Method and device for suppressing water acoustic signal motion noise based on vector magnetic sensor

Abstract

The application discloses a water acoustic signal motion noise suppression method and device based on a vector magnetic sensor and belongs to the technical field of low-frequency ocean environment physical field monitoring. The method uses a vector magnetic sensor carried on a mooring platform to collect magnetic field signals in real time, obtains a relative angle change sequence representing platform attitude change by calculating the minimum rotation angle of magnetic field vectors at adjacent moments. The sequence is taken as a reference input, and acoustic signals collected synchronously by a hydrophone are input into a variable step adaptive filter; the filter dynamically adjusts the step according to the energy of the reference input, estimates and offsets low-frequency noise introduced by platform motion in the acoustic signals, and outputs the suppressed water acoustic signals. The application realizes real-time and adaptive suppression of platform motion noise through joint processing of magnetic fields and acoustic fields, solves the problem that a traditional method is difficult to distinguish low-frequency motion interference from effective acoustic signals, and significantly improves the water acoustic observation precision and data quality of the mooring platform under complex sea conditions.

Description

Technical Field

[0001] This invention belongs to the field of low-frequency marine environmental physical field monitoring technology, specifically relating to a method and device for suppressing motion noise of underwater acoustic signals based on vector magnetic sensors. Background Technology

[0002] Moored platforms are widely used in marine environmental monitoring and underwater observation due to their advantages such as flexible deployment and long-term continuous operation. These platforms are typically maintained in a relatively fixed position by mooring cables and anchoring structures. However, during actual operation, they are still inevitably affected by external disturbances such as waves and currents, resulting in periodic or random platform movements. These movements, through structural transmission and fluid coupling, introduce significant low-frequency background noise into the hydrophone measurements mounted on the platform, severely interfering with the stable acquisition of weak signals.

[0003] To address the measurement noise problem caused by moored platforms, existing technologies mainly focus on two aspects: engineering design and signal processing. At the engineering design level, optimizing the platform's shape, adding damping devices, or improving the mooring structure can reduce the platform's motion response under wave action, thereby reducing noise introduction. However, these methods are highly sensitive to environmental conditions; when external disturbances exceed the design limits, the vibration suppression effect significantly decreases, and it is difficult to dynamically compensate for motion noise already transmitted to the sensor.

[0004] At the signal processing level, existing methods mostly employ noise suppression techniques targeting single sensors, such as high-pass filtering, band-stop filtering, or detrending processing. However, due to the severe spectral overlap between platform motion noise and real-world environmental signals in the low-frequency range (especially 1Hz to 10Hz), traditional filtering methods often weaken useful signal components while suppressing noise, leading to measurement information loss. In recent years, some studies have attempted to introduce auxiliary devices such as inertial measurement units or attitude and heading reference systems to perform post-processing data correction by recording platform attitude information. However, these methods mostly remain at the data comparison level, making it difficult to form a real-time, closed-loop noise compensation mechanism, and increasing system complexity and power consumption.

[0005] In summary, existing technologies still struggle to simultaneously maintain the integrity of low-frequency signals and effectively suppress motion noise in complex marine dynamic environments. There is an urgent need for a highly efficient noise suppression solution that can sense the platform's motion status in real time and dynamically adapt to environmental changes. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a method and apparatus for suppressing motion noise in underwater acoustic signals based on a vector magnetic sensor. For the first time, a vector magnetic sensor is used to sense changes in the attitude of a tethered platform. By constructing a dynamic response model of magnetic field changes and acoustic noise, a physically interpretable identification and adaptive suppression of low-frequency motion noise in hydrophone signals is achieved.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] A method for suppressing motion noise in underwater acoustic signals based on a vector magnetic sensor, applied to a tethered platform equipped with a hydrophone and a vector magnetic sensor, the method comprising:

[0009] Step 1: Obtain the triaxial magnetic field data output in real time by the vector magnetic sensor and the raw underwater acoustic signal synchronously collected by the hydrophone. Based on the triaxial magnetic field data, calculate the relative angle change sequence that characterizes the real-time attitude change of the moored platform under the action of waves.

[0010] Step 2: The relative angle change sequence is used as the reference input, and the original underwater acoustic signal is used as the desired input. Both are input to the variable step size adaptive filter. The variable step size adaptive filter dynamically adjusts the filtering step size according to the energy change of the reference input, estimates and cancels the low-frequency motion noise component introduced by the platform attitude change in the original underwater acoustic signal, and outputs the noise-suppressed underwater acoustic signal.

[0011] Furthermore, the calculation of the relative angle change sequence characterizing the real-time attitude change of the moored platform under the action of waves includes: normalizing the three-axis magnetic field vectors at adjacent moments to obtain normalized magnetic field vectors; calculating the angle between the normalized magnetic field vectors at adjacent moments and using it as the relative angle change between adjacent moments; repeating the relative angle change calculation steps for all adjacent moments to obtain the relative angle change sequence.

[0012] Furthermore, the angle between the normalized magnetic field vectors at adjacent moments is calculated using inverse trigonometric functions. Specifically, the dot product and inner product of the two normalized magnetic field vectors are calculated first, and then the angle is determined based on the ratio of the dot product to the inner product.

[0013] Furthermore, the variable step-size adaptive filter employs a variable step-size normalized least mean square algorithm, and the step-size factor in its filter weight update formula is dynamically adjusted based on the energy of the reference input vector at the current moment.

[0014] Furthermore, the dynamic adjustment method of the step size factor is as follows: when the energy of the reference input vector is lower than the preset threshold, it is determined that the tethered platform is in a weak motion state, and the step size factor is increased to above the reference step size to accelerate the convergence speed of the filter; when the energy of the reference input vector is not lower than the preset threshold, the reference step size is maintained.

[0015] Furthermore, the weight update formula for the variable step-size adaptive filter is:

[0016] The filter output at the current time is equal to the transpose of the filter weight vector at the current time multiplied by the reference input vector at the current time;

[0017] The error signal at the current moment is equal to the original underwater acoustic signal at the current moment minus the filtered output at the current moment.

[0018] Furthermore, the filter weight vector at the next time step is equal to the filter weight vector at the current time step step factor multiplied by the error signal multiplied by the reference input vector at the current time step step step factor multiplied by the reference input vector at the current time step step step factor multiplied by the reference input vector at the current time step step step factor and a constant to prevent the denominator from being zero.

[0019] On the other hand, the present invention provides a motion noise suppression device for underwater acoustic signals based on a vector magnetic sensor, comprising:

[0020] The calculation module is used to acquire the triaxial magnetic field data output in real time by the vector magnetic sensor and the raw underwater acoustic signal synchronously collected by the hydrophone. Based on the triaxial magnetic field data, it calculates the relative angle change sequence that characterizes the real-time attitude change of the moored platform under the action of waves.

[0021] The output module is used to input the relative angle change sequence as a reference input and the original underwater acoustic signal as the desired input to a variable step size adaptive filter. The variable step size adaptive filter dynamically adjusts the filtering step size according to the energy change of the reference input, estimates and cancels the low-frequency motion noise component introduced by the platform attitude change in the original underwater acoustic signal, and outputs the noise-suppressed underwater acoustic signal.

[0022] Thirdly, the present invention provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs; wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the aforementioned underwater acoustic signal motion noise suppression method based on a vector magnetic sensor.

[0023] Fourthly, the present invention provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, enable the processor to implement the aforementioned method for suppressing motion noise of underwater acoustic signals based on a vector magnetic sensor.

[0024] The beneficial effects of this invention are as follows:

[0025] First, this invention establishes for the first time a joint magnetic field-sound field processing mechanism based on a vector magnetic sensor. By converting the platform attitude changes sensed by the vector magnetic sensor into a sequence of relative angle changes, which serves as the reference input for adaptive filtering, physically interpretable identification of platform motion noise is achieved. Compared with traditional denoising methods that rely on a single acoustic feature, this invention can distinguish between platform motion interference and real background sound signals at the signal source, effectively solving the signal distortion problem caused by low-frequency spectrum overlap.

[0026] Secondly, the variable step-size adaptive filtering algorithm proposed in this invention exhibits higher response sensitivity and stability. Addressing the sensitivity degradation of vector magnetic sensors at high frequencies, a dynamic step-size adjustment mechanism based on reference input energy is introduced: when platform motion is weak or sensor response decreases, the step size is automatically increased to accelerate convergence; when motion is strong, the step size is automatically decreased to ensure steady-state accuracy. This design overcomes the shortcomings of traditional fixed-step-size algorithms, such as slow convergence or over-filtering under complex sea conditions, achieving fast and stable adaptive noise suppression.

[0027] Third, the method of this invention has a simple structure, low computational overhead, and strong real-time performance. It requires no additional attitude sensors or inertial measurement units; precise perception of motion and dynamic noise cancellation can be achieved solely using the platform's existing vector magnetic sensors. Experimental results show that this method can effectively eliminate buoy motion noise below 10Hz, and the spectral characteristics of the processed acoustic signal match well with those of the stationary reference platform. This significantly improves the accuracy and reliability of underwater acoustic observations by moored platforms in complex sea conditions, providing a stable preprocessing algorithm foundation for distributed buoy network detection. Attached Figure Description

[0028] Figure 1 This is a flowchart of the underwater acoustic signal motion noise suppression method based on a vector magnetic sensor according to the present invention;

[0029] Figure 2 This is a schematic diagram illustrating the principle of a vector magnetic sensor sensing attitude changes.

[0030] Figure 3 The correlation curves of acoustic and attitude signals during a typical time period are shown.

[0031] Figure 4 This is a schematic diagram of the variable step size adaptive filtering principle.

[0032] Figure 5(a) shows a comparison of the frequency domains before and after noise reduction of the tethered platform;

[0033] Figure 5(b) is a comparison of the frequency domain data of different platforms. Detailed Implementation

[0034] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0035] This invention uses a vector magnetic sensor to monitor the platform's motion state in real time, and uses the relative angle change output as a reference input for acoustic signal processing. Through synchronous acquisition and correlation analysis of magnetic and acoustic signals, a dynamic response model is constructed between the two, enabling the identification and suppression of motion noise components. Based on this, the invention further improves the traditional adaptive filtering algorithm by introducing a variable step-size adaptive mechanism driven by relative angle changes. This allows the filtering process to automatically adjust and update coefficients according to the platform's motion amplitude, achieving dynamic attenuation of low-frequency noise without destroying effective acoustic information. This design overcomes the limitation of traditional methods that rely solely on the characteristics of the acoustic signal itself to distinguish motion interference, accurately eliminating low-frequency noise caused by platform attitude changes under complex motion conditions. Through magnetic field-sound field collaborative processing, this invention provides a real-time, physically interpretable, and engineering-feasible motion noise suppression scheme for underwater acoustic measurements on buoy platforms, effectively improving the reliability and data quality of low-frequency acoustic detection.

[0036] like Figure 1 The diagram shows the overall flowchart of the underwater acoustic signal motion noise suppression method based on a vector magnetic sensor according to the present invention. A three-axis vector magnetic sensor installed on a tethered platform is used to sense and quantify the platform's attitude changes in the geomagnetic field in real time. By analyzing the variation law of the vector magnetic sensor output signal, the relative angle change sequence of the platform is calculated, thereby obtaining a reference signal characterizing the platform's motion state. This reference signal and the acoustic signal collected by the hydrophone are synchronously input into a variable step-size adaptive filter. The filter dynamically adjusts the step-size parameter according to the platform's motion intensity to achieve adaptive suppression of low-frequency motion noise, ultimately obtaining an acoustic signal free from platform motion interference. Specifically, the method is applied to a tethered platform equipped with a hydrophone and a vector magnetic sensor, including:

[0037] Step 1: Obtain the triaxial magnetic field data output in real time by the vector magnetic sensor and the raw underwater acoustic signal synchronously collected by the hydrophone. Based on the triaxial magnetic field data, calculate the relative angle change sequence that characterizes the real-time attitude change of the moored platform under the action of waves.

[0038] A vector magnetic sensor is a high-sensitivity vector magnetic field sensor based on the periodic saturation characteristics of soft magnetic materials, capable of accurately measuring the component of an external magnetic field along its sensitive axis. Its basic structure consists of a high-permeability iron core, an excitation coil, and an induction coil. When an alternating current is applied to the excitation coil, the magnetization state of the iron core periodically switches between positive and negative saturation, thereby generating a secondary induced voltage signal in the induction coil. When a static magnetic field (such as the Earth's magnetic field) exists externally, the symmetry of this magnetization curve is broken, and the signal amplitude generated in the induction coil is approximately proportional to the component of the external magnetic field. By using three sets of mutually orthogonally arranged vector magnetic sensors, the three-dimensional vector components of the external magnetic field in the sensor coordinate system can be simultaneously acquired.

[0039] The Earth's magnetic field exhibits good homogeneity within a finite spatial scale (such as within a few meters of a moored platform), and its direction and intensity can be considered a constant vector field. When a buoy equipped with a vector magnetic sensor changes its attitude on the sea surface, the components of the geomagnetic vector measured in the buoy's coordinate system will also change accordingly. In other words, although the actual geomagnetic vector remains fixed in the geographic reference coordinate system, its projection direction will rotate accordingly in the coordinate system that rotates with the buoy's attitude.

[0040] Based on this characteristic, the triaxial output signals of the vector magnetic sensor measured at different times can be regarded as the projection results of the same spatial geomagnetic vector under different attitudes. By normalizing the magnetic vectors measured at different times and calculating their included angle and minimum rotation matrix relationship, the degree of change in buoy attitude can be quantitatively described, which can be used to characterize the tilting, swaying or flipping trend of the buoy under the action of waves, flow fields, etc.

[0041] The buoy's motion can be considered as the superposition of its rotation around its center of gravity and its translation. According to the global geomagnetic field gradient model, the spatial gradient of the geomagnetic field is extremely small at the ocean scale; therefore, the influence of the vector magnetic sensor's translation with the center of gravity on the measurement results is negligible. Similarly, the accompanying hydrophone is also insensitive to translation in far-field sound perception. Therefore, in the buoy's motion analysis, the attitude change around the center of gravity can be mainly considered.

[0042] The invention employs a three-axis vector magnetic sensor as the motion sensing unit. Vector magnetic sensors offer advantages such as high sensitivity, low drift, and wide bandwidth response, accurately responding to minute changes in the magnetic field at low frequencies. Compared to common accelerometers or gyroscopes, vector magnetic sensors are sensitive to low-frequency fluctuations and can directly obtain stable vector direction information without integration calculations, making them suitable for detecting slow, slightly swaying buoy movements.

[0043] Therefore, by using a vector magnetic sensor to measure changes in the geomagnetic vector in real time, an approximate characterization of buoy attitude changes can be achieved without introducing a complex inertial measurement unit (IMU). This method is simple in structure, has low power consumption, and good long-term stability, making it suitable for attitude monitoring and dynamic analysis of long-term ocean observation platforms.

[0044] Since the geomagnetic vector has a fixed direction in the geographic system, the difference in the direction of the total geomagnetic field measured by the triaxial vector magnetic sensor at different times is only caused by changes in the buoy's attitude. When the buoy's attitude undergoes a slight rotation, the minimum rotation matrix and the corresponding rotation angle change between adjacent times can be calculated using the Rodrigues rotation formula, thus obtaining the buoy's motion change.

[0045] like Figure 2 As shown, the specific calculation process is as follows:

[0046] (1) Calculate the normalized total field measured by the triaxial vector magnetic sensor. Wherein and This represents the normalized total field measured by the vector magnetic sensor. " represents the magnitude of the vector, This indicates that the vector magnetic sensor measures the total field. , , These represent the magnetic field components in the x, y, and z directions measured by the vector magnetic sensor, respectively.

[0047] ,

[0048] ,

[0049] ,

[0050] (2) Calculate the axis of rotation between the coordinate systems at two time points. and minimum rotation angle 。 " represents the cross product between vectors, "" indicates the inner product between vectors.

[0051] ,

[0052] ,

[0053] (3) By assembling a sequence of the minimum rotation angle changes obtained at all sampling times, the relative angle change sequence of the buoy's real-time attitude change under the action of waves can be obtained. .

[0054] Step 2: The relative angle change sequence is used as the reference input, and the original underwater acoustic signal is used as the desired input. Both are input to the variable step size adaptive filter. The variable step size adaptive filter dynamically adjusts the filtering step size according to the energy change of the reference input, estimates and cancels the low-frequency motion noise component introduced by the platform attitude change in the original underwater acoustic signal, and outputs the noise-suppressed underwater acoustic signal.

[0055] When a hydrophone and a vector magnetic sensor are mounted together on a buoy platform, the platform experiences irregular motion due to external forces such as waves and wind, causing the hydrophone's spatial position to change over time. This results in significant low-frequency interference superimposed on the sound pressure signal. This type of interference is commonly referred to as motion noise, and its main sources include:

[0056] (1) Overall movement of the platform: The periodic up-and-down movement of the buoy will cause changes in the hydrostatic pressure on the underwater sensing element, thus forming a low-frequency drift component in the acoustic signal;

[0057] (2) Structural vibration coupling caused by attitude change: The change of buoy angle will cause vibration energy to be coupled into the sensitive direction of the transducer;

[0058] (3) Relative flow-induced noise: The change in attitude causes the water flow direction to change, which in turn causes micro-flow disturbances on the surface of the hydrophone.

[0059] The main characteristics of these motion interference signals are low frequency, significant amplitude variation with sea state, and strong correlation with buoy angle changes.

[0060] like Figure 3 As shown, the present invention utilizes a vector magnetic sensor to calculate the relative angle change sequence. This method can effectively reflect the dynamic process of buoy attitude. Calculation results show that the correlation coefficient between the relative angle change sequence and the low-frequency noise amplitude in the hydrophone acoustic signal has a significant correlation in multiple observation periods, verifying the effectiveness of using the vector magnetic sensor output for buoy motion feature extraction.

[0061] Using the calculated angle change time series as the reference input signal for the subsequent variable step size adaptive filter can accurately reflect the intensity and direction of the interference of platform motion on the acoustic signal.

[0062] To effectively suppress low-frequency noise caused by buoy motion, this invention proposes an adaptive filtering method based on the Variable Step-Size Normalized Least Mean Square (VSS-NLMS) algorithm. This method uses the relative angle change sequence calculated by a vector magnetic sensor as the reference input signal and the acoustic signal collected by a hydrophone as the desired input signal. By dynamically adjusting the filtering step size, it achieves adaptive separation of motion noise and target acoustic signal.

[0063] The principle of variable step size adaptive filtering is as follows: Figure 4 As shown, the update of the adaptive filter weights depends on two parts of data: one is the noisy angle input signal reflecting the buoy's attitude change. Right now Secondly, acoustic observation signals containing motion noise. The filter generates an estimated output based on the input signal. This represents the predicted motion noise component; subtracting this output from the acoustic observation signal yields the error signal. This refers to the effective acoustic signal after removing motion interference. The algorithm, at each sampling time, calculates the effective acoustic signal based on the input signal vector. and filter weight vector The output value is calculated, and the weights are updated in real time according to the normalized least mean square principle to achieve self-learning suppression of buoy motion noise. Its mathematical expression is:

[0064] ,

[0065] ,

[0066] ,

[0067] in, Step size factor A constant to prevent the denominator from approaching zero.

[0068] Traditional NLMS algorithms use a fixed step size, which often leads to slow convergence or over-adjustment when the input signal energy changes or the reference signal sensitivity is insufficient. Given that the output sensitivity of vector magnetic sensors drops significantly above approximately 4Hz, weakening their response to high-frequency motion, using a fixed step size makes it difficult to balance stability and convergence. Therefore, this invention introduces a variable step size mechanism based on real-time changes in input signal energy: when the input vector energy is low, indicating weak buoy motion or reduced vector magnetic sensor response, the step size is appropriately increased to accelerate filtering convergence; when the signal energy is strong, the step size is decreased to prevent oscillation. The specific variable step size strategy can be expressed as:

[0069] ,

[0070] in, As the reference step size, This is the magnification factor. The energy threshold is used. By adaptively adjusting the step size based on energy, the filter can maintain stable and rapid convergence at low frequencies, ensuring the stability of the filtering process under different signal energy conditions.

[0071] This variable step-size adaptive filtering method establishes a real-time correlation between the angle change signal and the acoustic signal. The angle change sequence serves as an external reference for platform motion, and the VSS-NLMS structure directly estimates and cancels correlated motion noise in the acoustic signal, resulting in significant noise reduction in the low-frequency band. The algorithm is characterized by its simple structure, low computational overhead, and strong real-time performance, making it particularly suitable for scenarios requiring low-frequency noise suppression, such as buoy arrays and combined acoustic-magnetic observations. Experimental results show that this filter effectively eliminates buoy motion noise below 10Hz while preserving the acoustic signal fidelity, providing a stable and reliable preprocessing algorithm foundation for subsequent distributed buoy network detection.

[0072] To verify the effectiveness of the adaptive low-frequency noise suppression method for vector magnetic sensors proposed in this invention, a sea trial was conducted in a certain sea area. The trial included two sets of marine acoustic-magnetic joint observation systems: a submerged platform and a moored platform. The submerged platform is minimally affected by fluid disturbances, and its acoustic field record can be regarded as a relatively stationary reference baseline. The method was verified using data from the moored platform. If the denoised data from the moored platform is close to that from the submerged platform in terms of spectral distribution and energy level, it can be proven that the main component suppressed by the algorithm is indeed motion noise.

[0073] The processing results are shown in Figures 5(a) and 5(b). The quality of the acoustic signal of the tethered platform after processing by the proposed method is significantly improved, and its spectral characteristics are in good agreement with those of the bottom-mounted reference platform, which verifies the reliability and engineering application value of the algorithm.

[0074] In summary, this application not only achieves a fundamental breakthrough in algorithm structure, moving from single-signal filtering to magneto-acoustic collaborative processing, but also makes substantial improvements in motion noise identification, parameter self-adjustment, and real-time application. It significantly enhances the accuracy and reliability of low-frequency acoustic observation and provides a new solution for the long-term stable operation of tethered acoustic measurement platforms in complex sea conditions.

[0075] On the other hand, the present invention provides a motion noise suppression device for underwater acoustic signals based on a vector magnetic sensor, wherein the various modules included are capable of implementing the various steps of the aforementioned method, specifically including:

[0076] The calculation module is used to acquire the triaxial magnetic field data output in real time by the vector magnetic sensor and the raw underwater acoustic signal synchronously collected by the hydrophone. Based on the triaxial magnetic field data, it calculates the relative angle change sequence that characterizes the real-time attitude change of the moored platform under the action of waves.

[0077] The output module is used to input the relative angle change sequence as a reference input and the original underwater acoustic signal as the desired input to a variable step size adaptive filter. The variable step size adaptive filter dynamically adjusts the filtering step size according to the energy change of the reference input, estimates and cancels the low-frequency motion noise component introduced by the platform attitude change in the original underwater acoustic signal, and outputs the noise-suppressed underwater acoustic signal.

[0078] Thirdly, the present invention provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs; wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the aforementioned underwater acoustic signal motion noise suppression method based on a vector magnetic sensor.

[0079] Fourthly, the present invention provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, enable the processor to implement the aforementioned method for suppressing motion noise of underwater acoustic signals based on a vector magnetic sensor.

[0080] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for suppressing motion noise in underwater acoustic signals based on a vector magnetic sensor, applied to a tethered platform equipped with a hydrophone and a vector magnetic sensor, characterized in that, The method includes: Step 1: Obtain the triaxial magnetic field data output in real time by the vector magnetic sensor and the raw underwater acoustic signal synchronously collected by the hydrophone. Based on the triaxial magnetic field data, calculate the relative angle change sequence that characterizes the real-time attitude change of the moored platform under the action of waves. Step 2: The relative angle change sequence is used as the reference input, and the original underwater acoustic signal is used as the desired input. Both are input to the variable step size adaptive filter. The variable step size adaptive filter dynamically adjusts the filtering step size according to the energy change of the reference input, estimates and cancels the low-frequency motion noise component introduced by the platform attitude change in the original underwater acoustic signal, and outputs the noise-suppressed underwater acoustic signal.

2. The method for suppressing motion noise in underwater acoustic signals based on a vector magnetic sensor according to claim 1, characterized in that, The calculation of the relative angle change sequence characterizing the real-time attitude change of the moored platform under the action of waves includes: normalizing the three-axis magnetic field vectors of adjacent moments to obtain normalized magnetic field vectors; calculating the angle between the normalized magnetic field vectors of adjacent moments and using it as the relative angle change between adjacent moments; repeating the relative angle change calculation steps for all adjacent moments to obtain the relative angle change sequence.

3. The method for suppressing motion noise in underwater acoustic signals based on a vector magnetic sensor according to claim 2, characterized in that, The angle between the normalized magnetic field vectors at adjacent moments is calculated using inverse trigonometric functions. Specifically, the dot product and inner product of the two normalized magnetic field vectors are calculated first, and then the angle is determined based on the ratio of the dot product to the inner product.

4. The method for suppressing motion noise in underwater acoustic signals based on a vector magnetic sensor according to claim 1, characterized in that, The variable step-size adaptive filter employs a variable step-size normalized least mean square algorithm, and the step-size factor in its filter weight update formula is dynamically adjusted based on the energy of the reference input vector at the current moment.

5. The method for suppressing motion noise in underwater acoustic signals based on a vector magnetic sensor according to claim 4, characterized in that, The dynamic adjustment method of the step size factor is as follows: when the energy of the reference input vector is lower than the preset threshold, it is determined that the tethered platform is in a weak motion state, and the step size factor is increased to above the reference step size to accelerate the convergence speed of the filter. Maintain the baseline step size when the energy of the reference input vector is not lower than the preset threshold.

6. The method for suppressing motion noise in underwater acoustic signals based on a vector magnetic sensor according to claim 1, characterized in that, The weight update formula for the variable step size adaptive filter is: The filter output at the current time is equal to the transpose of the filter weight vector at the current time multiplied by the reference input vector at the current time; The error signal at the current moment is equal to the original underwater acoustic signal at the current moment minus the filtered output at the current moment.

7. The method for suppressing motion noise in underwater acoustic signals based on a vector magnetic sensor according to claim 6, characterized in that, The filter weight vector at the next time step is equal to the filter weight vector at the current time step step factor multiplied by the error signal multiplied by the reference input vector at the current time step step step factor multiplied by the reference input vector at the current time step step step factor multiplied by the reference input vector at the current time step step step factor and a constant to prevent the denominator from being zero.

8. A device for suppressing motion noise in underwater acoustic signals based on a vector magnetic sensor, characterized in that, include: The calculation module is used to acquire the triaxial magnetic field data output in real time by the vector magnetic sensor and the raw underwater acoustic signal synchronously collected by the hydrophone. Based on the triaxial magnetic field data, it calculates the relative angle change sequence that characterizes the real-time attitude change of the moored platform under the action of waves. The output module is used to input the relative angle change sequence as a reference input and the original underwater acoustic signal as the desired input to the variable step size adaptive filter. The variable step size adaptive filter dynamically adjusts the filtering step size according to the energy change of the reference input, estimates and cancels the low-frequency motion noise component introduced by the platform attitude change in the original underwater acoustic signal, and outputs the noise-suppressed underwater acoustic signal.

9. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When one or more programs are executed by the one or more processors, the one or more processors implement the underwater acoustic signal motion noise suppression method based on a vector magnetic sensor as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, It stores executable instructions that, when executed by a processor, enable the processor to implement the underwater acoustic signal motion noise suppression method based on a vector magnetic sensor as described in any one of claims 1-7.

Images