality evaluation function based on the AIS digital baseband signal and to dynamically adjust the radio frequency front end.

[0015] Compared with existing technologies, the satellite-borne AIS receiving method and system based on channel state prediction provided by this invention has the following beneficial effects: This invention receives AIS analog RF signals through the RF front-end of a spaceborne AIS receiver and sequentially performs gain linear amplification, mixing filtering, and analog-to-digital conversion to generate an AIS digital received signal. Based on the AIS digital received signal, a dynamic channel state vector is constructed. This dynamic channel state vector is input into a channel state prediction model for Doppler prediction, and the channel state prediction result is output. A two-dimensional Doppler joint search window is determined based on the channel state prediction result, and frequency offset compensation is performed on the AIS digital received signal to generate an AIS digital baseband signal. Based on the AIS digital baseband signal, a signal quality evaluation function is constructed, and the RF front-end is dynamically adjusted. This enables adaptive reception control for dynamic channel states, improving the acquisition success rate, demodulation accuracy, and reception stability of AIS analog RF signals under dynamic satellite links.

[0016] This invention is applied to the reception, channel state prediction, and adaptive dynamic adjustment of AIS analog RF signals in the VHF band. The invention receives VHF AIS analog RF signals through the RF front-end of an onboard AIS receiver and sequentially performs linear gain amplification, mixing filtering, and analog-to-digital conversion, effectively improving the reception sensitivity and dynamic range of weak AIS analog RF signals while suppressing out-of-band noise and image frequency interference. The invention uses a channel state prediction model for Doppler prediction, enabling early perception of dynamic satellite link changes. By constructing a two-dimensional joint Doppler search window using predicted Doppler frequency shift characteristics and predicted Doppler change rate, the invention transforms global blind search into a two-dimensional grid fine search, reducing computational complexity and improving frequency offset correction accuracy. Based on the AIS digital baseband signal, the invention constructs a signal quality evaluation function and dynamically adjusts the RF front-end, establishing a closed-loop feedback mechanism for baseband quality assessment and RF parameter control. It adaptively adjusts the step size and direction according to the channel degradation correction coefficient, enabling the RF front-end to track and adapt to dynamic channel states in real time, ensuring the stability and demodulation accuracy of AIS analog RF signal reception under highly dynamic satellite links. Attached Figure Description

[0017] Figure 1 A flowchart of a satellite-borne AIS receiving method based on channel state prediction provided in an embodiment of the present invention; Figure 2 A system block diagram of a spaceborne AIS receiving system based on channel state prediction is provided for an embodiment of the present invention; Figure 3 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

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

[0019] like Figure 1 The diagram shown is a flowchart of a satellite-borne AIS reception method based on channel state prediction provided by an embodiment of the present invention. Figure 1The execution subject of the method shown can be a software and / or hardware device. The execution subject of this invention can include, but is not limited to, at least one of the following: user equipment, network equipment, etc. User equipment can include, but is not limited to, computers, smartphones, personal digital assistants (PDAs), and the aforementioned electronic devices. Network equipment can include, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers. Cloud computing is a type of distributed computing, consisting of a super virtual computer composed of a group of loosely coupled computers. This embodiment does not limit this. Steps S1 to S4 are detailed as follows: S1 receives the AIS analog radio frequency signal through the radio frequency front-end of the onboard AIS receiver and performs linear gain amplification, mixing and filtering and analog-to-digital conversion in sequence to generate the AIS digital received signal; The process of receiving AIS analog radio frequency signals through the radio frequency front-end of the onboard AIS receiver and sequentially performing gain linear amplification, mixing filtering, and analog-to-digital conversion to generate AIS digital received signals specifically includes: The AIS analog radio frequency signal is received through the radio frequency front-end of the onboard AIS receiver; The AIS analog RF signal is linearly amplified by a low-noise amplifier to output an AIS RF amplified signal. The AIS RF amplified signal is then input to a phase-locked loop frequency synthesizer for mixing and down-conversion, and finally bandpass filtered by an adaptive RF filter network to obtain the AIS analog received signal. The AIS analog received signal is bandpass sampled using a high-speed analog-to-digital converter to generate the AIS digital received signal.

[0020] The radio frequency (RF) front-end refers to the analog signal processing link in the spaceborne AIS receiver responsible for receiving, amplifying, frequency conversion, and filtering. The topology of the RF front-end includes hardware such as an antenna receiving port, a low-noise amplifier, a phase-locked loop frequency synthesizer, an adaptive RF filter network, and a high-speed analog-to-digital converter.

[0021] The shipboard terminal of the Automatic Identification System (AIS) transmits an analog AIS radio frequency signal in the VHF band. After propagating over a long distance, the signal reaches the radio frequency front-end of the onboard AIS receiver. Due to the high-speed relative motion between the satellite and the ship, the analog AIS signal undergoes Doppler shift and amplitude attenuation by the time it reaches the antenna receiving port. A low-noise amplifier linearly amplifies the analog AIS signal, outputting the amplified AIS radio frequency signal while maintaining its linearity during the amplification process. The AIS RF amplified signal is input to a phase-locked loop (PLL) frequency synthesizer for mixing and down-conversion. A high-precision, low-phase-noise local oscillator reference signal generated by the PLL frequency synthesizer is multiplied with the AIS RF amplified signal to generate a mixed output signal containing a difference frequency component. This mixed output signal is then bandpass filtered by an adaptive RF filter network to remove high-frequency image components and out-of-band noise, yielding the AIS analog received signal. Finally, a high-speed analog-to-digital converter (ADC) performs bandpass sampling on the AIS analog received signal and quantizes and encodes it according to a preset sampling clock frequency, converting the AIS analog received signal into an AIS digital received signal, thereby reducing the sampling rate of the high-speed ADC.

[0022] S2, construct a dynamic channel state vector based on the AIS digital received signal, input the dynamic channel state vector into the channel state prediction model for Doppler prediction, and output the channel state prediction result; The construction of the dynamic channel state vector based on the AIS digital received signal specifically includes: A preset time sliding window is used to segment the AIS digital received signal based on the time sliding window to obtain an AIS digital signal sequence. The AIS digital signal sequence is subjected to spectral transformation using a fast Fourier transform algorithm to extract Doppler frequency shift features; The AIS digital signal sequence is subjected to quadrature demodulation to obtain in-phase component data and quadrature component data, and phase change analysis is performed to extract phase drift trend characteristics; The sampling amplitude data corresponding to the AIS digital signal sequence is obtained and the signal amplitude change is calculated. The trend analysis of the signal amplitude change is performed to obtain the signal attenuation trend. The signal amplitude change and the signal attenuation trend are normalized respectively to generate signal amplitude fluctuation characteristics and signal attenuation rate characteristics, which are then combined to obtain signal amplitude attenuation characteristics. Retrieve the local reference sequence, perform time-domain sliding correlation detection on the AIS digital signal sequence and the local reference sequence, and extract the propagation delay variation characteristics; The dynamic channel state vector is constructed by integrating the Doppler frequency shift characteristics, the phase drift trend characteristics, the signal amplitude attenuation characteristics, and the propagation delay variation characteristics.

[0023] The dynamic channel state vector is a multi-dimensional feature vector composed of Doppler frequency shift characteristics, phase drift trend characteristics, signal amplitude attenuation characteristics, and propagation delay variation characteristics. It comprehensively describes the real-time dynamic characteristics of the communication link between satellites and ships.

[0024] A preset time sliding window is used, with its length set to an integer multiple of the frame period of the AIS digital received signal. A non-overlapping sliding interception mechanism is employed to move the time sliding window along the time axis of the AIS digital received signal, dividing the AIS digital received signal into multiple continuous signal segments, resulting in multiple sets of AIS digital signal sequences. Each signal segment is a set of AIS digital signal sequences, and each set of AIS digital signal sequences corresponds to the AIS digital received signal within a time sliding window. This ensures the temporal continuity between the AIS digital signal sequences and reflects the dynamic changes of the AIS digital received signal.

[0025] A fast Fourier transform frequency domain analysis is performed on the AIS digital signal sequence to obtain the signal spectrum curve and locate the spectral peak position in the signal spectrum curve. The frequency value corresponding to the spectral peak position is taken as the actual carrier frequency. The difference between the actual carrier frequency and the preset carrier frequency is calculated to obtain the carrier frequency offset and is used as the Doppler frequency shift feature, thereby obtaining the frequency change information caused by the relative motion between the satellite and the ship.

[0026] The AIS digital signal sequence is subjected to quadrature demodulation processing. The in-phase component data and quadrature component data are separated by low-pass filtering. The signal phase value corresponding to each AIS digital signal sampling point is calculated based on the in-phase component data and the quadrature component data. The signal phase value between consecutive AIS digital signal sampling points is analyzed to obtain the phase change trend of the AIS digital signal sequence over time. The phase drift trend feature is then extracted to characterize the phase change caused by the relative motion between the satellite and the ship, channel changes, and carrier synchronization errors.

[0027] The sampling amplitude data corresponding to the AIS digital signal sequence is acquired, and the signal amplitude change between adjacent AIS digital signal sampling points is calculated. Linear fitting is performed on the signal amplitude change to determine the signal attenuation trend of the AIS digital signal sequence over time. The signal amplitude fluctuation characteristics corresponding to the signal amplitude change are fused with the signal attenuation rate characteristics corresponding to the signal attenuation trend to obtain the signal amplitude attenuation characteristics, which are used to characterize the dynamic changes in signal strength caused by changes in propagation distance and spatial environment.

[0028] The local reference sequence is retrieved. This local reference sequence refers to a pre-stored known sequence that matches the format of the AIS digital signal sequence; it is typically the preamble field sequence or training sequence from the AIS digital signal sequence. The AIS digital signal sequence and the local reference sequence are then matched point-by-point in the time domain, and the correlation between them at different sliding positions is calculated to obtain the corresponding correlation detection results. Based on the position of the maximum correlation peak in the correlation detection results, the time offset of the AIS digital signal sequence relative to the local reference sequence is determined, and the changes in this time offset are analyzed to extract propagation delay variation characteristics, reflecting the spatial link transmission delay fluctuations.

[0029] The Doppler frequency shift features, phase drift trend features, signal amplitude attenuation features, and propagation delay variation features extracted from each AIS digital signal sequence are integrated in a fixed-dimensional order to generate a dynamic channel state vector.

[0030] The step of inputting the dynamic channel state vector into the channel state prediction model for Doppler prediction and outputting the channel state prediction result specifically includes: The preamble field in the AIS digital signal sequence is located, and the preamble field is equally divided along the time axis into a preamble segment and a preamble segment. Fast Fourier Transform spectral analysis is then performed on both the preamble segment and the preamble segment to extract the estimated Doppler frequency shift of the preamble segment. and the estimated Doppler frequency shift of the preceding sub-segment And calculate the measured Doppler rate of change. In the formula, This indicates the time interval between the preamble's preceding sub-segment and the preamble's following sub-segment; Historical channel state sequences are retrieved, and the historical channel state sequences are time-aligned with the dynamic channel state vector to construct a multi-scale channel state evolution sequence, which is then input into the channel state prediction model. The dependencies of the multi-scale channel state evolution sequence are mined through a time-series feature extraction layer to generate channel state evolution trend features. The measured Doppler change rate is concatenated with the channel state evolution trend feature to obtain the fused prediction feature F. This fused prediction feature is then input into the state transition equation, and joint predictions are performed on the Doppler frequency shift feature, the phase drift trend feature, the signal amplitude attenuation feature, and the propagation delay change feature for the future prediction window. The channel state prediction result is then generated by summarizing these predictions. The corresponding calculation formula is as follows: In the formula, These represent the predicted Doppler frequency shift characteristics, predicted Doppler rate of change, predicted phase drift trend characteristics, predicted signal amplitude attenuation characteristics, and predicted propagation delay variation characteristics of the future prediction window, respectively. This represents the Doppler frequency shift characteristic in the dynamic channel state vector; This represents the Doppler acceleration obtained based on the fitting of fused prediction features; This represents the phase drift trend characteristics in the dynamic channel state vector; This represents the phase drift rate obtained based on the fitting of fused prediction features; represent the amplitude attenuation mapping function and the propagation delay mapping function, respectively; t represents the reception time of the AIS digital received signal; This indicates the step length of the future forecast window.

[0031] During the acquisition of the measured Doppler rate of change, due to the high-speed relative motion between the satellite and the ship, the AIS digital signal sequence exhibits obvious time-varying Doppler characteristics at the onboard AIS receiver. If a single spectrum analysis is performed on the entire preamble field, only the average Doppler frequency shift can be obtained, and the dynamic changes of the Doppler frequency shift within the duration of the preamble cannot be captured. Using the local reference sequence as a fixed template, point-by-point time-domain sliding correlation operations are performed between the local reference sequence and the AIS digital signal sequence, and the correlation peak is detected. By setting a peak detection threshold and searching for the position of the maximum correlation peak, the start and end sampling points of the preamble field are located. When the correlation peak is greater than or equal to the peak detection threshold, it is determined that the AIS digital signal sampling point contains a preamble field, thereby locking the sampling interval of the preamble field. The preamble field is equally divided into a preamble segment and a preamble segment along the time axis of the AIS digital received signal. The preamble segment refers to the sub-segment located at the beginning of the time axis after the preamble field is equally divided along the time axis. It has the same number of sampling points as the preamble segment and covers the sampling data of the first half of the preamble field's duration, used to characterize the carrier frequency state near the start of the preamble. The preamble segment refers to the sub-segment located at the end of the time axis after the preamble field is equally divided along the time axis. It has the same number of sampling points as the preamble segment and covers the sampling data of the second half of the preamble field's duration, used to characterize the carrier frequency state near the end of the preamble.

[0032] There is a time interval between the preamble and the preamble, and the relative motion between the satellite and the ship causes the Doppler frequency shift to change continuously within this time interval. Therefore, there is a difference between the estimated Doppler frequency shift values ​​of the preamble and the preamble, which reflects the degree of change in Doppler frequency shift within this time interval. Fast Fourier Transform is performed on the preamble and the preamble separately to generate corresponding spectrum curves. The actual carrier frequency corresponding to the spectral peak is searched in the spectrum curves and the difference is calculated with the AIS nominal carrier frequency to obtain the estimated Doppler frequency shift value. The estimated Doppler frequency shift value of the preamble is subtracted from the estimated Doppler frequency shift value of the preamble by the estimated Doppler frequency shift value of the preamble, and then divided by the time interval between the preamble and the preamble to obtain the measured Doppler change rate. This allows for the acquisition of real-time Doppler change trends, overcoming the limitation of existing technologies that can only obtain frequency offset at a single moment and cannot capture the trend of frequency offset changes.

[0033] The historical channel state sequence refers to multiple historical dynamic channel state vectors arranged chronologically before the current AIS digital signal sequence, based on their reception times. After obtaining the measured Doppler rate of change, it is necessary to combine the channel change patterns corresponding to the AIS digital received signals within the historical time period to perform time-series correlation analysis on the current dynamic channel state vector, thereby improving the accuracy of future channel state prediction. The acquisition timestamp corresponding to each historical dynamic channel state vector in the historical channel state sequence is obtained, and the historical dynamic channel state vectors are rearranged using the reception time of the AIS digital received signal corresponding to the current dynamic channel state vector as the time reference. After time alignment, the rearranged historical dynamic channel state vectors are concatenated with the current dynamic channel state vector to construct a multi-scale channel state evolution sequence. This multi-scale channel state evolution sequence simultaneously includes short-term real-time observation features and long-term historical evolution features. The multi-scale channel state evolution sequence is input into the channel state prediction model, and the time-series feature extraction layer tracks the dependencies and causal relationships of the channel state at different time scales, uncovering hidden time-series dependencies and outputting channel state evolution trend features.

[0034] The measured Doppler rate of change is fused with the channel state evolution trend characteristics to obtain fused prediction features. The measured Doppler rate of change provides the instantaneous frequency change rate information based on the preamble at the current moment, while the channel state evolution trend characteristics provide the long-term evolution law and trend inertia of the channel learned from historical data, exhibiting high stability and strong noise suppression capabilities. The fusion of these two features improves the adaptability of the channel state prediction model to dynamic satellite link changes. The fused prediction features are input into the state transition equation, which extrapolates and predicts the channel parameters within the future prediction window based on the fused prediction features. For the Doppler frequency shift characteristics, a first-order linear extrapolation algorithm is used to add the Doppler frequency shift characteristics in the dynamic channel state vector to the product of the measured Doppler rate of change and the step time, obtaining the predicted Doppler frequency shift characteristics at the end of the future prediction window. For the Doppler rate of change, the product of Doppler acceleration and step time is superimposed on the measured Doppler rate of change to obtain the predicted Doppler rate of change at the end of the future prediction window. This Doppler acceleration is obtained by nonlinear fitting of the fused prediction features and is used to correct the trend of the Doppler rate of change itself. For the phase drift trend feature, the product of the phase drift trend feature in the dynamic channel state vector and the phase drift rate and step time is accumulated for prediction. This phase drift rate is also obtained by nonlinear fitting of the fused prediction features and is used to characterize the speed of phase accumulation. For the signal amplitude attenuation feature and propagation delay variation feature, since the correlation between the signal amplitude attenuation feature and propagation delay variation feature and the channel environment is complex and nonlinearly coupled, the fused prediction features are nonlinearly mapped end-to-end through amplitude attenuation mapping function and propagation delay mapping function, respectively, to output the predicted signal amplitude attenuation feature and predicted propagation delay variation feature within the future prediction window.

[0035] Through the aforementioned joint prediction method, the channel state prediction model can use the measured Doppler change rate and channel state evolution trend characteristics to predict the future satellite communication link state in advance. This enables the onboard AIS receiver to adjust frequency compensation, phase correction, gain control, and synchronization parameters in advance based on the predicted channel state results, thereby improving the acquisition success rate, demodulation accuracy, and reception stability of AIS digital received signals in dynamic satellite link environments.

[0036] S3, determine the Doppler two-dimensional joint search window based on the channel state prediction result and perform frequency offset compensation on the AIS digital received signal to generate the AIS digital baseband signal; The step of determining the Doppler two-dimensional joint search window based on the channel state prediction result specifically includes: Preset Doppler frequency shift error Doppler rate of change error The search range for predicted Doppler frequency shift features is limited based on the channel state prediction results. and the search range for predicting Doppler rate of change The corresponding formula is as follows: In the formula, These represent the lower limit and upper limit of the search for predicting Doppler frequency shift features, respectively. These represent the lower limit of the search for predicting the Doppler rate of change and the upper limit of the search for predicting the Doppler rate of change, respectively. Using the predicted Doppler frequency shift feature as the horizontal axis and the predicted Doppler rate of change as the vertical axis, the search range of the predicted Doppler frequency shift feature is determined. Search range of the predicted Doppler rate of change Perform orthogonal cross-multiplication to generate the Doppler two-dimensional joint search window.

[0037] Based on the channel state prediction results, a two-dimensional joint search window for Doppler is determined. By combining the predicted Doppler frequency shift characteristics and predicted Doppler change rate within the future prediction time window, the possible range of Doppler changes is limited in the two-dimensional parameter space, and joint search and frequency offset compensation are performed on the AIS digital received signal to improve the acquisition accuracy of AIS digital received signal in dynamic satellite link environment.

[0038] The two-dimensional Doppler joint search window refers to using predicted Doppler frequency shift characteristics and predicted Doppler rate of change as two search dimensions. It forms a rectangular grid search area by orthogonally intersecting the predicted Doppler frequency shift characteristic variation range and the predicted Doppler rate of change variation range. The horizontal axis corresponds to the predicted Doppler frequency shift characteristics, and the vertical axis corresponds to the predicted Doppler rate of change. The interior of the two-dimensional Doppler joint search window consists of equally spaced two-dimensional grid search points. Each two-dimensional grid search point represents a set of Doppler compensation parameter combinations to be verified, which is used to limit the dynamic frequency shift compensation range and improve the accuracy of optimal compensation parameter selection.

[0039] Based on historical prediction deviation statistics from the onboard AIS receiver, the Doppler frequency shift error and Doppler rate of change error are adaptively updated. Predicted Doppler frequency shift features and predicted Doppler rate of change for the future prediction window are extracted from the channel state prediction results. Using the predicted Doppler frequency shift feature as the center value in the frequency domain, the Doppler frequency shift error is subtracted or added to calculate the lower and upper limits of the predicted Doppler frequency shift feature search, thus limiting the search range. Similarly, using the predicted Doppler rate of change as the center value, the Doppler rate of change error is subtracted or added to calculate the lower and upper limits of the predicted Doppler rate of change search, thus limiting the search range. Furthermore, a Cartesian coordinate system is established on the two-dimensional parameter plane, with the predicted Doppler frequency shift feature as the horizontal axis and the predicted Doppler rate of change as the vertical axis. The search range of the predicted Doppler frequency shift feature and the search range of the predicted Doppler rate of change are combined by Cartesian product to generate a regular rectangular region on the two-dimensional parameter plane with the predicted Doppler frequency shift feature and the predicted Doppler rate of change as the geometric center. This results in a two-dimensional joint Doppler search window, which shrinks the Doppler parameter space of the global blind search into this two-dimensional joint Doppler search window. This reduces the computational complexity of frequency offset compensation processing and the time consumption of parameter search, and improves search efficiency and signal acquisition reliability.

[0040] Based on the Doppler two-dimensional joint search window, frequency offset compensation is performed on the AIS digital received signal to generate the AIS digital baseband signal, specifically including: Traverse the two-dimensional grid search points in the two-dimensional joint Doppler search window, construct Doppler compensation parameters based on the predicted Doppler frequency shift characteristics and the predicted Doppler change rate corresponding to the two-dimensional grid search points, and use the Doppler compensation parameters to perform frequency offset compensation on the AIS digital received signal to obtain an intermediate compensated digital signal; The local reference sequence is invoked, and the intermediate compensated digital signal corresponding to each two-dimensional grid search point is subjected to conjugate sliding correlation operation with the local reference sequence to generate a correlation peak sequence corresponding to each two-dimensional grid search point. The corresponding calculation formula for the correlation peak is as follows: In the formula, This represents the correlation peak value corresponding to the number of time-domain sliding sampling offset points m; This represents the intermediate compensation digital signal corresponding to the search point in the two-dimensional grid; The local reference sequence is represented by ; m represents the number of time-domain sliding sampling offset points of the local reference sequence relative to the intermediate compensated digital signal; N represents the number of sampling points of the local reference sequence; n represents the sampling point index of the local reference sequence; * represents the complex conjugate operator. The relevant peak sequence corresponding to the search points in the two-dimensional grid is traversed, and the Doppler compensation parameter corresponding to the largest relevant peak is selected as the optimal compensation parameter. The optimal compensation parameter is then used to perform frequency offset compensation on the AIS digital received signal, and the AIS digital baseband signal is output.

[0041] The two-dimensional Doppler joint search window is discretized at equal intervals according to a preset frequency domain step size and a preset rate of change step size, generating a regularly distributed set of two-dimensional grid search points. The preset frequency domain step size and preset rate of change step size are set according to the width of the predicted Doppler frequency shift feature search range and the predicted Doppler rate of change search range, as well as real-time constraints. The preset frequency domain step size can be set to 1 / 10 to 1 / 20 of the width of the predicted Doppler frequency shift feature search range. Each two-dimensional grid search point in the set of two-dimensional grid search points is traversed sequentially. Doppler compensation parameters are constructed based on the predicted Doppler frequency shift feature and the predicted Doppler rate of change corresponding to the two-dimensional grid search point and represented in the form of a complex exponential sequence. The Doppler compensation parameters are multiplied by each sample point of the AIS digital signal received signal using a complex multiplication to compensate for the frequency offset of the AIS digital signal received, resulting in an intermediate compensated digital signal. In the specific implementation process, the Doppler compensation parameter corresponding to each two-dimensional grid search point is applied to the AIS digital received signal, so that the frequency offset phase rotation direction in the AIS digital received signal is opposite to the Doppler compensation parameter, thereby canceling the dynamic Doppler effect.

[0042] To determine the compensation effect at each 2D grid search point and evaluate whether the intermediate compensated digital signal truly eliminates Doppler frequency offset, a local reference sequence is invoked and slid along the time axis of the intermediate compensated digital signal, point by point. For each time-domain sliding sampling offset of *m* points, the N consecutive sampling points of the intermediate compensated digital signal starting from the *m*th sampling point are multiplied by the N sampling points of the local reference sequence using complex conjugate multiplication and summed at the corresponding time-domain sliding sampling offset position to generate the correlation peak sequence corresponding to that 2D grid search point. The set of 2D grid search points is traversed, and the correlation peak sequences corresponding to each 2D grid search point are compared. The Doppler compensation parameter corresponding to the largest correlation peak is selected as the optimal compensation parameter. The predicted Doppler frequency shift characteristic and predicted Doppler change rate corresponding to this optimal compensation parameter are the best estimates of the true Doppler state of the AIS digital received signal. Using this optimal compensation parameter, frequency offset compensation is re-performed on the AIS digital received signal, outputting an AIS digital baseband signal with accurate Doppler correction, thus achieving dynamic frequency offset compensation.

[0043] S4. Construct a signal quality evaluation function based on the AIS digital baseband signal and dynamically adjust the RF front end.

[0044] The process of constructing a signal quality evaluation function based on the AIS digital baseband signal and dynamically adjusting the RF front-end specifically includes: The AIS digital baseband signal is demodulated to generate the baseband real-time signal-to-noise ratio (SNR) and the mean symbol phase error. Inter-symbol interference amplitude The bit error rate (BER) is verified, and combined with the predicted phase drift trend characteristics. The predicted signal amplitude attenuation characteristics The predicted propagation delay variation characteristics Construct the signal quality evaluation function and output the comprehensive quality score Q of the AIS digital baseband signal; A preset high-quality reception threshold and a low-quality reception threshold are defined. When the overall quality score Q is greater than or equal to the high-quality reception threshold, the RF parameters of the RF front-end remain unchanged. When the overall quality score Q is greater than the low-quality reception threshold but less than the high-quality reception threshold, a small adjustment command is output and the RF parameters are updated. When the overall quality score Q is less than or equal to the low-quality reception threshold, a large adjustment command is output and the RF parameters are updated.

[0045] The AIS digital baseband signal is demodulated, and the ratio of average signal power to average noise power is calculated to generate the real-time signal-to-noise ratio (SNR). After symbol decision and phase recovery of the AIS digital baseband signal, the arithmetic mean of the differences between the actual carrier phase and the ideal reference phase of each demodulated symbol is calculated to generate the mean symbol phase error. This is used to characterize the tracking accuracy and residual phase jitter of the carrier synchronization loop. The amplitude of the interference signal generated by the superposition of adjacent demodulated symbols due to statistical channel multipath effects, propagation delay variations, etc., is used to generate the inter-symbol interference amplitude. This allows for the assessment of the crosstalk effect of multipath propagation and channel distortion on demodulated symbols. After decoding and CRC checking the AIS digital baseband signal, the bit error rate (BER) is obtained by calculating the ratio of the number of erroneous bits to the total number of bits, which reflects the reliability and data integrity of the link transmission.

[0046] A positive weight is assigned to the baseband real-time signal-to-noise ratio (SNR), and the mean symbol phase error is... Inter-symbol interference amplitude , Bit Error Rate (BER) and Predicted Phase Shift Trend Features Predicted signal amplitude attenuation characteristics Predicting the characteristics of propagation delay changes Negative weights are assigned to each signal, and a signal quality evaluation function is constructed using a linear weighted summation algorithm. The signal is then fused according to preset weights to obtain the comprehensive quality score of the AIS digital baseband signal. The corresponding calculation formula is as follows: In the formula, These represent the preset weights corresponding to each evaluation indicator, which can be calibrated offline or updated online adaptively based on the historical statistical data of the onboard AIS receiver.

[0047] Pre-configured high-quality reception thresholds and low-quality reception thresholds are used to classify the RF front-end's operating state into either the high-quality reception range or the significantly degraded range. When the overall quality score is greater than or equal to the high-quality reception threshold, it indicates that the current link signal-to-noise ratio is sufficient, phase distortion and inter-symbol interference are weak, and there is no significant trend of future channel degradation. The current RF parameters are suitable for reception requirements, and no additional adjustment is needed. When the overall quality score is greater than the low-quality reception threshold but less than the high-quality reception threshold, it is determined that the current channel has slight distortion and a slight decrease in reception quality. A small adjustment command is output to the RF front-end, driving it to slightly update its RF parameters to adapt to the slightly degraded channel. When the overall quality score is less than or equal to the low-quality reception threshold, it is determined that the current channel is in a poor reception state with strong attenuation and strong interference, and the baseband demodulation performance is on the verge of failure. A large adjustment command is output to the RF front-end, driving it to significantly update its RF parameters to improve the reception capability of weak signals and reconstruct the operating state of the reception link.

[0048] Updating the radio frequency parameters via the small adjustment command and the large adjustment command respectively includes: Based on the predicted phase drift trend characteristics The predicted signal amplitude attenuation characteristics The predicted propagation delay variation characteristics Construct channel degradation correction coefficients; Read the basic small-amplitude adjustment step size from the small-amplitude adjustment instruction and multiply it by the channel degradation correction coefficient to calculate the actual small-amplitude adjustment step size of the radio frequency parameters and update the radio frequency parameters; Read the basic large adjustment step size in the large adjustment command and multiply it by the channel degradation correction coefficient to calculate the actual large adjustment step size of the radio frequency parameters and update the radio frequency parameters.

[0049] Radio frequency (RF) parameters refer to the operating parameters of the analog signal processing link dynamically configured in the RF front end, including but not limited to the gain control word of the low-noise amplifier, the local oscillator frequency control word of the phase-locked loop frequency synthesizer, and the center frequency and bandwidth control word of the adaptive RF filter network, to achieve real-time matching and adaptation between the hardware receiving link and the dynamic satellite channel.

[0050] If a fixed basic adjustment step size is used to update RF parameters, excessively large basic adjustment steps may introduce additional phase noise and gain fluctuations during slight channel fluctuations, or insufficient basic adjustment steps may prevent timely restoration of link availability during rapid channel deterioration. Therefore, the basic adjustment step size needs to be adaptively adjusted based on the future channel degradation trend corresponding to the channel state prediction results to achieve a precise match between the RF front-end control strength and the degree of channel degradation. Predicted phase drift trend features, predicted signal amplitude attenuation features, and predicted propagation delay variation features are extracted from the channel state prediction results. The absolute values ​​of the predicted phase drift trend features, predicted signal amplitude attenuation features, and predicted propagation delay variation features are multiplied by their respective preset weights and summed, then superimposed with a reference unit of 1 to construct the channel degradation correction coefficient. The corresponding calculation formula is as follows: .

[0051] The basic small-amplitude adjustment step size built into the small-amplitude adjustment command is read and multiplied by the channel degradation correction coefficient to obtain the actual small-amplitude adjustment step size of the adaptively amplified RF parameters. The basic large-amplitude adjustment step size built into the large-amplitude adjustment command is read and multiplied by the channel degradation correction coefficient to obtain the actual large-amplitude adjustment step size of the adaptively amplified RF parameters. The basic small-amplitude adjustment step size and the basic large-amplitude adjustment step size are reference adjustment increment values ​​obtained during the orbital initialization phase of the onboard AIS receiver based on the physical adjustment resolution of the RF front-end hardware and on-orbit operation experience data. By introducing the channel degradation correction coefficient, the basic adjustment step size is dynamically scaled to improve the adaptive control robustness of the onboard AIS receiver in complex space electromagnetic environments.

[0052] Based on the current RF parameters and channel state prediction results, the direction of RF parameter adjustment is determined. Following this direction, the actual small or large adjustment step size of the RF parameters is accumulated and added to the current RF parameters, causing the RF front-end's operating state to converge towards adapting to the dynamic channel environment. After updating the RF parameters, the RF front-end reconfigures the analog signal processing link according to the updated parameters and re-acquires the AIS analog RF signal in the VHF band during the next reception cycle.

[0053] like Figure 2 The diagram shown is a system block diagram of a spaceborne AIS receiving system based on channel state prediction according to an embodiment of the present invention. The system includes: The radio frequency acquisition module is used to receive AIS analog radio frequency signals through the radio frequency front end of the onboard AIS receiver and sequentially perform gain linear amplification, mixing filtering and analog-to-digital conversion to generate AIS digital received signals; The channel prediction module is used to construct a dynamic channel state vector based on the AIS digital received signal, input the dynamic channel state vector into the channel state prediction model for Doppler prediction, and output the channel state prediction result. The offset compensation module is used to determine the Doppler two-dimensional joint search window based on the channel state prediction result and to perform frequency offset compensation on the AIS digital received signal to generate an AIS digital baseband signal. The radio frequency control module is used to construct a signal quality evaluation function based on the AIS digital baseband signal and to dynamically adjust the radio frequency front end.

[0054] Figure 2 The system of the illustrated embodiment can be used to perform corresponding operations. Figure 1 The steps in the method embodiments shown are implemented in a similar manner and have similar technical effects, and will not be repeated here.

[0055] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor performs the steps of a satellite AIS reception method based on channel state prediction as described in any of the preceding claims.

[0056] like Figure 3 The diagram shown is a hardware structure schematic of an electronic device according to an embodiment of the present invention. The electronic device 30 includes: a processor 31, a memory 32, and a computer program; wherein... The memory 32 is used to store the computer program, and the memory may also be flash memory. The computer program is, for example, an application program or functional module that implements the above method.

[0057] Processor 31 is configured to execute the computer program stored in the memory to implement the various steps performed by the device in the above method. For details, please refer to the relevant descriptions in the preceding method embodiments.

[0058] Alternatively, the memory 32 can be either standalone or integrated with the processor 31.

[0059] When the memory 32 is a device independent of the processor 31, the device may further include: Bus 33 is used to connect the memory 32 and the processor 31.

[0060] A readable storage medium storing a computer program, which, when executed by a processor, is used to implement the steps of a channel state prediction-based spaceborne AIS reception method as described in any of the preceding claims.

[0061] The readable storage medium can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of computer programs from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application-Specific Integrated Circuit (ASIC). Alternatively, the ASIC can be located in a user equipment. Of course, the processor and the readable storage medium can also exist as discrete components in a communication device. The readable storage medium can be a read-only memory (ROM), random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0062] The present invention also provides a program product including executable instructions stored in a readable storage medium. At least one processor of the device can read the executable instructions from the readable storage medium, and the at least one processor executes the executable instructions to cause the device to implement the methods provided in the various embodiments described above.

[0063] In the embodiments of the above-described device, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A satellite-borne AIS receiving method based on channel state prediction, characterized in that, The method includes: The AIS analog radio frequency signal is received by the radio frequency front-end of the onboard AIS receiver and then sequentially subjected to gain linear amplification, mixing filtering and analog-to-digital conversion to generate the AIS digital received signal. A dynamic channel state vector is constructed based on the AIS digital received signal. The dynamic channel state vector is input into the channel state prediction model for Doppler prediction, and the channel state prediction result is output. Based on the channel state prediction results, a two-dimensional Doppler joint search window is determined and frequency offset compensation is performed on the AIS digital received signal to generate an AIS digital baseband signal. A signal quality evaluation function is constructed based on the AIS digital baseband signal, and the radio frequency front end is dynamically adjusted.

2. The satellite-borne AIS reception method based on channel state prediction according to claim 1, characterized in that, The process of receiving AIS analog radio frequency signals through the radio frequency front-end of the onboard AIS receiver and sequentially performing gain linear amplification, mixing filtering, and analog-to-digital conversion to generate AIS digital received signals specifically includes: The AIS analog radio frequency signal is received through the radio frequency front-end of the onboard AIS receiver; The AIS analog RF signal is linearly amplified by a low-noise amplifier to output an AIS RF amplified signal. The AIS RF amplified signal is then input to a phase-locked loop frequency synthesizer for mixing and down-conversion, and finally bandpass filtered by an adaptive RF filter network to obtain the AIS analog received signal. The AIS analog received signal is bandpass sampled using a high-speed analog-to-digital converter to generate the AIS digital received signal.

3. The satellite-borne AIS reception method based on channel state prediction according to claim 1, characterized in that, The construction of the dynamic channel state vector based on the AIS digital received signal specifically includes: A preset time sliding window is used to segment the AIS digital received signal based on the time sliding window to obtain an AIS digital signal sequence. The AIS digital signal sequence is subjected to spectral transformation using a fast Fourier transform algorithm to extract Doppler frequency shift features; The AIS digital signal sequence is subjected to quadrature demodulation to obtain in-phase component data and quadrature component data, and phase change analysis is performed to extract phase drift trend characteristics; The sampling amplitude data corresponding to the AIS digital signal sequence is obtained and the signal amplitude change is calculated. The trend analysis of the signal amplitude change is performed to obtain the signal attenuation trend. The signal amplitude change and the signal attenuation trend are normalized respectively to generate signal amplitude fluctuation characteristics and signal attenuation rate characteristics, which are then combined to obtain signal amplitude attenuation characteristics. Retrieve the local reference sequence, perform time-domain sliding correlation detection on the AIS digital signal sequence and the local reference sequence, and extract the propagation delay variation characteristics; The dynamic channel state vector is constructed by integrating the Doppler frequency shift characteristics, the phase drift trend characteristics, the signal amplitude attenuation characteristics, and the propagation delay variation characteristics.

4. The satellite-borne AIS receiving method based on channel state prediction according to claim 3, characterized in that, The step of inputting the dynamic channel state vector into the channel state prediction model for Doppler prediction and outputting the channel state prediction result specifically includes: The preamble field in the AIS digital signal sequence is located, and the preamble field is equally divided along the time axis into a preamble segment and a preamble segment. Fast Fourier Transform spectral analysis is then performed on both the preamble segment and the preamble segment to extract the estimated Doppler frequency shift value of the preamble segment. and the estimated Doppler frequency shift of the preceding sub-segment And calculate the measured Doppler rate of change. In the formula, This indicates the time interval between the preamble's preceding sub-segment and the preamble's following sub-segment; Historical channel state sequences are retrieved, and the historical channel state sequences are time-aligned with the dynamic channel state vector to construct a multi-scale channel state evolution sequence, which is then input into the channel state prediction model. The dependencies of the multi-scale channel state evolution sequence are mined through a time-series feature extraction layer to generate channel state evolution trend features. The measured Doppler change rate is concatenated with the channel state evolution trend feature to obtain the fused prediction feature F. This fused prediction feature is then input into the state transition equation, and joint predictions are performed on the Doppler frequency shift feature, the phase drift trend feature, the signal amplitude attenuation feature, and the propagation delay change feature for the future prediction window. The channel state prediction result is then generated by summarizing these predictions. The corresponding calculation formula is as follows: In the formula, These represent the predicted Doppler frequency shift characteristics, predicted Doppler rate of change, predicted phase drift trend characteristics, predicted signal amplitude attenuation characteristics, and predicted propagation delay variation characteristics of the future prediction window, respectively. This represents the Doppler frequency shift characteristic in the dynamic channel state vector; This represents the Doppler acceleration obtained based on the fitting of fused prediction features; This represents the phase drift trend characteristics in the dynamic channel state vector; This represents the phase drift rate obtained based on the fitting of fused prediction features; represents the amplitude attenuation mapping function and the propagation delay mapping function, respectively; t represents the reception time of the AIS digital received signal; This indicates the step length of the future forecast window.

5. The satellite-borne AIS receiving method based on channel state prediction according to claim 4, characterized in that, The step of determining the Doppler two-dimensional joint search window based on the channel state prediction result specifically includes: Preset Doppler frequency shift error Doppler rate of change error The search range for predicted Doppler frequency shift features is limited based on the channel state prediction results. and the search range for predicting Doppler rate of change The corresponding formula is as follows: In the formula, These represent the lower limit and upper limit of the search for predicting Doppler frequency shift features, respectively. These represent the lower limit of the search for predicting the Doppler rate of change and the upper limit of the search for predicting the Doppler rate of change, respectively. Using the predicted Doppler frequency shift feature as the horizontal axis and the predicted Doppler rate of change as the vertical axis, the search range of the predicted Doppler frequency shift feature is determined. Search range of the predicted Doppler rate of change Perform orthogonal cross-multiplication to generate the Doppler two-dimensional joint search window.

6. The satellite-borne AIS receiving method based on channel state prediction according to claim 5, characterized in that, Frequency offset compensation is performed on the AIS digital received signal based on the Doppler two-dimensional joint search window to generate the AIS digital baseband signal, specifically including: Traverse the two-dimensional grid search points in the two-dimensional joint Doppler search window, construct Doppler compensation parameters based on the predicted Doppler frequency shift characteristics and the predicted Doppler change rate corresponding to the two-dimensional grid search points, and use the Doppler compensation parameters to perform frequency offset compensation on the AIS digital received signal to obtain an intermediate compensated digital signal; The local reference sequence is invoked, and the intermediate compensated digital signal corresponding to each two-dimensional grid search point is subjected to conjugate sliding correlation operation with the local reference sequence to generate a correlation peak sequence corresponding to each two-dimensional grid search point. The corresponding calculation formula for the correlation peak is as follows: In the formula, This represents the correlation peak value corresponding to the number of time-domain sliding sampling offset points m; This represents the intermediate compensation digital signal corresponding to the search point in the two-dimensional grid; The local reference sequence is represented by ; m represents the number of time-domain sliding sampling offset points of the local reference sequence relative to the intermediate compensated digital signal; N represents the number of sampling points of the local reference sequence; n represents the sampling point index of the local reference sequence; * represents the complex conjugate operator. The relevant peak sequence corresponding to the search points in the two-dimensional grid is traversed, and the Doppler compensation parameter corresponding to the largest relevant peak is selected as the optimal compensation parameter. The optimal compensation parameter is then used to perform frequency offset compensation on the AIS digital received signal, and the AIS digital baseband signal is output.

7. The satellite-borne AIS receiving method based on channel state prediction according to claim 4, characterized in that, The process of constructing a signal quality evaluation function based on the AIS digital baseband signal and dynamically adjusting the RF front-end specifically includes: The AIS digital baseband signal is demodulated to generate the baseband real-time signal-to-noise ratio (SNR) and the mean symbol phase error. Inter-symbol interference amplitude The bit error rate (BER) is verified, and combined with the predicted phase drift trend characteristics. The predicted signal amplitude attenuation characteristics The predicted propagation delay variation characteristics Construct the signal quality evaluation function and output the comprehensive quality score Q of the AIS digital baseband signal; A preset high-quality reception threshold and a low-quality reception threshold are defined. When the overall quality score Q is greater than or equal to the high-quality reception threshold, the RF parameters of the RF front-end remain unchanged. When the overall quality score Q is greater than the low-quality reception threshold but less than the high-quality reception threshold, a small adjustment command is output and the RF parameters are updated. When the overall quality score Q is less than or equal to the low-quality reception threshold, a large adjustment command is output and the RF parameters are updated.

8. A satellite-borne AIS receiving method based on channel state prediction according to claim 7, characterized in that, Updating the radio frequency parameters via the small adjustment command and the large adjustment command respectively includes: Based on the predicted phase drift trend characteristics The predicted signal amplitude attenuation characteristics The predicted propagation delay variation characteristics Construct channel degradation correction coefficients; Read the basic small-amplitude adjustment step size from the small-amplitude adjustment instruction and multiply it by the channel degradation correction coefficient to calculate the actual small-amplitude adjustment step size of the radio frequency parameters and update the radio frequency parameters; Read the basic large adjustment step size in the large adjustment command and multiply it by the channel degradation correction coefficient to calculate the actual large adjustment step size of the radio frequency parameters and update the radio frequency parameters.

9. A spaceborne AIS receiving system based on channel state prediction, characterized in that, The system, applied to a satellite-borne AIS reception method based on channel state prediction as described in any one of claims 1-8, comprises: The radio frequency acquisition module is used to receive AIS analog radio frequency signals through the radio frequency front-end of the onboard AIS receiver and sequentially perform gain linear amplification, mixing filtering and analog-to-digital conversion to generate AIS digital received signals; The channel prediction module is used to construct a dynamic channel state vector based on the AIS digital received signal, input the dynamic channel state vector into the channel state prediction model for Doppler prediction, and output the channel state prediction result. The offset compensation module is used to determine the Doppler two-dimensional joint search window based on the channel state prediction result and to perform frequency offset compensation on the AIS digital received signal to generate an AIS digital baseband signal. The radio frequency control module is used to construct a signal quality evaluation function based on the AIS digital baseband signal and to dynamically adjust the radio frequency front end.

Images