A low-frequency communication anti-interference control method and system based on a multi-layer defense body

Abstract

The application provides a low-frequency communication anti-interference control method and system based on a multilayer defense body, relates to the technical field of communication, and the method comprises the following steps: based on a signal anti-interference compensation coefficient, a multimode redundant architecture is used to preprocess a low-frequency communication signal, and a signal stream with time-frequency redundancy characteristics is generated; the signal stream with time-frequency redundancy characteristics is encrypted by a quantum key distribution protocol, and a quantum encrypted signal stream is generated; a blockchain-based identity authentication mechanism and a dynamic access control strategy are used to safely control the quantum encrypted signal stream, so that real-time protection and trusted access control of a communication link are realized. The application improves the stability and credibility of the communication link.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a low-frequency communication anti-interference control method and system based on a multi-layered defense system. Background Technology

[0002] During a transoceanic voyage, the ship's traditional low-frequency communication system transmits navigation data, exchanges route instructions, and conducts daily communication via satellite and ground base stations. The system's authentication and access control rely on a centralized server: the ship's terminal device identity information and communication permission rules are centrally stored in the land control center's database. Before each communication, a verification request must be sent to the central server, and a link connection is established after the verification is successful.

[0003] When a vessel was sailing in a certain offshore area, the ground control center server experienced a sudden data synchronization failure, resulting in partial corruption of the stored vessel identity keys and permission records. At this time, the vessel's terminal initiated a communication request according to the normal procedure. Due to the data anomaly, the center server could not correctly verify the vessel's identity and mistakenly identified the legitimate terminal as an unauthorized node. After the identity verification failed, the communication link was forcibly interrupted, and the vessel could not receive real-time navigation correction instructions from the ground center, nor could it upload vessel operation status data. At the same time, because the system lacked distributed evidence storage for the identity verification process, the ground center technicians had difficulty quickly locating the cause of the failure and initially misjudged it as a hardware malfunction of the vessel's terminal, delaying the opportunity to troubleshoot the problem. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a low-frequency communication anti-interference control method and system based on a multi-layer defense system, which avoids the single point of failure risk of centralized storage, can verify node identity in real time, record permission changes and trace abnormal behavior, form a security control closed loop, and improve the stability and reliability of the communication link.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] A first aspect is a low-frequency communication anti-interference control method based on a multi-layered defense system, the method comprising:

[0007] Low-frequency communication signal data is collected and converted into a set of spatiotemporal signal samples. A signal reference point is determined based on the spatiotemporal signal sample set. Two spatial orientation vectors are determined based on the signal reference point. The phase difference between the two spatial orientation vectors is calculated based on the spatiotemporal signal sample set to delineate a signal feature domain. Three signal monitoring nodes are determined using a distributed detection mechanism, with each node located in the effective and ineffective regions of the signal feature domain. A closed-loop signal phase path is constructed based on the phase correlation of each monitoring node in the time domain. A signal anti-interference compensation coefficient is generated based on the closed-loop signal phase path.

[0008] Based on the signal anti-interference compensation coefficient, a multi-mode redundancy architecture is used to preprocess low-frequency communication signals to generate a signal stream with time-frequency redundancy characteristics.

[0009] A quantum encrypted signal stream is generated by encrypting a signal stream with time-frequency redundancy using a quantum key distribution protocol.

[0010] A blockchain-based identity authentication mechanism and dynamic access control strategy are used to securely manage quantum encrypted signal streams, enabling real-time protection and trusted access control of communication links.

[0011] Secondly, a low-frequency communication anti-interference control system based on a multi-layered defense system includes:

[0012] The acquisition module is used to acquire low-frequency communication signal data and convert the low-frequency communication signal data into a set of spatiotemporal signal samples; determine a signal reference point based on the spatiotemporal signal sample set; determine two spatial orientation vectors based on the signal reference point; calculate the phase difference between the two spatial orientation vectors based on the spatiotemporal signal sample set to delineate a signal feature domain; determine three signal monitoring nodes using a distributed detection mechanism, with the three signal monitoring nodes located in the effective and invalid intervals of the signal feature domain, respectively; construct a closed signal phase path based on the phase correlation of each signal monitoring node in the time domain; and generate a signal anti-interference compensation coefficient based on the closed signal phase path.

[0013] The generation module is used to preprocess low-frequency communication signals based on the signal anti-interference compensation coefficient and adopt a multi-mode redundancy architecture to generate a signal stream with time-frequency redundancy characteristics.

[0014] The encryption module is used to encrypt a signal stream with time-frequency redundancy characteristics using a quantum key distribution protocol, thereby generating a quantum encrypted signal stream.

[0015] The control module is used to securely manage the quantum encrypted signal stream using a blockchain-based identity authentication mechanism and dynamic access control strategy, thereby achieving real-time protection and trusted access control of the communication link.

[0016] The above-described solution of the present invention has at least the following beneficial effects:

[0017] By analyzing spatiotemporal signal samples, calculating phase differences, and generating anti-interference compensation coefficients, signal characteristics can be accurately characterized and interference effects can be dynamically compensated. This effectively solves the problems of low-frequency signal distortion and attenuation in complex environments, enhancing signal transmission stability. Combining the time-frequency redundancy characteristics of a multi-mode redundancy architecture with the unconditional security of quantum key distribution, the risk of signal interception and cracking is significantly reduced, ensuring the confidentiality and integrity of data transmission in high-security scenarios. Based on blockchain-based identity authentication mechanisms and dynamic access control strategies, a distributed trusted architecture is constructed, avoiding the single-point failure risk of centralized storage. It can verify node identities in real time, record permission changes, and trace abnormal behavior, forming a closed-loop security management system and improving the stability and reliability of communication links. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a low-frequency communication anti-interference control method based on a multi-layered defense system, provided by an embodiment of the present invention.

[0019] Figure 2 This is a schematic diagram of a low-frequency communication anti-interference control system based on a multi-layered defense system provided by an embodiment of the present invention. Detailed Implementation

[0020] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0021] like Figure 1 As shown, an embodiment of the present invention proposes a low-frequency communication anti-interference control method based on a multi-layered defense system, the method comprising the following steps:

[0022] Step S1: Collect low-frequency communication signal data and convert it into a set of spatiotemporal signal sample sets; determine a signal reference point based on the spatiotemporal signal sample set; determine two spatial orientation vectors based on the signal reference point; calculate the phase difference between the two spatial orientation vectors based on the spatiotemporal signal sample set to delineate a signal feature domain; determine three signal monitoring nodes using a distributed detection mechanism, with the three monitoring nodes located in the effective and ineffective intervals of the signal feature domain, respectively; construct a closed-loop signal phase path based on the phase correlation of each monitoring node in the time domain; and generate a signal anti-interference compensation coefficient based on the closed-loop signal phase path.

[0023] Step S2: Based on the signal anti-interference compensation coefficient, a multi-mode redundancy architecture is used to preprocess the low-frequency communication signal to generate a signal stream with time-frequency redundancy characteristics.

[0024] Step S3: Encrypt the signal stream with time-frequency redundancy characteristics using a quantum key distribution protocol to generate a quantum encrypted signal stream;

[0025] Step S4: A blockchain-based identity authentication mechanism and dynamic access control strategy are used to securely manage the quantum encrypted signal stream, thereby achieving real-time protection and trusted access control of the communication link.

[0026] In this embodiment of the invention, by dividing the spatiotemporal signal feature domain, constructing the phase closed path, and generating compensation coefficients, signal interference characteristics are accurately captured, thereby improving the stability and anti-interference benchmark capability of the signal from the bottom layer. A multi-mode redundancy architecture preprocesses and generates a time-frequency redundant signal stream, using redundancy characteristics to offset signal distortion caused by interference, thereby enhancing the system's fault tolerance and self-healing capability against interference. Combined with quantum key distribution protocol encryption, relying on the unbreakable nature of quantum encryption, the confidentiality of signal transmission is greatly improved, preventing information from being stolen or tampered with. Identity authentication and dynamic access control supported by blockchain technology ensure that the identity of communication nodes is trustworthy and access is controllable. Combined with real-time protection strategies, it effectively resists malicious interference and illegal access, ensuring the trustworthiness and continuity of the communication link.

[0027] In a preferred embodiment of the present invention, step S1 involves acquiring low-frequency communication signal data and converting the low-frequency communication signal data into a set of spatiotemporal signal samples; determining a signal reference point based on the spatiotemporal signal sample set, including:

[0028] Step S100: Real-time acquisition of low-frequency communication signal data and preprocessing of the low-frequency communication signal data to obtain preprocessed signal data. Specifically, during transoceanic navigation, when acquiring low-frequency communication signal data in real time, the ship-mounted low-frequency communication terminal and satellite relay receiving equipment are used. The low-frequency communication terminal includes a signal receiving antenna and a radio frequency front-end module. For navigation data between the ship and the ground base station, such as latitude and longitude correction information, route instructions, such as turning instructions, and status reporting signals, such as ship speed and fuel consumption data, raw signal data are continuously collected at a sampling frequency of 1kHz. The sampling frequency is adapted to the low-frequency communication band of 3-30kHz. The data is transmitted to the signal processing unit in real time through the ship's local data bus, such as the CAN bus, and the timestamps corresponding to the data are recorded synchronously, accurate to the millisecond level. The data is associated with the ship's GPS positioning information to make the data spatiotemporally matched with the navigation status.

[0029] When preprocessing low-frequency communication signal data, the following operations are performed in combination with the characteristics of the offshore environment. Due to the presence of mechanical vibration interference from sea waves, atmospheric electromagnetic noise, and low-frequency signal interference from other ships in the offshore environment, non-target frequency band noise is first filtered through a hardware-level LC bandpass filter circuit (the center frequency matches the nominal communication frequency of the ground base station). Then, an adaptive noise cancellation algorithm (using the signal from a historical interference-free period as a reference template) is used to dynamically suppress sudden noise (such as instantaneous interference caused by lightning electromagnetic pulses) and retain the effective modulation components in the signal.

[0030] More specifically, since the ship's power supply system (such as a generator) may introduce DC offset, digital DC blocking processing is used (the average value of the first 100 sampling points of the signal sequence is calculated as the DC component, and this value is subtracted point by point) to avoid signal amplitude distortion caused by DC components (especially affecting the pulse amplitude recognition of navigation commands); according to the change in distance between the ship and the base station (such as from nearshore to offshore, the signal strength may decrease by more than 30%), the signal amplitude is mapped to the [-1, 1] interval through linear transformation (the difference between the maximum and minimum values ​​of the signal is calculated to eliminate the influence of the intensity difference caused by distance on subsequent feature extraction, and finally the preprocessed signal data is obtained).

[0031] Step S101: The preprocessed signal data is decomposed using the Daubechies wavelet basis function to obtain wavelet coefficients at each scale. The wavelet coefficients are then filtered, and the filtered wavelet coefficients at each scale are synthesized into a time-domain signal. A sliding time window is used to divide the time-domain signal into equal-length time intervals, and frequency domain features are extracted from the time-domain signal within each time interval to obtain a spatiotemporal signal sample set containing signal amplitude, phase, and frequency characteristic parameters. Specifically, this includes, when performing multi-resolution decomposition of the preprocessed signal data using the Daubechies wavelet basis function, targeting low-frequency communication signals from ships... The mixed characteristics of the signal (including continuous modulation signals of navigation data and pulse signals of commands) are analyzed. The db6 wavelet basis (whose symmetry and support length are adapted to the steep edges of pulse signals and the smooth changes of continuous signals) are selected. The preprocessed signal is decomposed into components of 5 scales: scales 1-2 correspond to high-frequency interference (such as instantaneous pulses caused by wave impact), and scales 3-5 correspond to low-frequency effective signals (10-20kHz modulation signals of navigation data and 5-10kHz pulse signals of commands). This yields the wavelet coefficient matrix at each scale. The row index of the matrix is ​​the time sampling point, the column index is the scale, and the element value is the characteristic amplitude at the corresponding time.

[0032] Among them, when filtering the wavelet coefficients, the characteristics of sudden interference in the open sea are taken into account:

[0033] For high-frequency coefficients at scales 1-2, a dynamic threshold based on signal-to-noise ratio (using the mean of the coefficient at that scale plus three times the standard deviation as the threshold) is used, and coefficients below the threshold are zeroed (judged as interference). For low-frequency coefficients at scales 3-5, all valid values ​​are retained because they contain the core features of navigation and command signals. The filtered coefficients at each scale are reconstructed through wavelet inverse transform to obtain the time-domain signal after removing high-frequency burst interference (this signal can clearly distinguish the continuous waveform of navigation data from the pulse waveform of commands). When using a sliding time window to segment the time-domain signal, combined with the transmission cycle of communication data, the ground base station sends navigation correction data every 10 seconds and route commands every 30 seconds. The window length is set to 10 seconds (covering one navigation data transmission cycle), and the sliding step size is 5 seconds (to achieve 50% overlap and avoid loss of command signals due to data truncation). Starting from the beginning position of the time-domain signal, the window is slid according to the step size, and the signal segment in each window is extracted, containing 10,000 sampling points, corresponding to 1kHz × 10 seconds, to obtain several time-segment signals of equal length.

[0034] When extracting frequency domain features from the time-domain signal within each time period, the differentiated features between navigation and command signals are determined, namely:

[0035] Amplitude characteristics are calculated by determining the average amplitude of the navigation data signal (reflecting signal transmission quality), the peak amplitude of the command pulse signal (used to distinguish different priority commands, such as emergency turn commands with higher amplitude), and the amplitude ratio of the two (to determine if the signal is affected by attenuation). Phase characteristics are extracted by extracting the carrier phase angle of the navigation signal (used for synchronous demodulation of data), the phase transition time of the command pulse (corresponding to the start / end marker of the command), and phase stability parameters; the smaller the standard deviation, the more stable the signal. Frequency characteristics are identified by identifying the main frequency of the navigation signal (which must match the base station's nominal frequency; a deviation exceeding ±0.5kHz indicates possible interference), the frequency bandwidth of the command signal (usually 2kHz, used to distinguish valid commands from interference), and the frequency interval between the two. The above characteristic parameters for each time period are organized according to the structure of time window number, amplitude, phase, and frequency to form a spatiotemporal signal sample set, with each row corresponding to a time window and each column corresponding to a characteristic parameter.

[0036] Step S102, based on the spatiotemporal signal sample set, extracts signal energy distribution features and determines the region of concentrated signal intensity by calculating the energy density of each frequency band, including:

[0037] Step S1021 involves windowing preprocessing of the spatiotemporal signal sample set. The Hanning window function is used to window each signal sample to obtain windowed signal samples. Specifically, when preprocessing each signal sample in the spatiotemporal signal sample set (i.e., a 10-second signal segment), due to the potential non-stationarity of ship signals (such as sudden attenuation of signals in the open sea), a Hanning window function with the same length as the sample (10,000 sampling points) is generated. The numerical distribution of this window function is as follows: the first 10% of sampling points smoothly rise from 0 to 0.5, and the middle 8... The 0% sampling points are stabilized at 0.5-1.0 to preserve the main body of the signal. The last 10% of sampling points smoothly decrease from 1.0 to 0 to avoid truncation at the end of the signal. For each signal sample, point-by-point multiplication is performed, that is, the continuous waveform segment of the navigation data and the steep segment of the command pulse in the sample are multiplied by the corresponding position of the Hanning window, so that the attenuation at both ends of the sample is more gradual (reducing spectral leakage caused by sudden signal attenuation). The effective signal in the middle segment (such as the peak part of the command pulse) maintains a higher amplitude. Finally, the windowed signal sample is obtained, and its waveform has no obvious abrupt change at the edge of the window.

[0038] Step S1022: Based on the windowed signal samples, convert them to the frequency domain using Fast Fourier Transform to obtain the spectral distribution of each signal sample. Specifically, when performing frequency domain conversion based on the windowed signal samples, the characteristics of navigation signals and command signals coexisting in the same frequency band in ship communication are taken into account.

[0039] For each windowed sample (10,000 points), a Fast Fourier Transform (FFT) is applied to convert the time-domain signal into the frequency domain (frequency range 0-500Hz to 25kHz, covering the low-frequency communication band). The amplitude of each frequency point is calculated by FFT. The navigation data signal corresponds to the main frequency peak of 15kHz, with a relatively stable amplitude, while the command signal corresponds to the pulse peak of 10kHz, with the amplitude fluctuating with the command transmission time. The frequencies and corresponding amplitudes are arranged in ascending order of frequency to form a spectral distribution curve, with the horizontal axis representing frequency and the vertical axis representing amplitude. The two main peaks on the curve correspond to the frequency components of the navigation and command signals, respectively.

[0040] Step S1023: Calculate the power spectral density of each frequency band based on the spectral distribution, and obtain the energy density value of each frequency band through integration. Specifically, when calculating the power spectral density of each frequency band based on the spectral distribution, the frequency planning for ship communication is considered. Navigation signals are 14-16kHz, command signals are 9-11kHz, and the 0-30kHz frequency range is divided into 300 frequency bands at 100Hz intervals. The 14-16kHz interval corresponds to 20 navigation signal frequency bands, the 9-11kHz interval corresponds to 20 command signal frequency bands, and the remainder are potential interference frequency bands. For each frequency band, calculate the energy density value within that band. The total power is obtained by summing the squares of the amplitudes at each frequency point; then, it is divided by the bandwidth of 100Hz to obtain the power spectral density. The power spectral density of the navigation and command bands is significantly higher than that of the interference bands, normally 5-10 times higher. The power spectral density of each band is integrated over its frequency range (e.g., 14000-14100Hz), i.e., the area enclosed by the power spectral density curve and this range. The integration result is the energy density of that band. The total energy density of the navigation signal band should account for 40%-50% of the total energy of the sample, and the command signal band should account for 20%-30%. If the ratio is abnormal, it indicates possible interference.

[0041] Step S1024: Based on the energy density values ​​of each frequency band, determine the energy density threshold and filter out frequency bands whose energy density exceeds the threshold. Specifically, when determining the energy density threshold based on the energy density values ​​of each frequency band, consider the environmental differences in ship navigation, where there is more near-shore interference and less offshore interference.

[0042] The energy density values ​​of all frequency bands in the current sea area (determined to be offshore by ship GPS positioning) are statistically analyzed, and their average value μ and standard deviation σ are calculated. The energy density of the interference frequency bands is concentrated around μ, while the energy density of the navigation / command frequency band is significantly higher than μ. An adaptive threshold rule is adopted, with the threshold set to μ+2σ in offshore areas (to reduce missed detection of weak signals) and μ+3σ in near-shore areas (to reduce false positives of interference frequency bands). Since the current scenario is offshore, the threshold is set to μ+2σ. All frequency bands are traversed, and frequency bands with energy density exceeding the threshold are selected, mainly the 14-16kHz navigation frequency band and the 9-11kHz command frequency band. If a frequency band exceeds the threshold even though it is not within the planned range, it is judged as abnormal interference.

[0043] Step S1025: Based on the frequency bands where the energy density exceeds the threshold, determine the region where the signal strength is concentrated. Specifically, this includes frequency planning for navigation and command signals when determining the region where the signal strength is concentrated based on the frequency bands where the energy density exceeds the threshold.

[0044] The selected frequency bands are sorted by frequency. Twenty consecutive frequency bands in the 14-16kHz range (intervals of 100Hz) are merged into the navigation signal strength concentration area of ​​14-16kHz; twenty consecutive frequency bands in the 9-11kHz range are merged into the command signal strength concentration area of ​​9-11kHz. If there are other discrete high-frequency bands with high energy density, such as around 18kHz, their interval with the planned frequency band is calculated. If it exceeds 5kHz, it is determined to be an interference band and is not included in the strength concentration area. Finally, two signal strength concentration areas are determined, corresponding to the main energy distribution ranges of navigation data and route commands, respectively.

[0045] Step S103: Based on the area of ​​concentrated signal strength, determine the core area of ​​signal energy distribution, and calculate and determine the centroid of the core area as the signal reference point. Specifically, when determining the core area of ​​signal energy distribution based on the area of ​​concentrated signal strength, the priority of command signals is higher than that of navigation data, taking into account the priority of ship communication.

[0046] The total energy of the two regions is calculated: the total energy of the navigation region is the sum of the energy densities of all frequency bands within 14-16kHz, and the total energy of the command region is the sum of the energy densities of all frequency bands within 9-11kHz. Since command signals (such as emergency turns) directly affect navigation safety, even if their total energy is slightly lower than that of the navigation region (e.g., 10%), the command region is still prioritized as the candidate core region. If the energy of the command region drops by more than 30% due to interference, the system switches to the navigation region to stabilize the core region. Since there is no significant interference in the current scenario, the command region of 9-11kHz is selected as the core region for signal energy distribution.

[0047] More specifically, when the center of gravity of the core area is calculated and determined as the signal reference point, the transmission characteristics of the associated command signal are as follows:

[0048] In the frequency dimension, the center frequency of each band within the 9-11kHz core region is calculated. For example, the center frequency of the 9000-9100Hz band is 9050Hz. The energy density of each band is used as a weight, with higher energy resulting in a larger weight. A weighted average is then used to obtain the centroid frequency, typically close to 10kHz, which is the nominal frequency of the command signal. In the time dimension, the occurrence times of the core region within the spatiotemporal sample set are statistically analyzed. Since the command is sent every 30 seconds, the midpoint of the 30-second period with the highest frequency of occurrence is selected, such as the 15th second, as the primary time coordinate. Combining the centroid frequency of 10kHz and the primary time coordinate of the 15th second, a two-dimensional coordinate point containing both time and frequency information is formed. This point serves as the signal reference point, providing a stable benchmark for subsequent analysis of the signal's spatial orientation vector, ensuring a stable basis for the phase and amplitude analysis of the command signal.

[0049] In this embodiment of the invention, hardware filtering and adaptive software noise reduction are combined to effectively eliminate interference from wave vibrations and electromagnetic noise in the offshore environment. Wavelet multi-resolution decomposition is used to accurately separate signal and noise components, ensuring that the preprocessed signal retains core information such as navigation data and commands. By using sliding time window segmentation and frequency domain feature extraction, the time-domain signal is transformed into a multi-dimensional sample set containing amplitude, phase, and frequency. Combined with Hanning window to suppress spectral leakage, the signal energy distribution analysis is more accurate, and the characteristic differences between navigation and command signals can be clearly distinguished. By calculating the frequency band energy density and thresholding, the signal intensity concentration area is accurately located, and the core energy distribution area is further determined and the centroid is calculated as a reference point, so that the signal analysis has a clear reference origin, improving the accuracy of subsequent phase, bearing, and other parameter calculations, especially adapting to the differentiated characteristic requirements of command and navigation signals in ship communication. Adaptive thresholds are used to address the differences between offshore and nearshore environments. Combined with ship communication frequency planning and data transmission cycle optimization, the processing parameters are optimized to ensure that effective features can still be extracted stably under scenarios such as signal strength attenuation and sudden interference.

[0050] In a preferred embodiment of the present invention, two spatial orientation vectors are determined based on the signal reference point; the phase difference between the two spatial orientation vectors is calculated according to the spatiotemporal signal sample set to delineate a signal feature domain, including:

[0051] Step S104: Based on the signal reference point, determine the main direction vector and generate the first spatial orientation vector through orthogonal decomposition. Specifically, this may include: based on the signal reference point, including the two-dimensional coordinate point of the core area centroid frequency of 10kHz and the main time coordinate at the 15th second, when determining the main direction vector, considering the propagation characteristics of ship low-frequency communication, the signal is mainly transmitted along the straight propagation path between the ship and the ground base station. The specific operation is as follows:

[0052] The main direction vector is defined by taking the signal reference point as the starting point and performing spatial calculations on the ship's GPS positioning information and the fixed coordinates of the ground base station (pre-set in the ship navigation system) to obtain the azimuth angle (such as the true north deflection angle calculated by the difference between latitude and longitude) of the ship's current position pointing to the base station. The direction of the straight line corresponding to this azimuth angle is defined as the main direction of signal propagation, forming the main direction vector. The vector direction is along the main path of the communication link, and the vector length is related to the relative proportion of the signal propagation distance.

[0053] The first spatial orientation vector is generated by orthogonal decomposition. In the ship's three-dimensional coordinate system (with the ship's center as the origin, the bow direction as the X-axis, the right side of the hull as the Y-axis, and the vertical direction upward as the Z-axis), the main direction vector is orthogonally decomposed, and the component along the X-axis (the bow points towards the base station, which has the highest degree of coincidence with the main direction) is extracted as the first spatial orientation vector. This vector specifically represents the energy transmission path of the signal in the main propagation direction. Its starting point is the signal reference point, and its direction is along the ship's X-axis towards the base station. The vector parameters include the direction angle (the angle with true north) and the propagation delay characteristics of the signal in this direction (calculated based on the reference point time coordinates).

[0054] Step S105, based on the first spatial azimuth vector, determines a second spatial azimuth vector orthogonal to it. Specifically, this may include: when determining the second spatial azimuth vector based on the first spatial azimuth vector, i.e., the main propagation direction along the ship's X-axis, according to the definition of an orthogonal direction, i.e., the direction perpendicular to the first vector in space, and considering the characteristics of lateral interference in ship communication (such as low-frequency signal interference from approaching ships), the specific operation is as follows:

[0055] Orthogonal direction determination: In the ship's three-dimensional coordinate system, the first vector is along the X-axis (forward and backward direction), and the direction orthogonal to it is the Y-axis (left and right direction, perpendicular to the bow direction). This direction is easily affected by the ship's rolling caused by crosswinds and waves, and may also be subject to lateral electromagnetic interference. A second spatial orientation vector is generated. Starting from the signal reference point, the second spatial orientation vector is defined along the positive Y-axis of the ship (right side of the hull). Its direction is perpendicular to the first vector (X-axis), and its length maintains the same relative proportion as the first vector. This vector is used to capture signal change characteristics perpendicular to the main propagation direction, especially for interference signals from the side, such as low-frequency communication signals from other ships.

[0056] Step S106: Based on the spatiotemporal signal sample set, calculate the signal phase characteristics along the first and second spatial azimuth vectors respectively, and obtain the phase difference between the two spatial azimuth vectors through differential operation. Specifically, this may include: when calculating the signal phase characteristics along the two spatial azimuth vectors based on the spatiotemporal signal sample set, considering the phase change pattern of the ship signal in different directions, i.e., the phase is stable in the main direction, but is easily affected by swaying interference in the lateral direction, causing phase fluctuations. The specific operation is as follows:

[0057] For the first spatial orientation vector (X-axis main direction), signal segments associated with this direction (i.e., effective command signals transmitted along the main propagation path) are selected from the spatiotemporal signal sample set. The instantaneous phase value corresponding to the 10kHz centroid frequency within each time window is extracted (analyzed from the time domain signal through Hilbert transform) to form the main direction phase sequence (containing continuous phase change data over time, such as the phase angle per millisecond within the 10th-20th second). For the second spatial orientation vector (Y-axis horizontal direction), signal components perpendicular to the main direction are extracted from the spatiotemporal sample set (such as those caused by the ship's swaying). The signal reflection component and the interference signal from the side are analyzed. Similarly, the instantaneous phase value at the 10kHz frequency is analyzed to form a transverse phase sequence (including phase fluctuation data caused by transverse interference or ship motion). The phase sequences of the two vector directions are calculated by time-by-time difference, that is, the phase value of the main direction at the same time point is subtracted from the transverse phase value to obtain the phase difference sequence. This difference reflects the phase difference between the main propagation direction and the transverse direction of the signal (in normal communication, the phase of the main direction is stable, the transverse phase fluctuation is small, and the difference is concentrated in the range of ±5°; if there is strong lateral interference, the difference will deviate significantly from this range).

[0058] Step S107, based on the phase difference value, divide the signal characteristic interval and determine the signal characteristic domain. Specifically, this may include: when dividing the signal characteristic interval and determining the signal characteristic domain based on the phase difference value, combining the phase stability threshold during normal ship communication, i.e., obtained through statistical analysis of historical interference-free navigation data, as follows:

[0059] The phase difference sequence during historical normal communication periods is statistically analyzed, and its mean (e.g., 2°) and standard deviation (e.g., 3°) are calculated. The effective phase difference interval is defined as the mean ± 2 times the standard deviation, i.e., 2° ± 6°, ranging from -4° to 8°. The phase difference values ​​within this interval correspond to the effective signal propagation characteristics. Intervals outside this range, such as <-4° or >8°, are determined to be invalid intervals due to interference or signal distortion. The spatiotemporal range (including the corresponding time window and frequency range) where the phase difference values ​​fall within the effective interval is defined as the effective subdomain of the signal characteristic domain. The signal within this subdomain contains stable command and navigation components. The spatiotemporal range where the phase difference values ​​fall within the invalid interval is defined as the invalid subdomain of the signal characteristic domain. The signal within this subdomain is significantly affected by interference or ship motion. The final signal characteristic domain encompasses both the effective and invalid subdomains, fully reflecting the phase characteristic distribution of the signal in different spatial directions.

[0060] In this embodiment of the invention, spatial azimuth vectors of the main direction and orthogonal direction are constructed with the signal reference point as the origin. This accurately maps the main propagation path and lateral interference direction of the ship and the base station, extending signal analysis from a single time / frequency domain to a spatial dimension. This better reflects the actual scenario in long-range maritime communication where signals propagate along specific paths and are susceptible to lateral interference. By calculating the phase characteristics and difference of the two orthogonal vectors, the phase difference in the spatial direction is transformed into a quantifiable index. During normal communication, the main direction phase is stable, the lateral phase fluctuation is small, and the phase difference is concentrated in the effective range. When interference occurs, the difference value deviates significantly, enabling the detection of interference. Accurate differentiation between effective signals and lateral interference; dividing effective and invalid subdomains based on phase difference values, clearly defining the spatial distribution range of the signal characteristic domain, with the effective subdomain corresponding to the stable navigation and command signal propagation area, and the invalid subdomain marking the interference or distortion area, improving the targeting of anti-interference control; through spatial vector orthogonal decomposition and phase difference analysis, directional identification of interference in different directions is achieved, avoiding the limitation of traditional single-dimensional analysis in distinguishing spatial interference, enabling low-frequency communication systems to more accurately adapt to complex scenarios such as main direction signal attenuation and sudden lateral interference during ship navigation, and improving overall anti-interference capability.

[0061] In a preferred embodiment of the present invention, a distributed detection mechanism is used to determine three signal monitoring nodes, which are respectively located in the effective and invalid regions of the signal feature domain, including:

[0062] Step S108: Based on the spatial distribution characteristics of the signal feature domain, the feature domain is divided into multiple detection units, and the energy density value of each unit is calculated. Specifically, this includes: when dividing the detection units based on the spatial distribution characteristics of the signal feature domain (including effective and invalid subdomains, extending along the main X-axis and distributed laterally along the Y-axis in the ship coordinate system), the physical layout of the ship's communication equipment is considered (e.g., the low-frequency receiving antenna is located at the bow, the signal processing unit is located in the middle, and interference sources are mostly concentrated in the stern engine room). The specific operation is as follows:

[0063] In the ship's three-dimensional coordinate system, the signal feature domain is divided into longitudinal units along the X-axis (bow to stern) at 5-meter intervals, and transverse units along the Y-axis (left and right sides of the ship) at 2-meter intervals, forming a 5m×2m rectangular detection unit grid (covering the core communication path from the bow receiving antenna to the ship's processing unit, as well as the possible interference area at the stern). Each unit corresponds to a specific physical space region (e.g., X = 10-15m, Y = 0-2m corresponds to a 5-meter range on the right side of the bow). For each detection unit, signal segments within the corresponding spatial range of the spatiotemporal signal sample set are extracted (the unit to which the signal belongs is located by the signal arrival time difference). The energy density calculation method of step S1023 is reused to calculate the total energy density value (in V) of the 10kHz command signal band and the 14-16kHz navigation signal band within the unit. 2 · s / m2 (After normalization based on the unit area), the energy density distribution of each unit is obtained (the unit energy density within the effective subdomain is typically higher than 10). -4 V 2· s / m 2 The energy density of the cells in the invalid subdomain and interference region is less than 5 × 10⁻⁶. -5 V 2· s / m 2 ).

[0064] Step S109: Based on the energy density values ​​of each unit, identify the effective and ineffective intervals of the signal feature domain and mark the boundaries of each interval. Specifically, this includes: when identifying the effective and ineffective intervals and marking the boundaries based on the energy density values ​​of each unit, combining the energy threshold during normal ship communication (through historical data statistics, the energy density of the effective signal must be consistently higher than 8 × 10⁻⁶). -5 V 2· s / m 2 The specific steps are as follows:

[0065] By traversing all detection units, the energy density value is consistently higher than 8 × 10⁻⁶. -5 V 2· s / m 2 Furthermore, continuous clusters of cells whose phase difference falls within the effective range (-4° to 8°) are determined to be within the effective range of the signal characteristic domain (mainly concentrated in the range of X = 5-25m and Y = -3 to 3m from the bow receiving antenna to the midship processing unit, corresponding to the core region of the main propagation path); energy density values ​​below 5×10 -5 V 2· s / m 2 Clusters of cells whose phase difference exceeds the effective range are identified as invalid ranges (mainly distributed in the X=30-40m range near the stern engine room and the edge areas of Y>±5m on both sides of the hull, which are susceptible to mechanical vibration and electromagnetic interference). In the ship coordinate system, the boundary curves of the effective range are fitted by the cell coordinates (e.g., X=5m is the front boundary, X=25m is the rear boundary, and Y=±3m is the left and right boundary), and the boundary coordinates are recorded in the physical space by hull markings (e.g., marking the effective range line on the deck) and the system database to clarify the spatial boundary between the effective and invalid ranges.

[0066] Step S110: Based on the distribution characteristics of the effective and invalid intervals, select two monitoring node locations within the effective interval and one monitoring node location outside the invalid interval. Specifically, when selecting three monitoring node locations based on the distribution characteristics of the effective and invalid intervals, the reliability requirements of ship communication (monitoring the main signal, backup path, and interference reference) are considered. The specific operation is as follows:

[0067] Select two nodes within the valid interval:

[0068] The first node (the main monitoring node) is selected at the core location of the effective range (X=15m, Y=0m, i.e., at the centerline of the ship, corresponding to the spatial projection of the signal reference point), where the signal energy density is highest (typically >1.5×10). -4 V 2· s / m 2 It also has the best phase stability and is used for real-time monitoring of command and navigation signals on the main propagation path.

[0069] The second node (backup monitoring node) is selected at the edge of the effective range (X=25m, Y=2m, near the right boundary of the rear end of the effective range). The signal energy density at this location is slightly lower (approximately 1×10⁻⁶). -4 V 2· s / m 2 However, it is still within the effective range and is used to monitor the attenuation trend of the main signal and the stability of the edge area to avoid monitoring failure due to a single node failure.

[0070] A node is selected outside the invalid region, at a location far from the invalid region and with extremely low interference (X=5m, Y=-6m, 6 meters to the port side of the bow, far from the stern engine room and the interference zone at the edge of the hull). Although the energy density at this location is low (approximately 6×10⁻⁶), the energy density is still low (approximately 6×10⁻⁶). - 5 V 2· s / m 2 However, the phase fluctuation is extremely small (difference < ±2°), serving as a reference node to distinguish between real interference and signal changes caused by the ship's own motion.

[0071] Step S111: Based on the selected locations of the three monitoring nodes, perform spatiotemporal synchronization calibration on each node and verify whether the node locations meet the distribution requirements of the effective and invalid intervals to obtain the final three signal monitoring nodes, of which two nodes are located within the effective interval of the signal characteristic domain and one node is located outside the invalid interval of the signal characteristic domain. Specifically, when performing spatiotemporal synchronization calibration and verification based on the selected locations of the three monitoring nodes, the ship's GPS timing and inertial navigation system are combined (to ensure time and position accuracy). The specific operations are as follows:

[0072] The ship obtains standard UTC time through its GPS receiver module and sends synchronization pulses to the clock modules of the three nodes to keep the timestamp error of each node within 1ms (ensuring the consistency of phase difference calculation). The ship's inertial navigation system (INS) measures the actual physical coordinates of the three nodes and compares them with the preset positions. By fine-tuning the installation positions of the points (such as adjusting the antenna bracket), the deviation between the actual coordinates and the planned coordinates is kept within 0.5 meters (ensuring the accuracy of the spatial orientation vector).

[0073] For two nodes within the effective interval, signals are continuously acquired for 10 minutes to verify whether their phase difference falls within the range of -4° to 8° for more than 90% of the time, and whether the energy density is consistently higher than 8×10⁻⁶. -5 V 2· s / m 2 The system confirms that the reference node is located within the valid range. For reference nodes outside the invalid range, it verifies that more than 95% of their phase difference falls within the ±2° range and that there are no overlapping units in the invalid range in their area. If all three nodes pass the verification, they are designated as official signal monitoring nodes and marked as the main monitoring node, backup monitoring node, and reference reference node, respectively, and connected to the ship's communication anti-interference control system. If a node fails the verification, units in adjacent locations are re-selected for testing until the requirements are met.

[0074] In this embodiment of the invention, by spatially gridding the detection units and calculating the energy density, and combining the phase difference interval, the effective signal interval (core area of ​​the main propagation path) and the invalid interval (interference concentration area) are clearly defined, achieving precise spatial differentiation between signals and interference, and providing a clear target for anti-interference strategies. Within the effective interval, primary and backup monitoring nodes are set up to cover the core and edge areas, ensuring both the real-time monitoring accuracy of the main signal and avoiding the risk of single-node failure through backup nodes. Reference nodes outside the invalid interval provide low-interference references, effectively distinguishing between true interference and ship motion noise, and improving the accuracy of interference judgment. GPS timing and the inertial navigation system (INS) ensure that the node timestamp error is ≤1ms and the position deviation is ≤0.5 meters, guaranteeing the spatiotemporal consistency of phase difference calculation and energy density analysis.

[0075] In a preferred embodiment of the present invention, a closed-loop signal phase path is constructed based on the phase correlation of each signal monitoring node in the time domain; based on the closed-loop signal phase path, a signal anti-interference compensation coefficient is generated, including:

[0076] Step S112: Based on the spatiotemporal location information of the three signal monitoring nodes, signal phase data of each node at the same time interval are collected. Specifically, this includes: using the standard UTC time of the ship's GPS timing as a reference, a unified collection period is set, and a typical cycle of the ground base station sending navigation correction commands is selected, such as 30 consecutive seconds, to cover a complete command signal transmission, so that each node collects data within a completely synchronized time window; for the main monitoring node (X=15m, Y=0m) and the backup monitoring node (X=25m, Y=2m), the instantaneous phase data of the 10kHz command signal and the 14-16kHz navigation signal are collected, since these two nodes are located in the effective range, and the phase characteristics of the main propagation path need to be captured; the reference node (X=5m, Y=-6m) synchronously collects phase data of the same frequency band, but its data mainly reflects the phase reference under low interference environment; during the collection process, each node records the phase value at a sampling frequency of 1kHz, and synchronously associates its own spatiotemporal coordinates (updated in real time through INS) and timestamp (error controlled within 1ms), finally forming three sets of datasets containing three-dimensional information of time, position and phase.

[0077] Step S113: Based on the signal phase data, calculate the relative phase difference between each monitoring node and construct a complete phase distribution map. Specifically, this includes: based on the collected phase data, focusing on the correlation characteristics between phase stability and interference in ship communication, calculating the difference and constructing the distribution map. The specific operations are as follows:

[0078] For the three sets of node data, the relative phase difference between the three pairs of nodes is calculated at the same time point: the difference between the main monitoring node and the backup monitoring node, which reflects the spatial consistency of the signal phase within the effective interval; the difference between the main monitoring node and the reference reference node, which reflects the phase deviation between the main signal and the low-interference reference; and the difference between the backup monitoring node and the reference reference node, which reflects the phase difference between the edge of the effective interval and the reference.

[0079] During the calculation, the phase data of the reference node is used as the zero-interference reference template. If the phase of the main monitoring node is φ1 and the phase of the reference node is φ0 at a certain moment, the relative difference is φ1-φ0, thereby eliminating the synchronous influence of the ship's common motion (such as overall sway) on the phase.

[0080] Subsequently, phase difference curves for three pairs of nodes are plotted with time on the horizontal axis and phase difference on the vertical axis to form a complete phase distribution map. The effective phase range (-4° to 8°) and the low-interference reference range ±2° are marked in the map to intuitively show the phase fluctuation pattern between different nodes. During normal communication, the node difference within the effective range is concentrated in the effective range, and the difference with the reference node fluctuates less. When there is interference, the difference will deviate significantly from the range.

[0081] Step S114: Based on the phase distribution map, connect the phase feature points of each monitoring node to form a closed path for the signal phase. Specifically, this includes: based on the characteristic patterns of the phase distribution map and combined with the spatial location of the nodes, constructing a closed path that reflects the phase continuity. The specific operation is as follows:

[0082] First, key phase feature points are selected in the phase distribution map: the phase stability point of the main monitoring node (the moment when the phase difference fluctuation of 10 consecutive sampling points is <1°), the phase edge point of the backup monitoring node (the moment when the difference is close to the boundary of the effective interval), and the phase reference point of the reference node (the moment when the difference is stable within ±1°). This ensures that the feature points cover three scenarios: signal stability, edge fluctuation, and reference reference. Then, in three-dimensional space (X / Y axes are the node position coordinates, Z axis is the phase difference), the stability point of the main monitoring node, the edge point of the backup monitoring node, and the reference point of the reference node are connected in sequence to form a triangular closed path. The path connection must meet the principle of spatiotemporal continuity: the time interval between adjacent feature points does not exceed 5 seconds (to ensure the continuity of phase change), and the path extends in space along a straight line along the actual physical location of the node. This ensures that the path can reflect the phase characteristics within the effective interval and anchor a low-interference reference through the reference node, completely covering the phase change trajectory of the signal from the main propagation path to the interference reference.

[0083] Step S115: Based on the closed path of the signal phase, calculate the curvature characteristics of each point on the path, and obtain the overall curvature parameter of the path through integration. Specifically, this includes: quantifying the severity of signal interference by analyzing the curvature of the closed path. The specific operations are as follows:

[0084] For the constructed phase closed path, discrete points on the path are taken at 1-meter spatial intervals, and the curvature value of each point is calculated. The greater the curvature, the more drastic the phase change at that location, and the stronger the interference. During the calculation, special attention is paid to line segments in the path that deviate from the effective phase interval: if the phase difference of a certain segment of the path is consistently >8° or <-4°, the curvature value of that segment is weighted and amplified according to the actual fluctuation amplitude to highlight the influence of the interference area. Subsequently, the curvature values ​​of all points on the path are integrated, and the integration interval covers the entire spatial length of the closed path (from the main monitoring node to the backup node, then to the reference node, and finally back to the main node). The integration result is the overall curvature parameter of the path. The larger the value of this parameter, the more significant the cumulative impact of interference on the signal phase during spatial propagation. Under normal conditions without interference, the curvature parameter is usually <5° / m; when there is strong lateral interference or stern mechanical noise, the parameter will increase to over 10° / m.

[0085] Step S116: Based on the overall curvature parameters of the path, generate signal anti-interference compensation coefficients. Specifically, this includes generating anti-interference compensation coefficients adapted to the ship scenario based on the correlation between the overall curvature parameters and the degree of interference. The specific operations are as follows:

[0086] The baseline value of the compensation coefficient is set to 1 (no compensation is needed when there is no interference). A positive correlation mapping relationship between the curvature parameter and the compensation coefficient is established: when the curvature parameter is <5° / m, the compensation coefficient is 1.0-1.2, only fine-tuning is performed for slight phase fluctuations; when the parameter is between 5° / m and 10° / m, the coefficient is 1.2-1.5, enhancing phase correction for moderate interference; when the parameter is >10° / m, the coefficient is 1.5-2.0, providing deep compensation for strong interference. Simultaneously, compensation weights are allocated differently based on node characteristics. The primary monitoring node, with the highest energy density, has a compensation coefficient applied at 1.0 times the baseline value; the backup monitoring node, located in the edge region, has a coefficient amplified to 1.1 times, strengthening its anti-interference capability against edge signals. The phase data of the reference node is used as the compensation calibration benchmark, not directly involved in compensation but used to verify the compensation effect. The final generated compensation coefficient must ensure that, after correction, more than 90% of the phase difference between the primary and backup nodes returns to the effective range, achieving accurate cancellation of complex interference in the open sea.

[0087] In this embodiment of the invention, by synchronously collecting phase data from three nodes at the same time period and establishing phase correlation with spatiotemporal location information, the phase differences of main signal propagation, edge attenuation, and interference noise in ship communication can be effectively distinguished. By calculating the relative phase difference between nodes and constructing a distribution map, the phase fluctuation pattern can be presented intuitively. Based on the closed path formed by feature points, combined with curvature features and overall curvature parameters, the abstract phase interference is transformed into a quantifiable physical index, accurately reflecting the interference intensity and spatial distribution characteristics. The anti-interference compensation coefficient generated based on the curvature parameters can dynamically adjust the compensation intensity according to the degree of interference (slight, moderate, strong interference), and differentiate the weights for the different characteristics of the main monitoring node and the backup node, ensuring that the signal phase within the effective range stably returns to the effective range after correction, significantly improving the signal anti-interference capability.

[0088] In a preferred embodiment of the present invention, step S2, based on the signal anti-interference compensation coefficient, preprocesses the low-frequency communication signal using a multi-mode redundancy architecture to generate a signal stream with time-frequency redundancy characteristics, including:

[0089] Step S200 involves performing preliminary filtering on the original low-frequency communication signal based on the signal anti-interference compensation coefficient to obtain the filtered signal. Specifically, this includes: based on the signal anti-interference compensation coefficient generated in step S1 (reflecting the intensity and type of interference in the current offshore environment), targeted preliminary filtering is performed on the original low-frequency communication signal of the ship (including navigation correction commands, course turning commands, ship speed / fuel consumption, and other status data). The specific operations are as follows:

[0090] First, the interference level is determined based on the compensation coefficient: if the coefficient is <1.2, it is a minor interference, and a conventional hardware-level LC bandpass filter circuit is activated, with the center frequency locked to the nominal communication frequency of the ground base station, such as 10kHz command signal and 15kHz navigation signal, and the bandwidth set to 2kHz, filtering out electromagnetic noise from sea waves in non-target frequency bands; if the coefficient is between 1.2 and 1.5, it is a moderate interference, and software adaptive filtering is superimposed on the hardware filtering, using the low-interference signal of the reference node in step S1 as a template to dynamically eliminate sudden noise, such as instantaneous amplitude jumps caused by lightning electromagnetic pulses; if the coefficient is >1.5, it is a strong interference, and the filtering strength is further enhanced, using digital notch filtering to remove specific frequency interference introduced by the stern engine room, such as 50Hz power frequency interference and its harmonics, while retaining the effective modulation components in the signal, such as the pulse amplitude changes of the command signal and the continuous phase modulation characteristics of the navigation signal; finally, the filtered signal is obtained, with its noise power reduced by more than 30% compared to the original signal, and the pulse waveforms of key commands and the phase changes of navigation signals are completely preserved.

[0091] Step S201: Based on the filtered signal, the filtered signal is distributed to three parallel processing channels. Each channel uses a different signal modulation method for processing to obtain the output signals of the three parallel processing channels. Specifically, based on the filtered signal and according to the multi-mode redundancy architecture design, the signal is evenly distributed to three independent parallel processing channels. Each channel uses a differentiated modulation method for different signal types in ship communication to ensure that a single type of interference cannot simultaneously affect all channels. The specific operation is as follows:

[0092] Channel 1 (Amplitude Shift Keying Channel) is used for ship status reporting signals, such as speed and fuel consumption data, which are sensitive to amplitude changes. It adopts ASK (Amplitude Shift Keying) modulation mode, which uses the high and low amplitude changes of the signal to represent data 0 and 1. The modulation rate is adapted to the low frequency communication requirements, such as 1200bps, to ensure that the amplitude characteristics of the status data are stable and recognizable.

[0093] Channel 2 (Frequency Shift Keying Channel) is used for navigation correction commands, such as latitude and longitude correction information. It needs to resist frequency drift interference and adopts FSK (Frequency Shift Keying) modulation mode. It uses two different low-frequency carrier frequencies (such as 9kHz representing 0 and 11kHz representing 1) to transmit data. The carrier frequency interval is 2kHz to avoid frequency interference in the open sea and ensure that the frequency characteristics of navigation commands are clearly distinguishable.

[0094] Channel 3 (Phase Shift Keying Channel) is designed for high-priority route commands, such as emergency turn commands, which require resistance to phase jitter interference. It adopts PSK (Phase Shift Keying) modulation, which uses the 0° or 180° change of the carrier phase to represent the data. The stability of the phase change is used to counteract the phase fluctuations caused by ship vibration, ensuring that the phase characteristics of the emergency command are accurate. After each channel independently completes signal modulation, it outputs parallel signals containing the original information but with different modulation characteristics.

[0095] Step S202 involves extracting the time-domain and frequency-domain features of each channel's output signal based on the output signals of the three parallel processing channels, and generating composite signal features. Specifically, this includes extracting time-domain and frequency-domain features from the output signals of the three parallel channels, and generating composite features that comprehensively reflect the signal characteristics through feature fusion, providing a basis for subsequent consistency detection. The specific operations are as follows:

[0096] Time-domain feature extraction captures key time-domain parameters for the output signal of each channel. Channel 1 (ASK) focuses on extracting the amplitude peak (amplitude threshold to distinguish between 0 and 1) and pulse duration (bit width of status data); Channel 2 (FSK) focuses on extracting the frequency transition moment (start / end flag of navigation command) and frequency hold duration (duration of each data bit); Channel 3 (PSK) focuses on extracting the phase transition angle (phase difference of 0° or 180°) and phase stabilization duration (phase continuity segment of command signal).

[0097] Frequency domain feature extraction involves analyzing the spectral distribution of each channel signal using Fast Fourier Transform. Channel 1 focuses on the main frequency amplitude (amplitude ratio of the 10kHz command signal) and spectral sidelobe suppression ratio (reducing spectral spread of amplitude modulation). Channel 2 focuses on the power ratio of the two carrier frequencies (power difference between 9kHz and 11kHz) and frequency bandwidth (ensuring the concentration of the navigation signal spectrum). Channel 3 focuses on the spectral peak position corresponding to phase modulation (phase modulation spectrum of the 15kHz navigation signal) and phase noise power (reflecting phase stability). The time domain features (9 items in total) and frequency domain features (9 items in total) of the three channels are integrated into a composite feature set containing 18 parameters. These parameters are recorded according to channel number, feature type, and feature value, forming a feature matrix that comprehensively reflects the signal quality and information integrity of each channel.

[0098] Step S203: Based on the composite signal characteristics, perform consistency detection on the signals of each channel and generate an intermediate signal stream with time-frequency redundancy characteristics. Specifically, this includes: performing consistency detection on the output signals of the three channels based on the composite signal characteristics, ensuring the accuracy of signal information through redundancy verification, and simultaneously constructing an intermediate signal stream with time-frequency redundancy characteristics. The specific operations are as follows:

[0099] First, establish consistency criteria: In the time domain, the pulse width deviation of key commands in the three channels must be ≤10%, such as the pulse duration of turning commands should be 50ms±5ms; in the frequency domain, the main frequency deviation must be ≤±0.3kHz, such as the main frequency of navigation signals should be within 15kHz±0.3kHz; in terms of information content, by comparing the demodulated binary data of each channel, such as the binary code corresponding to a 30° turn in the route command, ensure that the consistency of the core command code is ≥95%.

[0100] For signal segments that pass consistency testing, such as navigation correction data that is consistent across all three channels, the data is directly retained and repeatedly embedded in the time domain. For example, each segment of key data is transmitted twice consecutively to form time domain redundancy. For segments with slight deviations, such as a channel whose frequency domain characteristics deviate slightly but whose time domain characteristics are consistent, the signal from channel 2 (FSK channel, which has stronger anti-interference capabilities) is used as a reference for correction. The corrected signal is retained and the spectrum coverage is extended in the frequency domain. For example, weak repetitive signals are added at ±0.5kHz on both sides of the original main frequency to form frequency domain redundancy. For abnormal segments with large deviations, such as a channel whose signal is distorted by strong interference, the signal is directly removed and redundant information from the other two channels is used to fill the gap, ensuring that the intermediate signal stream does not lose key content. The final generated intermediate signal stream contains repetitive segments of key information in the time domain and covers multiple sub-frequency bands within the target frequency band in the frequency domain, possessing preliminary time-frequency redundancy characteristics.

[0101] Step S204 involves optimizing the intermediate signal stream using a signal anti-interference compensation coefficient to ultimately generate a signal stream with time-frequency redundancy characteristics. Specifically, this includes: performing targeted optimization of the intermediate signal stream based on the signal anti-interference compensation coefficient to enhance time-frequency redundancy characteristics and ensure the signal can withstand complex interference in the open sea. The specific operations are as follows:

[0102] The redundancy strength is dynamically adjusted based on the magnitude of the compensation coefficient: If the coefficient is <1.2, there is slight interference, and the optimization focus is on reducing redundancy to save bandwidth. It is sufficient to retain two repetitions in the time domain and two sub-band coverages in the frequency domain. At the same time, the signal amplitude deviation is corrected by the coefficient, such as enhancing the signal strength of the corresponding frequency band of the main monitoring node; If the coefficient is between 1.2 and 1.5, there is moderate interference, and the number of repetitions in the time domain is increased to three to ensure that there is still an effective backup after a certain interference. The frequency domain is expanded to cover three sub-bands, such as 10kHz, 10.5kHz, and 9.5kHz. The phase deviation is corrected by the coefficient, and the signal phase is adjusted to a stable range with reference to the phase closure path result in step S1; If the coefficient is >1.5, there is strong interference, and redundancy is further strengthened: the key instructions in the time domain are repeated four times, and the frequency domain covers five sub-bands, expanding to the 9-11kHz range. At the same time, the signal's anti-attenuation capability is enhanced by the coefficient, such as increasing the peak amplitude of the pulse signal to ensure that it can still be identified when the signal attenuates by 30%.

[0103] During the optimization process, the time-frequency redundancy effectiveness of the synchronous verification signal is verified. In the time domain, the original information can still be fully recovered after any consecutive loss of a repetitive signal. In the frequency domain, when a certain sub-frequency band is interfered with, the signals of other frequency bands can be used to complete the content through feature matching. The final generated signal stream has significant time-frequency redundancy characteristics while maintaining the integrity of the original communication data, which can effectively offset the problems of noise, attenuation, and sudden interference in the offshore environment.

[0104] In this embodiment of the invention, a filtering strategy based on dynamic adjustment of the anti-interference compensation coefficient is employed. Hardware filtering, adaptive software filtering, and digital notch filtering are activated for slight, moderate, and strong interference, respectively. This effectively filters out sea wave noise and electromagnetic interference while fully preserving the modulation characteristics of key signals such as navigation commands and status data. By using differentiated modulation methods (ASK, FSK, and PSK) for three parallel channels, the different resistances of these modulation methods to amplitude, frequency, and phase interference are utilized to prevent a single type of interference from affecting all channels simultaneously. This constructs an anti-interference barrier at the hardware architecture level, reducing the risk of overall signal distortion. The time domain (pulse width, phase transition) of each channel is extracted. A composite feature set is generated by combining the frequency domain (main frequency amplitude, spectrum distribution) features with the frequency domain features. The accuracy of the signal information is ensured through strict consistency detection (time domain deviation ≤10%, frequency domain deviation ≤±0.3kHz). Consistent signals are repeatedly embedded in the time domain (2-4 times) and the frequency domain spectrum is expanded (2-5 sub-bands). The redundancy strength is dynamically adjusted by combining the compensation coefficient to achieve the fault tolerance effect of recovering lost time domain segments and compensating for single frequency disturbances in the frequency domain. Abnormal signals are corrected based on channels with stronger anti-interference capabilities to ensure the integrity of the intermediate signal stream. The final generated signal stream can effectively offset the signal distortion caused by complex interference in the open sea, and improve the system's anti-interference self-healing capability.

[0105] In a preferred embodiment of the present invention, step S3, which involves encrypting a signal stream with time-frequency redundancy characteristics using a quantum key distribution protocol to generate a quantum encrypted signal stream, includes:

[0106] Step S300: Based on the signal stream with time-frequency redundancy characteristics, a quantum random number generator is used to generate a quantum true random number sequence as the initial encryption key. The quantum random number generator (hereinafter referred to as the quantum random number module) is the core device for generating the initial encryption key. Its deployment and initialization must be adapted to complex environments such as ship vibration and electromagnetic interference to ensure physical randomness and stability. Specifically, this includes:

[0107] The quantum random number module adopts an industrial-grade protective design, with overall dimensions of 30cm×20cm×10cm and a weight of no more than 2kg. It is compatible with the standard mounting rails of ship communication control cabinets. The deployment location follows the principle of being close to the signal source and far from the interference source. It is fixed in the upper area of ​​the ship's communication control cabinet, adjacent to the low-frequency signal processing unit, with a distance of no more than 50cm to shorten the physical transmission distance with the signal stream and reduce data latency. At the same time, it is at least 5m away from the stern engine room and at least 1.5m away from the ship's power supply module to reduce the impact of mechanical vibration and electromagnetic noise, such as 50Hz power frequency interference from the generator, on the core components of the module.

[0108] The module is installed using a shock-resistant bracket with a 3mm thick silicone damping pad embedded at the bottom. It is rigidly connected to the control cabinet chassis by four M6 stainless steel bolts to ensure that the module displacement does not exceed 0.5mm when the hull rolls at a pitch angle of no more than 15 degrees, thus preventing misalignment of the optical components. The module shell is made of 0.8mm thick galvanized steel sheet and is stamped as a whole. The key internal optical component, the single-photon source beam splitter, is covered with a copper mesh shield. An electromagnetic interference filter with a cutoff frequency of 1MHz is installed at the input end to block stray signals from entering the ship's electromagnetic environment.

[0109] The quantum random number module contains five core functional components, which are physically connected and pre-configured in sequence to ensure that the components work together. These components include:

[0110] The single-photon source is connected to the input of the beam splitter via an FC / APC fiber optic patch cord. The two outputs of the beam splitter are connected to two single-photon detectors via 2m long polarization-maintaining fibers. The detector output signals are connected to the signal amplification circuit via a 50Ω coaxial cable. The amplified digital signals are transmitted to the digital conversion unit via the LVDS interface. The digital conversion unit establishes data interaction with the low-frequency signal processing unit via the ship's local CAN bus with a transmission rate of 1Mbps to receive signal stream characteristic parameters.

[0111] The single-photon source has a fixed emission wavelength of 850nm to adapt to the atmospheric window in the open sea with an attenuation rate of no more than 0.2dB / km. The emission frequency is set to 1kHz, emitting one photon per millisecond to balance randomness and energy consumption. The transmitted or reflected light power is monitored in real time by a power meter, and mechanical fine-tuning is automatically triggered when the deviation exceeds the limit. The detector trigger threshold is set to 30mV to filter electromagnetic noise pulses below this value, and the response time is locked at 1ns.

[0112] To address the voltage fluctuations in ship power supply systems, such as diesel generators, which typically range from AC220V±10%, multi-stage voltage stabilization is employed to ensure stable power supply to the modules. Specifically, this includes:

[0113] The main power supply link is designed with DC 12V power supply, and the input is connected to the ship's DC regulated power supply, which has a wide input range of AC85-264V and overvoltage or overcurrent protection functions. The output of the regulated power supply is connected in series with a 100μF electrolytic capacitor and a 10μH inductor to form a π-type filter circuit, which controls the output ripple voltage to no more than 5mV peak-to-peak value, so as to avoid voltage fluctuations causing instability in the power of the single photon source. Redundant power supply configuration is provided with a spare DC 12V lithium battery pack with a capacity of 10Ah, which is connected in parallel with the main power supply link through an automatic switching switch. When the main power supply is interrupted, such as by a ship's power grid failure, the switching time is no more than 10ms, ensuring that the module continues to work without interrupting the key generation process for 30 minutes.

[0114] After power-on, the quantum random number module automatically initiates a three-level self-test process when the standby current is no greater than 300mA and the operating current is no greater than 800mA. It only enters the working state after passing all tests, which specifically include:

[0115] Within 10 seconds of starting the Level 1 hardware self-test, the physical connection status of core components is checked, including the continuity of the fiber optic link, the input optical power measured by the optical power meter (not less than -60dBm), the detector bias voltage (5V±0.1V), and the communication status of the digital conversion unit (CAN bus heartbeat packet response time not greater than 100ms). If a component is abnormal, the red fault light on the module panel will flash at a frequency of 2Hz, and a fault code such as E01, representing that the single photon source has not been started, will be sent to the ship monitoring terminal via the CAN bus.

[0116] Within 20 seconds after the Level 2 performance calibration self-test, key performance indicators are quantitatively calibrated, including single-photon source stability (power deviation not exceeding 5% in 1000 consecutive transmissions), beam splitter consistency (transmission and reflection count difference not exceeding 5% in 1000 consecutive photons), and detector dark count (dark count not exceeding 5 times in 1 minute when there is no photon input to avoid ambient light interference). If the calibration fails, the module automatically starts internal fine-tuning, such as adjusting the detector cooling temperature to -10℃±1℃ to reduce the dark count. If it still fails after 3 retries, a manual maintenance alarm is triggered.

[0117] Within 30 seconds after the Level 3 environment adaptation test self-check, the module simulates typical ship operating conditions such as generator startup and hull vibration. The initial randomness of the output random number is tested by a simplified frequency test. The deviation of the 01 ratio in a 10,000-bit sequence is no more than 2%. After passing the test, the green running light on the module panel stays on, and a ready signal containing the module ID and current status code is sent to the low-frequency signal processing unit. The module then waits for the trigger command of the time-frequency redundant signal stream to enter the random number generation standby state.

[0118] The quantum random number module employs a dual mechanism of single-photon path selection randomness and quantum vacuum noise capture to generate random numbers in parallel, ensuring stability under extreme environments. Specifically, this includes:

[0119] The single-photon path selection mechanism involves a single-photon source continuously emitting single photons at a frequency of 1 kHz. After entering a 50:50 beam splitter, the photons theoretically have a 50% probability of being transmitted to detector A and a 50% probability of being reflected to detector B. The detectors capture the photon arrival signal in real time. When detector A is triggered, it outputs a digital 0, and when detector B is triggered, it outputs a digital 1. If atmospheric scattering at a certain time results in a probability of no photons being detected that is no greater than 0.1%, the quantum vacuum noise mechanism is triggered to supplement the signal.

[0120] The quantum vacuum noise mechanism captures electromagnetic fluctuation noise in the quantum vacuum using a high-sensitivity microwave receiver. The noise signal is then converted into a binary sequence via analog-to-digital conversion, serving as a backup to the path selection mechanism and ensuring continuous generation of random numbers even without photon input. The original sequence merging and filtering process merges the original random sequences generated by the two mechanisms in real time and removes periodic interference caused by ship mechanical vibrations, such as the 10Hz hull swaying frequency component, through digital filtering, initially forming an irregular binary stream.

[0121] The quantum random number module establishes real-time data interaction with the ship's signal processing unit, dynamically associating characteristic parameters of the time-frequency redundant signal stream to ensure accurate adaptation between key generation and signal encryption requirements. Specifically, this includes:

[0122] Time-domain periodic correlation: The signal processing unit extracts the time-domain repetition period of key instructions from the time-frequency redundant signal stream. For example, the ground base station sends a navigation correction instruction every 30 seconds. This instruction is repeated 4 times in the time domain in the signal stream. The 30-second period and the rising edge of the first repetition pulse are used as the key generation trigger signal to ensure that a fresh key is ready before each key instruction is transmitted.

[0123] Frequency domain main frequency correlation: through spectrum analysis, the core frequency band of the signal flow is determined to be 10kHz for command signals and 15kHz for navigation signals. The module uses the frequency domain main frequency value of 10kHz as the benchmark parameter for random number generation rate. When the power of 10kHz signal is detected to be greater than 60%, indicating that the current stage is a command-intensive transmission stage, the generation rate is increased to 2048 bits per second. When the proportion of 15kHz navigation signal is higher, the rate is maintained at 1024 bits per second to avoid resource waste.

[0124] Data volume matching control: The signal processing unit counts the instantaneous data volume of the time-frequency redundant signal stream in real time, such as counting the number of KB currently transmitted every 100ms. The module dynamically adjusts the generation speed according to the ratio of 128-bit random numbers corresponding to each KB of signal stream. For example, when the signal stream transmits an emergency turnaround command with a single command data volume of 2KB, the module generates a 256-bit random number within 200ms to ensure sufficient key length.

[0125] The generated random number sequence must meet both encryption and verification requirements, with strict control over the key length and segmentation identification, specifically including:

[0126] Key length calculation: The initial key length is calculated based on the total data volume of the signal stream. If the total data volume of the time-frequency redundant signal stream is NKB, including redundant data with time-domain repetition and frequency-domain expansion, then the initial encryption key sequence length is set to 1.5 × N × 128 bits. Of these, 1 × N × 128 bits are used for core encryption, and 0.5 × N × 128 bits are reserved as redundancy check bits for error correction in subsequent key distribution. For example, a signal stream with a total data volume of 10KB corresponds to an initial key of 1.5 × 10 × 128 = 1920 bits, including 1280 encryption bits and 640 check bits. Sequence segment identification: The generated key sequence is divided into encryption segments and check segments. Each segment is prefixed with an 8-bit identifier: 00000001 represents the encryption segment, and 00000010 represents the check segment, facilitating identification and retrieval in subsequent steps.

[0127] To ensure the initial key possesses quantum-level unpredictability, a rigorous randomness detection process is used to verify its validity, specifically including:

[0128] The key sequence undergoes comprehensive testing, focusing on 15 core indicators, including frequency testing (binary 0 / 1 ratio deviation not exceeding 1%), run-length testing (no obvious pattern in consecutive 0 / 1 lengths), and linear complexity testing (failure to predict using short linear feedback shift registers). If any indicator fails during testing, such as a frequency deviation exceeding 1%, the module is triggered to regenerate that key segment, and the reason for failure is recorded (e.g., temporary detector malfunction, archived through the ship's local log system). Once all indicators pass, the initial encrypted key sequence is marked as valid and stored in the secure storage area of ​​the ship's encryption chip, supporting hardware encryption to prevent physical tampering, awaiting use in the next step of quantum key distribution, ensuring security throughout the entire process from generation to storage.

[0129] Step S301: Based on the initial encryption key, a shared quantum key is established between the two communicating parties using the BB84 quantum key distribution protocol. The ship terminal, acting as the quantum signal transmitter, converts the binary information of the initial encryption key into a single-photon polarization state and transmits it via a satellite relay channel. Specifically, this includes:

[0130] Extract a 2048-bit valid binary sequence from the initial encryption key generated in step S300, removing redundant check bits to obtain the original information to be encoded. Divide this binary sequence into groups of one bit each to match the single-photon emission rhythm. Use two sets of orthogonal polarization bases for quantum encoding: the first set is a horizontal-vertical base (Z-base), where horizontal polarization states encode binary 0 and vertical polarization states encode binary 1; the second set is a 45-degree-135-degree base (X-base), where 45-degree polarization states encode binary 0 and 135-degree polarization states encode binary 1. The two polarization bases are randomly switched at a frequency synchronized with the single-photon emission frequency of 1kHz. Modulation is performed through a single-photon source with a wavelength of 1550nm and a polarization modulator with a response time of no more than 5ns in the ship's terminal quantum emission module. The single-photon source operates at a frequency of 1kHz. A single photon is emitted, and the polarization modulator modulates the photon's polarization state according to the basis selection sequence generated in real time by the quantum random number module. Under the Z basis, 0 corresponds to the horizontal polarization state and 1 corresponds to the vertical polarization state; under the X basis, 0 corresponds to the 45-degree polarization state and 1 corresponds to the 135-degree polarization state, so that each photon carries 1 bit of key information. The modulated single photon is transmitted to a low-Earth orbit communication satellite via a shipborne quantum transmission antenna with a gain of 15dBi and a beamwidth of 3 degrees. The satellite's quantum repeater transponder, with a low-noise optical amplifier containing a noise figure of no more than 3dB, forwards the signal to a ground base station. The transmission timestamp of each photon is recorded in real time, accurate to 10ns, and associated with the ship's GPS time, polarization basis selection (Z basis 0, X basis 1), and the corresponding encoding bit (0 or 1). This forms a transmission log format of timestamp plus basis selection plus encoding bit, which is stored in the ship's secure memory.

[0131] The ground base station, acting as the receiver, measures the quantum signal and performs a basis comparison with the ship's terminal via a classical channel, specifically including:

[0132] The ground base station quantum receiving module, which includes a high-precision optical antenna with a 50cm aperture and a tracking accuracy of no more than 0.1 degrees, and a polarization meter that supports Z-base and X-base switching, performs reception. After the optical antenna captures the single-photon signal relayed by the satellite, the polarization meter randomly selects the measurement base at a frequency of 1kHz. The base selection sequence is generated by the local quantum random number generator of the base station. Under the Z-base, the horizontal polarization state is judged as 0 and the vertical polarization state is judged as 1; under the X-base, the 45-degree polarization state is judged as 0 and the 135-degree polarization state is judged as 1. The measurement timestamp, measurement base selection, and measurement results are recorded in real time to form a reception log.

[0133] The ground base station transmits base selection information to the ship terminal via the idle sub-band of the redundant 9-11kHz command signal of the ship's low-frequency communication at a transmission rate of 9600bps. The information includes the measurement timestamp and the corresponding base selection sequence, but does not include the measurement result. After receiving the information, the ship terminal aligns the transmission log and the reception log according to the timestamp. If the timestamp deviation is no more than 50ns, the photons are considered to be the same. The photon records that are consistent with the transmission base and the measurement base are selected, and their encoded bits are retained. Together with the measurement result, they form an original key string with a length of approximately 50% of the total number of transmitted photons. The initial sequence of 2048 bits corresponds to the extraction of approximately 1024 bits of the original key. If the base matching consistency rate is lower than 45% in a certain period, the retransmission mechanism is triggered to retransmit the quantum signal for that period.

[0134] The original key string undergoes error correction and privacy amplification processing to generate a shared quantum key consistent with both parties, specifically including:

[0135] Both parties divide the original key string into 32-bit groups, and calculate an 8-bit parity check code for each group, including the first 16 even parity bits and the last 16 odd parity bits. The ship terminal sends the verification information, including each group's check code, to the base station via the classic channel. The base station compares its local check code with the received check code, and locates the error bits for inconsistent groups using a binary search method. Both parties exchange error position indices (excluding key content) via the classic channel and correct them bit by bit. After error correction, 20% of the key segments are randomly selected for comparison. If the consistency rate is not less than 99.9%, it is confirmed that there are no residual errors. Both parties use the SHA-256 hash function to perform privacy amplification on approximately 960 bits of the error-corrected key string. The process of generating a 256-bit hash value from the input key string is irreversible and cannot be passed through. The original key is deduced from the hash value; during the amplification process, hash function parameters such as the initial vector are exchanged through a classical channel to ensure algorithm consistency; after privacy amplification, each party randomly selects a 32-bit key fragment and compares it through a classical channel. If they are completely identical, the shared key is confirmed to be valid; if they are inconsistent, error correction and amplification are re-executed; the verified 256-bit shared key is stored in the ship's onboard encryption chip by the ship terminal and in the base station's AES-256 encryption level security database by the ground base station. The keys of both parties are completely identical and possess quantum non-cloning property. Through the above steps, the ship terminal and the ground base station establish a secure shared quantum key in the complex channel environment of the open sea, effectively resisting eavesdropping and noise interference.

[0136] Step S302, based on the shared quantum key, performs layered encryption processing on the time-frequency redundant signal stream to generate a primary encrypted signal stream. Specifically, this includes dividing the time-frequency redundant signal stream into three levels according to information importance, with differentiated data redundancy characteristics for each level.

[0137] The core layer is defined as containing the highest level of safety data, such as emergency turning commands and navigation correction parameters; this layer is repeated 4 times in the time domain and covers 5 sub-frequency bands in the frequency domain. The critical layer is defined as containing important data such as route planning and speed adjustment commands; the number of times the time domain is repeated and the number of sub-frequency bands covered in the frequency domain in this layer are between those of the core layer and the ordinary layer. The ordinary layer is defined as containing routinely reported data such as ship fuel consumption and equipment status; this layer is repeated 2 times in the time domain and covers 2 sub-frequency bands in the frequency domain.

[0138] The shared quantum key is split according to a preset ratio and differentiated encryption algorithms are applied to signals at different levels, specifically including:

[0139] Subkey splitting involves dividing the shared quantum key into three subkeys: a 128-bit subkey for the core layer, a 64-bit subkey for the critical layer, and a 64-bit subkey for the ordinary layer. Core layer encryption uses the AES-256 encryption algorithm, encrypting the key instruction segments that repeat in the time domain segment by segment using the core layer subkey. Simultaneously, a key check bit is added to the frequency-extended sub-band signal. Critical layer encryption uses the AES-128 encryption algorithm, performing an XOR operation between the critical layer subkey and the main frequency amplitude of the signal's frequency domain characteristics to ensure unpredictable spectral characteristics after encryption. Ordinary layer encryption uses stream cipher encryption, performing a bit-modulo-2 addition operation between the ordinary layer subkey and the regular data that repeats in the time domain, preserving the basic redundancy characteristics of the signal.

[0140] The encrypted signals from each level are integrated according to their original timing sequence to form a primary encrypted signal stream, specifically including:

[0141] The encryption signals of the core layer, key layer, and ordinary layer are reassembled according to the original transmission sequence to maintain the logical continuity of the data. Each layer's encryption result header carries a layer identifier and encryption length information. The layer identifier is used to distinguish different encryption levels, and the encryption length information is used to locate data boundaries. By integrating and attaching the identifier, a primary encryption signal stream containing multiple layers of encryption features is formed, ensuring that the leakage of a single subkey does not affect the security of data at other layers.

[0142] Step S303, based on the primary encrypted signal stream, verifies the integrity of the encryption process using a quantum authentication protocol and generates a message authentication code using a quantum hash algorithm. Specifically, this includes: verifying the encryption integrity of the primary encrypted signal stream using an entangled-state-based quantum authentication protocol.

[0143] The ship terminal and the ground base station share a pair of entangled photons in advance and store them in a quantum memory. The ship terminal performs quantum encoding on the first bit of the core layer encryption of the primary encrypted signal stream, performs Bell state measurement on the encoding result with one photon in the entangled photon pair, and sends the measurement result to the ground base station through a classical channel. The ground base station performs synchronous measurement on the other photon in the entangled photon pair, compares its own measurement result with the measurement result sent by the ship terminal, and if the matching degree is not less than 99%, it confirms that the encryption process has not been eavesdropped on or tampered with and the verification is successful.

[0144] The verified primary encrypted signal stream is processed using quantum hashing to generate a message authentication code, specifically including:

[0145] Signal quantum state transformation: The time-domain sampling points and frequency-domain spectrum values ​​of the verified primary encrypted signal stream are transformed into quantum state vectors. The quantum state vectors are then processed by multi-qubit gate operations of quantum circuits, such as CNOT gates and Tooffoli gates, and compressed into a fixed-length 256-bit quantum hash value. The quantum hash value is used as a message authentication code and bound to the primary encrypted signal stream. Any tampering with the signal stream will cause a significant change in the message authentication code value, which cannot be forged.

[0146] Step S304, based on the verified encrypted signal stream and message authentication code, converts the verified encrypted signal stream into a quantum state signal, ultimately generating a quantum encrypted signal stream, specifically including:

[0147] The verified primary encrypted signal stream, i.e., the classical electrical signal, is converted into a quantum state signal, specifically including:

[0148] For time-domain repeating encrypted segments, photonic polarization state encoding is used to encrypt them so that 0 corresponds to a 0-degree polarization state and 1 corresponds to a 90-degree polarization state. For frequency-domain extended encrypted sub-bands, photonic phase encoding is used so that the frequency offset of the sub-band corresponds to a phase difference of 0 degrees or π, ensuring that the quantum state corresponds one-to-one with the original encrypted information and maintaining information integrity.

[0149] The quantum state signal is redundantly processed by combining time-frequency redundancy characteristics. Specifically, the redundancy is enhanced in the core layer, in which the core layer quantum state is repeatedly sent twice in the time domain with an interval of 10ms each time, and the quantum state characteristics are copied in adjacent sub-frequency bands in the frequency domain; the redundancy is maintained in the ordinary layer, in which the basic redundancy of the ordinary layer quantum state is retained, so as to ensure that even if some quantum states are lost due to interference during transmission, such as photon absorption caused by electromagnetic noise in the ocean, the original information can still be recovered through the redundant quantum states.

[0150] The quantum state signal and the message authentication code are integrated to form the final quantum encrypted signal stream. Specifically, the converted quantum state signal and the message authentication code are appended to the end of the signal in the form of quantum state tags for integration. The integrated signal is then output through the transmission module of the ship's low-frequency quantum communication terminal in the 3-30kHz low-frequency band. The output signal stream has both quantum-unbreakable security and time-frequency redundancy anti-interference capability, and can be stably transmitted through satellite or ground link.

[0151] In a preferred embodiment of the present invention, step S4, employing a blockchain-based identity authentication mechanism and dynamic access control strategy to securely manage the quantum encrypted signal stream, thereby achieving real-time protection and trusted access control of the communication link, includes:

[0152] Step S400: Based on the quantum encrypted signal stream, establish a distributed identity authentication system and generate digital identity credentials. The specific process is as follows: Collect basic information such as hardware identifiers, equipment models, communication module serial numbers, network addresses, and operating system versions from ship terminals and ground base stations. Standardize this information, remove redundant fields according to a preset data format, unify field lengths, and integrate them into a structured identity information dataset. Use a cryptographic hash algorithm to perform a one-way mapping operation on the identity information dataset to generate a fixed-length identity hash value. Generate a unique asymmetric public-private key pair for each communication node. The private key is securely stored by the node through a local encryption chip. The public key is associated with the identity hash value and written as a node identity record into the blockchain distributed ledger. Perform a digital signature operation on the identity information dataset based on the node's public key to generate a signature result. Associate and integrate the identity hash value, public key, signature result, signature timestamp, and characteristic identifiers of the quantum encrypted signal stream, such as signal frequency range and transmission period, to form a digital identity credential, ensuring a unique correspondence between node identity and quantum encrypted signal stream.

[0153] Step S401: Based on the digital identity credential, the legality and trustworthiness of the communication node are verified through a consensus mechanism to obtain a verified communication node. The specific process is as follows: When a communication node initiates an identity verification request, it broadcasts the digital identity credential to multiple verification nodes in the blockchain network. The verification nodes retrieve the public key and identity hash value stored in the blockchain ledger, compare the retrieved historical information with the public key and identity hash value in the current credential field by field, and verify the validity of the digital signature. The verification nodes calculate a trustworthiness score based on indicators such as the number of abnormal transmissions, the number of abnormal access attempts, and the signal transmission success rate in the node's historical communication records. The number of abnormal transmissions has a higher weight than the number of abnormal attempts. A practical Byzantine fault-tolerant consensus mechanism is used to collect the verification results and trustworthiness scores of each verification node. When more than two-thirds of the verification nodes return a score that is not lower than the preset trustworthiness threshold and the identity information is completely consistent, a consensus is reached, the communication node is determined to be a legitimate and trustworthy node, and the verification result is written into a new block of the blockchain to obtain a verified communication node.

[0154] Step S402: Based on the verified communication nodes, implement dynamic access control. Adjust access permissions according to real-time network status and security policies to obtain the adjusted access permissions. Specifically, this includes: real-time collection of network status indicators such as the current communication load of verified nodes, the transmission error rate of quantum encryption signal streams, data transmission delay, and electromagnetic interference intensity in the sea area where the nodes are located through the network monitoring module; quantifying and weighting each indicator according to a preset weight ratio to generate a real-time risk value; when the communication load is too high, the error rate exceeds the threshold, or the interference intensity is high, the risk value increases accordingly; and access is determined based on the node's trustworthiness score. Access levels are defined as follows: Core permissions correspond to a risk value below the first threshold and a score above the first scoring standard, allowing the transmission of all types of quantum encrypted signal streams; Key permissions correspond to a risk value between the first and second thresholds and a score above the second scoring standard, allowing the transmission of important instruction-type signal streams; and Ordinary permissions correspond to a risk value below the second threshold and a score not lower than the basic standard, allowing the transmission of only regular status-type signal streams. When the risk value exceeds the threshold corresponding to the current permission level, an automatic permission downgrade process is triggered. When the risk value falls back to a safe range and remains stable for a preset duration, an automatic permission recovery process is executed, resulting in the adjusted access permissions.

[0155] Step S403: Based on the adjusted access permissions, the access control list is updated in real time through the blockchain transaction record mechanism, and the access control policy is automatically executed using smart contracts. Specifically, this includes: encapsulating the adjusted access permission level, node unique identifier, permission validity period, and allowed signal types into a permission update transaction according to the blockchain transaction format; after the transaction is verified for legality by consensus nodes in the blockchain network, it is written into the blockchain to form an immutable permission change record; the access control list is updated synchronously in real time according to the permission records in the blockchain; the list clearly records the current permission level of each node, the allowed signal types, transmission time periods, data volume limits, etc.; and deploying smart contracts with preset access control rules in the blockchain. When a node initiates a quantum encrypted signal stream transmission request, the smart contract automatically retrieves the access control list to verify the node's current permissions. If the permissions match, signal transmission is allowed and the transmission start time is recorded; if the permissions do not match, transmission is rejected and an abnormal request event is recorded in the blockchain, thus realizing the automatic execution of the access control policy.

[0156] Step S404: Based on the executed access control policy, the entire transmission process of the quantum encrypted signal stream is monitored and audited to achieve real-time protection and trusted access control of the communication link. Specifically, this includes: deploying monitoring modules at the main transmission path nodes of the quantum encrypted signal stream to collect key information such as node identifiers, transmission time, signal type, data volume, permission verification results, and transmission status in real time, generating detailed transmission logs. The monitoring modules encrypt the transmission logs at fixed time intervals and upload them to the blockchain audit ledger to ensure that the log content is tamper-proof. Pre-set anomaly monitoring rules are implemented. When situations such as mismatch between transmission content and permission level, signal transmission interruption frequency exceeding a preset threshold, or abnormal changes in node identity information occur during transmission, real-time alarms are triggered. The audit system periodically extracts transmission logs from the blockchain audit ledger, cross-compares them with the access control list and smart contract execution records, and calculates indicators such as permission compliance rate, abnormal event occurrence rate, and signal integrity compliance rate, generating a compliance audit report to ensure that the transmission behavior of the quantum encrypted signal stream is consistent with the access control policy, thereby achieving real-time protection and trusted access control of the communication link.

[0157] In this embodiment of the invention, a unique digital identity credential is generated by integrating the hardware and network characteristics of communication nodes and associating it with quantum encrypted signal stream characteristics, thereby achieving a strong binding between node identity and signal and avoiding the risk of identity forgery or signal theft. Based on multi-node consensus verification of communication node credibility, and combined with quantitative scoring of historical communication records, it ensures that only legitimate and trustworthy nodes access the network, effectively resisting malicious node intrusion and reducing the risk of signal leakage or tampering caused by unauthorized access. Real-time monitoring of network load, interference intensity, and other status indicators, combined with node credibility, dynamically adjusts access permissions, ensuring priority transmission of core signals while preventing high-risk nodes from excessively consuming resources, thus improving the anti-interference flexibility of the communication link. The immutability of blockchain records permission changes, and smart contracts automatically execute access control policies, achieving full traceability and no human intervention in permission updates and execution, reducing human error vulnerabilities, and ensuring the consistency and security of policy execution.

[0158] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A low-frequency communication anti-interference control method based on a multi-layered defense system, characterized in that, The method includes: The process involves: acquiring low-frequency communication signal data and converting it into a set of spatiotemporal signal samples; determining a signal reference point based on the spatiotemporal signal samples; determining two spatial orientation vectors based on the signal reference point; calculating the phase difference between the two spatial orientation vectors based on the spatiotemporal signal samples to delineate a signal feature domain; and employing a distributed detection mechanism to determine three signal monitoring nodes, each located within the effective and ineffective regions of the signal feature domain. This includes: dividing the feature domain into multiple detection units based on its spatial distribution characteristics and calculating the energy density value of each unit; identifying the effective and ineffective regions of the signal feature domain based on the energy density values ​​of each unit and marking the boundaries of each region; selecting two monitoring node positions within the effective region and one monitoring node position outside the ineffective region based on the distribution characteristics of the effective and ineffective regions; and performing spatiotemporal synchronization calibration on each selected monitoring node position and verifying whether the node positions meet the distribution requirements of the effective and ineffective regions to obtain the final three signal monitoring nodes, where two nodes are located within the effective region of the signal feature domain and one node is located outside the ineffective region of the signal feature domain. A closed-loop signal phase path is constructed based on the phase correlation of each signal monitoring node in the time domain; a signal anti-interference compensation coefficient is generated based on the closed-loop signal phase path. Based on the signal anti-interference compensation coefficient, a multi-mode redundancy architecture is used to preprocess low-frequency communication signals to generate a signal stream with time-frequency redundancy characteristics. A quantum encrypted signal stream is generated by encrypting a signal stream with time-frequency redundancy characteristics using a quantum key distribution protocol. A blockchain-based identity authentication mechanism and dynamic access control strategy are used to securely manage quantum encrypted signal streams, enabling real-time protection and trusted access control of communication links.

2. The low-frequency communication anti-interference control method based on a multi-layered defense system according to claim 1, characterized in that, Collect low-frequency communication signal data and convert the low-frequency communication signal data into a set of spatiotemporal signal samples; Determining a signal reference point based on the spatiotemporal signal sample set includes: Low-frequency communication signal data is acquired in real time and preprocessed to obtain preprocessed signal data. The preprocessed signal data is decomposed into multi-resolution components using the Daubechies wavelet basis function to obtain wavelet coefficients at each scale. The wavelet coefficients are then filtered and synthesized into a time-domain signal. A sliding time window is used to divide the time-domain signal into equal-length time intervals, and frequency domain features are extracted from the time-domain signal in each time interval to obtain a spatiotemporal signal sample set containing signal amplitude, phase, and frequency feature parameters. Based on the spatiotemporal signal sample set, the signal energy distribution characteristics are extracted, and the region where the signal intensity is concentrated is determined by calculating the energy density of each frequency band. Based on the region where the signal strength is concentrated, the core region of the signal energy distribution is determined, and the centroid of the core region is calculated and determined as the signal reference point.

3. The low-frequency communication anti-interference control method based on a multi-layered defense system according to claim 2, characterized in that, Based on the spatiotemporal signal sample set, signal energy distribution features are extracted, and regions with concentrated signal intensity are determined by calculating the energy density of each frequency band, including: The spatiotemporal signal sample set is preprocessed by windowing. The Hanning window function is used to window each signal sample to obtain the windowed signal sample. Based on the windowed signal samples, the signals are transformed to the frequency domain using a fast Fourier transform to obtain the spectral distribution of each signal sample. Based on the spectral distribution, the power spectral density of each frequency band is calculated, and the energy density value of each frequency band is obtained through integration. Based on the energy density values ​​of each frequency band, an energy density threshold is determined, and frequency bands with energy densities exceeding the threshold are selected. Based on the frequency bands where the energy density exceeds the threshold, the region where the signal strength is concentrated is determined.

4. The low-frequency communication anti-interference control method based on a multi-layered defense system according to claim 3, characterized in that, Based on the aforementioned signal reference point, two spatial orientation vectors are determined; Calculate the phase difference between two spatial orientation vectors based on the spatiotemporal signal sample set to delineate a signal feature domain, including: Based on the signal reference point, the main direction vector is determined, and the first spatial orientation vector is generated through orthogonal decomposition. Based on the first spatial orientation vector, determine the second spatial orientation vector that is orthogonal to it; Based on the spatiotemporal signal sample set, the signal phase characteristics in the directions of the first spatial azimuth vector and the second spatial azimuth vector are calculated respectively, and the phase difference between the two spatial azimuth vectors is obtained through differential operation. Based on the phase difference, the signal characteristic interval is divided, and the signal characteristic domain is determined.

5. The low-frequency communication anti-interference control method based on a multi-layered defense system according to claim 4, characterized in that, A closed-loop signal phase path is constructed based on the phase correlation of each signal monitoring node in the time domain; Based on the closed path of the signal phase, a signal anti-interference compensation coefficient is generated, including: Based on the spatiotemporal location information of the three signal monitoring nodes, signal phase data of each node are collected at the same time period. Based on the signal phase data, the relative phase difference between each monitoring node is calculated, and a complete phase distribution map is constructed. Based on the phase distribution map, the phase characteristic points of each monitoring node are connected to form a closed path of signal phase; Based on the closed path of the signal phase, the curvature characteristics of each point on the path are calculated, and the overall curvature parameter of the path is obtained by integration. Based on the overall curvature parameter of the path, a signal anti-interference compensation coefficient is generated.

6. The low-frequency communication anti-interference control method based on a multi-layered defense system according to claim 5, characterized in that, Based on the signal anti-interference compensation coefficient, a multi-mode redundancy architecture is used to preprocess low-frequency communication signals to generate a signal stream with time-frequency redundancy characteristics, including: Based on the signal anti-interference compensation coefficient, the original low-frequency communication signal is initially filtered to obtain the filtered signal. Based on the filtered signal, the filtered signal is distributed to three parallel processing channels, and each channel is processed using a different signal modulation method to obtain the output signal of the three parallel processing channels. Based on the output signals of the three parallel processing channels, the time-domain and frequency-domain features of each channel signal are extracted, and composite signal features are generated. Based on the characteristics of the composite signal, consistency detection is performed on the signals of each channel, and an intermediate signal stream with time-frequency redundancy is generated. The intermediate signal stream is optimized by using a signal anti-interference compensation coefficient to ultimately generate a signal stream with time-frequency redundancy characteristics.

7. The low-frequency communication anti-interference control method based on a multi-layered defense system according to claim 6, characterized in that, A quantum encrypted signal stream is generated by encrypting a signal stream with time-frequency redundancy using a quantum key distribution protocol, including: Based on a signal stream with time-frequency redundancy, a quantum random number generator is used to generate a quantum true random number sequence as the initial encryption key. Based on the initial encryption key, a shared quantum key is established between the two communicating parties using the BB84 quantum key distribution protocol; Based on the shared quantum key, the time-frequency redundant signal stream is subjected to layered encryption processing to generate a primary encrypted signal stream; Based on the aforementioned primary encrypted signal stream, the integrity of the encryption process is verified through a quantum authentication protocol, and a message authentication code is generated using a quantum hash algorithm. Based on the verified encrypted signal stream and message authentication code, the verified encrypted signal stream is converted into a quantum state signal, ultimately generating a quantum encrypted signal stream.

8. The low-frequency communication anti-interference control method based on a multi-layered defense system according to claim 7, characterized in that, A blockchain-based identity authentication mechanism and dynamic access control strategy are employed to securely manage quantum encrypted signal streams, achieving real-time protection and trusted access control of communication links, including: Based on the quantum encryption signal stream, a distributed identity authentication system is established, and digital identity credentials are generated; Based on the digital identity credentials, the legitimacy and trustworthiness of the communication nodes are verified through a consensus mechanism to obtain the verified communication nodes; Based on verified communication nodes, dynamic access control is implemented, and access permissions are adjusted according to real-time network status and security policies to obtain the adjusted access permissions. Based on the adjusted access permissions, the access control list is updated in real time through the blockchain transaction record mechanism, and the access control policy is automatically executed by smart contracts. Based on the executed access control policy, the transmission process of quantum encrypted signal streams is monitored and audited throughout, achieving real-time protection and trusted access control of the communication link.

9. A low-frequency communication anti-interference control system based on a multi-layered defense system, wherein the system implements the method as described in any one of claims 1 to 8, characterized in that, include: The acquisition module is used to acquire low-frequency communication signal data and convert the low-frequency communication signal data into a set of spatiotemporal signal samples; A signal reference point is determined based on the spatiotemporal signal sample set; two spatial orientation vectors are determined based on the signal reference point; the phase difference between the two spatial orientation vectors is calculated based on the spatiotemporal signal sample set to delineate a signal feature domain; three signal monitoring nodes are determined using a distributed detection mechanism, with the three signal monitoring nodes located in the effective and invalid intervals of the signal feature domain, respectively; a closed signal phase path is constructed based on the phase correlation of each signal monitoring node in the time domain. Based on the closed path of the signal phase, a signal anti-interference compensation coefficient is generated. The generation module is used to preprocess low-frequency communication signals based on the signal anti-interference compensation coefficient and adopt a multi-mode redundancy architecture to generate a signal stream with time-frequency redundancy characteristics. The encryption module is used to encrypt a signal stream with time-frequency redundancy characteristics using a quantum key distribution protocol, thereby generating a quantum encrypted signal stream. The control module is used to securely manage the quantum encrypted signal stream using a blockchain-based identity authentication mechanism and dynamic access control strategy, thereby achieving real-time protection and trusted access control of the communication link.

Images