Marine acoustic environment monitoring system based on noise map

Abstract

The invention provides a marine acoustic environment monitoring system based on a noise map. The marine acoustic environment monitoring system comprises a buoy network module, a data fusion processing module, a noise map generation module, a noise source identification module and an edge response module, the buoy network module serves as a sensing tail end, multi-source information is gathered to the data fusion processing module through the first data bus for standardization processing, and then the noise map generation module is driven through the second data bus to construct a three-dimensional sound field model. The output of the model is transmitted to a noise source identification module through a third data bus for intelligent analysis and traceability; the alarm decision of the noise source identification module is output to the edge response module, and the edge response module dynamically regulates and controls the communication and acquisition strategy of each buoy node through a distributed control circuit to form a complete closed loop. According to the invention, high-precision and automatic monitoring of the marine acoustic environment is realized, the problems of limited monitoring range and single data dimension of the traditional means are solved, and the reliability and emergency capability of the system are improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of marine environment monitoring, and particularly relates to a marine acoustic environment dynamic monitoring system fusing a noise map. BACKGROUND

[0002] Current marine acoustic environment monitoring mainly relies on fixed hydrophone arrays or single-point measurement devices, which have the following significant defects: first, the spatial coverage range is limited, and the spatial distribution characteristics of a large range of sound fields cannot be obtained; second, the measurement data and marine environment parameters are processed separately, resulting in a calculation error of sound propagation loss exceeding ±5dB; and third, the static deployment mode is difficult to adapt to dynamic changes in the marine environment.

[0003] Existing noise map technologies mainly rely on theoretical sound propagation models and lack dynamic calibration of measured data. The main problems are: 1) the map update cycle is long, and transient sound sources such as ship noise cannot be captured; 2) commercial software does not integrate intelligent separation modules for biological noise and artificial noise, resulting in distorted ecological assessment; and 3) the map generation process is not associated with AIS and other shipping data sources, which results in a high false alarm rate in existing systems in scenarios such as offshore wind farm supervision.

[0004] Although machine learning has been applied to marine acoustic signal processing, the existing technology has three major shortcomings: 1) the feature library sample is single, and the classification accuracy of overlapping noise is less than 50%; 2) the edge computing capability is weak, and the node end cannot realize real-time alarm of abnormal noise; and 3) no correlation model is established between acoustic fingerprints and environmental factors. SUMMARY

[0005] The present application aims to solve the technical problems of limited spatial coverage, lagging noise map updates, and ambiguous noise source identification in existing marine acoustic monitoring. By constructing a dynamic noise map generation and intelligent interpretation engine, a marine acoustic environment dynamic monitoring system capable of realizing large-scale sound field distribution and accurate noise source tracing is introduced.

[0006] The marine acoustic environment monitoring system based on a noise map comprises:

[0007] The buoy network module is composed of at least six buoy nodes, each buoy node serving as a distributed data acquisition unit. The buoy node acquires underwater noise signals through a built-in underwater acoustic sensor array, obtains temperature, salinity, depth, and seawater flow rate data through an environmental parameter sensor, and dynamically switches transmission links according to channel quality and transmits multi-source information to a data fusion processing module through a self-reconfigurable communication circuit.

[0008] The data fusion processing module is connected with the buoy network module through a first data bus and receives buoy network data, receives satellite remote sensing data through a satellite data interface, and accesses shore-based radar data through a radar input port.

[0009] The noise map generation module is connected with the output end of the data fusion processing module through a second data bus.

[0010] The noise source identification module is connected with the output end of the noise map generation module through a third data bus.

[0011] The edge response module is in communication connection with the noise source identification module and is connected with the self-reconstruction communication circuit of each buoy node through a distributed control line.

[0012] The marine acoustic environment monitoring system based on a noise map of the application realizes high-precision, real-time and automatic monitoring of a wide-area marine acoustic environment through the cooperative work of a distributed buoy network, multi-source data fusion, noise map visualization, noise source intelligent identification and edge response.

[0013] For the optimization scheme of the above monitoring system, the hydroacoustic sensor array, the environmental parameter sensor and the self-reconstruction communication circuit included in the buoy node constitute a signal processing link through the following relationship:

[0014] 1) The output end of the hydroacoustic sensor array accesses the analog input channel of the analog-to-digital converter through the analog multiplexer, and the digital output end of the analog-to-digital converter is connected to the ADC data interface of the node controller;

[0015] 2) The environmental parameter sensor is connected in parallel to the serial communication interface of the node controller through the RS-485 bus, and the synchronous clock output end of the environmental parameter sensor is directly connected to the sampling clock pin of the analog-to-digital converter;

[0016] 3) The input end of the self-reconfigurable communication circuit receives the fusion data packet sent by the node controller through the UART serial port, and the channel switching control end is connected to the GPIO output pin of the signal strength comparator, which monitors the signal-to-noise ratio of the underwater acoustic and wireless communication channels in real time;

[0017] The synchronous clock signal of the environmental parameter sensor triggers the synchronous collection of underwater acoustic signals by the analog-to-digital converter; the node controller transmits to the self-reconfigurable communication circuit module through the UART serial port; when the signal strength comparator detects that the signal-to-noise ratio of the underwater acoustic channel is <10 dB, the self-reconfigurable communication circuit is switched to the wireless radio frequency channel through the GPIO control line.

[0018] The buoy node here is designed with hardware synchronization and intelligent link switching mechanism, which ensures the high quality and high reliability of the collected data from the source. The synchronous clock of the environmental parameter sensor directly triggers the underwater acoustic signal collection, eliminating the time delay error caused by different clock synchronization, and laying a solid foundation for subsequent accurate noise map generation. At the same time, the signal strength comparator intelligently switches the communication channel, effectively overcoming the problem of environmental interference of the underwater acoustic channel, ensuring the continuity and stability of the data transmission link, and avoiding the loss of key monitoring data.

[0019] For the optimization scheme of the above monitoring system, the data fusion processing module includes a multi-source data receiving submodule, a space-time alignment submodule, and a normalized output submodule;

[0020] The multi-source data receiving submodule is used to receive satellite remote sensing data, shore-based radar data, and buoy network data in real time, complete physical layer signal demodulation and protocol analysis, and transmit the original data stream to the space-time alignment submodule through a high-speed serial bus;

[0021] The space-time alignment submodule adds a unified timestamp to the multi-source data based on the GPS timing signal, converts the spatial coordinates to the WGS-84 coordinate system, and generates space-time synchronization data. The input end of the space-time alignment submodule receives the output data of the multi-source data receiving submodule, and the output end of the space-time alignment submodule sends the space-time synchronization data to the normalized output submodule through the PCIe high-speed interface;

[0022] The normalization output submodule is used for converting heterogeneous data formats into a unified transmission protocol and performing dimension normalization processing, an input end of the normalization output submodule receives output data of the space-time alignment submodule, and an output end of the normalization output submodule outputs fusion data packets to the noise map generation module through a parallel data bus.

[0023] The data fusion processing module performs three-stage processing pipeline of "receiving-aligning-normalizing", and performs space-time alignment based on GPS and WGS-84 standards, integrates data of different sources, different formats and different dimensions into unified, standardized and highly available fusion data packets, which provides pure and standard data raw materials for the back-end noise map generation, eliminates fusion errors caused by inconsistent data benchmarks, and is a key prerequisite for realizing accurate analysis of the whole system.

[0024] For the optimization scheme of the monitoring system, the noise map generation module comprises a data buffering submodule, a grid processing submodule and a rendering output submodule, the data buffering submodule receives and buffers the fusion data packets output by the data fusion processing module, performs traffic shaping and priority scheduling, and an output end of the data buffering submodule provides a data stream to the grid processing submodule through a high-speed data transmission channel;

[0025] The grid processing submodule performs space interpolation calculation based on a sound propagation physical model and dynamically corrects sound attenuation errors, an input end of the grid processing submodule is connected to an output end of the data buffering submodule, and an output end of the grid processing submodule sends noise data to the rendering output submodule through a PCIe high-speed interface;

[0026] The rendering output submodule is used for converting noise data into a visual map, superimposing time coordinates and color scale information, an input end of the rendering output submodule receives output data of the grid processing submodule, and an output end of the rendering output submodule outputs a three-dimensional noise distribution map through a video interface.

[0027] The data flow output end of the data buffering sub-module is directly connected to the data input FIFO of the grid processing sub-module through a 64-bit parallel bus; the noise distribution matrix output end of the grid processing sub-module is connected to the texture memory of the rendering output sub-module through a PCIe x8 interface; the frame completion signal line of the rendering output sub-module is fed back to the cache clearing control end of the data buffering sub-module, triggering data emptying; the function chain timing is: multi-source data→data buffering→grid processing→rendering output (map generation). The noise map generation module has the significant advantages of high efficiency and low delay through the special hardware pipeline architecture of "buffering-processing-rendering", the traffic shaping and priority scheduling of the data buffering sub-module avoid data congestion, ensuring smooth processing of real-time data flow, the grid processing sub-module performs interpolation and sound attenuation correction based on a physical model, making the generated noise distribution matrix have clear physical meaning and high precision; the rendering output sub-module finally outputs a three-dimensional visual map with space-time and intensity information, intuitively presenting the sound field distribution, improving the explainability and human-computer interaction efficiency of the monitoring result, and the sub-modules are interconnected through high-speed buses and cooperated through feedback signals, ensuring the high-efficiency closed loop of the whole generation process.

[0028] For the optimization scheme of the above monitoring system, the noise source identification module includes a signal preprocessing sub-module, a feature matching sub-module and a decision output sub-module.

[0029] The input end of the signal preprocessing sub-module is connected to the data bus of the noise map generation module, the signal preprocessing sub-module performs frequency band selection and dynamic range compression on the input acoustic signal, and outputs a standard spectrum.

[0030] The feature matching sub-module is used for comparing the preprocessed standard spectrum with a voiceprint feature library, calculating the similarity and associating AIS ship information, the input end of the feature matching sub-module receives the output of the signal preprocessing sub-module, and the AIS interface of the feature matching sub-module is connected to a ship database through a serial bus.

[0031] The input end of the decision output sub-module receives the matching result of the feature matching sub-module, the decision output sub-module determines the noise source type according to the spectrum similarity threshold, and generates an alarm signal and a traceability report.

[0032] The connection relationship between the sub-modules in the noise source identification module is: the spectrum output end of the signal preprocessing sub-module is connected to the data buffer of the feature matching sub-module through a high-speed serial bus; the similarity matrix output end of the feature matching sub-module is connected to the threshold comparison unit of the decision output sub-module through a parallel decision bus; the ship query instruction end of the decision output sub-module is directly connected to the AIS data interface of the feature matching sub-module through an RS-485 bus; and the alarm trigger signal line of the decision output sub-module is fed back to the frequency band control end of the signal preprocessing sub-module. Here, by establishing a standardized signal processing and decision-making process, the automation, intelligence and high precision of noise source identification are realized; the signal preprocessing improves the signal-to-noise ratio, providing high-quality data for feature matching; the feature matching sub-module associates the real-time spectrum with the voiceprint library and AIS information, realizing accurate tracing from "noise" to "source"; the decision output sub-module makes objective judgments based on thresholds, avoiding human error, and through a feedback mechanism, an adaptive optimization closed loop is formed, which can dynamically adjust the identification strategy for different types of noise sources and continuously improve the identification accuracy.

[0033] For the optimization scheme of the above monitoring system, the edge response module includes a local decision sub-module, a data compression sub-module, a Beidou communication sub-module, and a power management sub-module.

[0034] The input end of the local decision sub-module receives the alarm signal and the trace report of the noise source identification module, and the local decision sub-module determines the noise event level in real time to generate a response instruction and a data packaging strategy.

[0035] The enable end of the data compression sub-module is controlled by the trigger signal of the local decision sub-module, and the data compression sub-module is used for lossless compression of data and packaging of Beidou communication protocol frames.

[0036] The input end of the Beidou communication sub-module receives the output of the data compression sub-module, and the Beidou communication sub-module is used for transmitting compressed data to a satellite and receiving positioning and timing information.

[0037] The state input end of the power management sub-module monitors the working state of the Beidou communication sub-module in real time, and the power management sub-module dynamically controls the power supply state of each sub-module to realize energy efficiency optimization.

[0038] The connection relationship among the sub-modules in the edge response module: the response instruction output end of the local decision sub-module is directly connected to the enable end of the data compression sub-module through a GPIO control line to trigger the compression start; the protocol frame output end of the data compression sub-module is connected to the data input buffer of the Beidou communication sub-module through an SPI bus (clock frequency >= 20MHz); the transmission state signal line of the Beidou communication sub-module is fed back to the voltage regulation control end of the power management sub-module for use in increasing the power supply voltage during transmission; and the power supply on-off control end of the power management sub-module is connected to the power supply pins of each sub-module through a MOSFET switch array for use in cutting off the power supply of unnecessary modules during standby. The edge response module realizes intelligent decision and energy efficiency optimization of the system at the edge side, the local decision sub-module enables the system to determine the event level and respond quickly on site, reduces the decision delay and dependence on the central platform, the data compression and Beidou communication private network transmission greatly save the expensive satellite communication bandwidth and power consumption, and the power management sub-module dynamically regulates the power supply according to the communication state to realize "on-demand power supply", greatly prolongs the continuous working time of the buoy network under the condition of no external power supply, and solves the energy bottleneck problem of long-term ocean monitoring.

[0039] Further optimization of the above scheme, the deployment structure of the buoy network module satisfies:

[0040] 1) The buoy nodes are arranged in a regular hexagonal grid with a node spacing of 5±0.5km;

[0041] 2) Double redundant communication links are established between adjacent nodes:

[0042] Primary link: underwater acoustic communication, carrier frequency 20kHz, bandwidth ±2kHz;

[0043] Backup link: LoRa radio frequency, frequency band 433MHz, transmission power 20dBm;

[0044] 3) Network self-reconstruction protocol: when a node fails, the routing path is automatically rebuilt by the adjacent three nodes, and the recovery time is <=30 seconds.

[0045] Here, the optimization of the deployment scheme brings the core advantages of full coverage and high reliability of the network to the system. The regular hexagonal grid layout realizes seamless coverage of the target sea area with the least number of nodes, and the optimized spacing balances the monitoring accuracy and cost; the double redundant communication links ensure that the network connectivity remains in any single channel failure, and the self-reconstruction protocol enables the network to have self-healing capability, so that a single node failure will not cause the monitoring network to be paralyzed, thereby improving the survivability and reliability of the entire system in harsh marine environments.

[0046] Further optimization of the above scheme also includes an environmental adaptability enhancement architecture, which comprises: 1) a buoy node shell is provided with a bioresistant electrode array, and a ±5V pulse voltage is periodically applied to inhibit barnacle attachment;

[0047] 2) the underwater acoustic sensor array is equipped with an acoustic wave-transparent material protective cover, and the acoustic impedance thereof is matched with the characteristics of seawater (impedance ratio 1.02-1.05);

[0048] 3) a data fusion processing module integrates an environmental compensation algorithm, which automatically corrects the sound wave absorption coefficient when the seawater pH value is less than 7.8;

[0049] 4) a self-reconfigurable communication circuit is built-in with a temperature difference compensation circuit, and the working temperature range is expanded to -20℃ to +55℃.

[0050] Through the multi-dimensional environmental adaptability design of the environmental adaptability enhancement architecture, the present application directly addresses and solves the common problems of marine instrument equipment, such as bioattachment, material corrosion, and environmental parameter fluctuation; the bioresistant electrode array actively inhibits bioattachment and avoids sensor performance degradation; the impedance matching design of the acoustic wave-transparent cover minimizes its interference with sound wave measurement from a passive level; the environmental compensation algorithm ensures the accuracy of monitoring data under different water quality conditions; the temperature difference compensation circuit expands the normal working temperature range of the equipment, ensuring that the system can work stably and reliably in different sea areas from the cold zone to the tropical zone, significantly reducing the maintenance frequency and the total life cycle cost. The environmental compensation algorithm and the temperature difference compensation circuit are both mature technical means.

[0051] Further optimization of the above scheme, the buoy node adopts a double-layer pressure-resistant sealed cabin structure:

[0052] 1) the outer cabin body is a titanium alloy grid skeleton filled with polyurethane foam, and the pressure resistance level is greater than or equal to 10MPa;

[0053] 2) the inner cabin body is divided into a device cabin and a buoyancy adjustment cabin, and the device cabin maintains a constant pressure and temperature environment;

[0054] 3) the outer surface of the cabin body is coated with a bioresistant coating, and the surface energy is less than or equal to 25mN / m.

[0055] The buoy structure design brings excellent pressure resistance, reliability and durability. The outer cabin body filled with polyurethane foam and the titanium alloy skeleton provides high structural strength and pressure resistance, ensuring that the equipment can work normally in a deep sea environment; the inner constant pressure and temperature device cabin provides a safe and stable working environment for precision electronic components; the low surface energy bioresistant coating can effectively reduce marine bioattachment, reduce sailing resistance and maintenance cost, and prolong the on-orbit working time of the equipment. This structure design is the physical basis for ensuring that the entire system can work stably and reliably for a long time.

[0056] Compared with the prior art, the present application has the beneficial effects that:

[0057] 1、The distributed buoy network in the present application adopts a hexagonal grid layout and a double-redundancy self-reconfiguration communication link, ensuring seamless monitoring coverage and extremely high reliability of data transmission, each buoy node integrates an array of underwater acoustic sensors and environmental parameter sensors, and through a hardware synchronous triggering mechanism, precise synchronous acquisition of acoustic data and hydrological data is realized, thereby guaranteeing data quality from the source, and the environmental adaptability enhanced design and double-layer pressure-resistant sealed cabin structure endow the system with the ability to work stably for a long time in complex and harsh marine environments, thereby providing a solid guarantee for obtaining continuous, real and high-quality raw data.

[0058] 2、The system performs spatio-temporal alignment and dimension normalization on multi-source heterogeneous data such as the buoy network, satellite remote sensing and shore-based radar through a professional data fusion processing module, generates a fusion data package with a unified spatio-temporal reference, the noise map generation module performs three-dimensional spatial interpolation and dynamic sound attenuation correction based on a sound propagation physical model, generates a visual noise distribution map with clear physical meaning and high precision, and then the noise source identification module realizes automatic and accurate identification and tracing of noise sources by feature library comparison and AIS information association, thereby converting massive raw monitoring data into accurate intelligence that can directly support decision-making.

[0059] 3、The system is not a passive data collector, but an intelligent terminal with local decision-making capability, the noise source identification module can trigger multi-level alarms in real time after discovering abnormalities, the edge response module starts the localization strategy immediately, losslessly compresses key data and uploads them through Beidou short message in priority, thereby ensuring reliable transmission of early warning information under extreme conditions; the intelligent response mechanism of "end-edge" cooperation greatly shortens the decision-making cycle from discovering abnormalities to starting disposal, and provides unprecedented efficient technical means for marine law enforcement, ecological environment protection, navigation safety maintenance and emergency event response. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 Fig. 1 is a schematic diagram of the overall structure of the marine acoustic environmental monitoring system based on a noise map according to the present application;

[0061] Figure 2 Fig. 2 is a schematic diagram of the structure of a buoy node in the present application;

[0062] Figure 3 Fig. 3 is a schematic diagram of the structure of a data fusion processing module in the present application;

[0063] Figure 4 Fig. 4 is a schematic diagram of the structure of a noise map generation module in the present application;

[0064] Figure 5A structural schematic diagram of a noise source identification module in the application;

[0065] Figure 6 A structural schematic diagram of an edge response module in the application;

[0066] Figure 7 A structural schematic diagram of an external structure of a buoy node in the application.

[0067] In the figure, 1, a buoy network module; 11, a buoy node; 111, an underwater acoustic sensor array; 112, an environmental parameter sensor; 113, a self-reconfigurable communication circuit; 114, an analog multiplexer; 115, an analog-to-digital converter; 116, a node controller; 117, a signal strength comparator; 118, an outer cabin; 119, an inner cabin; 2, a data fusion processing module; 21, a multi-source data receiving submodule; 22, a space-time alignment submodule; 23, a normalized output submodule; 3, a noise map generation module; 31, a data buffering submodule; 32, a grid processing submodule; 33, a rendering output submodule; 4, a noise source identification module; 41, a signal preprocessing submodule; 42, a feature matching submodule; 43, a decision output submodule; 5, an edge response module; 51, a local decision submodule; 52, a data compression submodule; 53, a Beidou communication submodule; 54, a power management submodule; 61, a first data bus; 62, a second data bus; 63, a third data bus. DETAILED DESCRIPTION

[0068] As Figure 1 shown, the ocean acoustic environment monitoring system based on a noise map in the application comprises: a buoy network module 1: composed of at least 6 buoy nodes 11, each buoy node 11 serving as a distributed data acquisition unit, acquiring underwater noise signals through a built-in underwater acoustic sensor array 111, obtaining temperature, salinity, depth and seawater flow rate data through an environmental parameter sensor 112, and dynamically switching transmission links according to channel quality and networking transmission of multi-source information to a data fusion processing module 2 by a self-reconfigurable communication circuit 113.

[0069] In the application, the deployment structure of the buoy network module 1 satisfies:

[0070] 1) The buoy nodes 11 are arranged in a regular hexagonal grid, and the node spacing is 5±0.5 km;

[0071] 2) Double-redundant communication links are established between adjacent nodes: the main link is underwater acoustic communication, and the backup link is LoRa radio frequency;

[0072] 3) Network self-reconfigurable protocol: when a node fails, the route path is automatically rebuilt by the adjacent 3 nodes, and the recovery time is ≤30 seconds.

[0073] Referring to Figure 7In the design of the buoy node, the buoy node 11 adopts a double-layer pressure-resistant sealed cabin structure: 1) the outer cabin body 118 is a titanium alloy grid skeleton filled with polyurethane foam;

[0074] 2) the inner cabin body 119 is divided into an equipment cabin and a buoyancy adjusting cabin, and the equipment cabin maintains a constant pressure and temperature environment;

[0075] 3) the outer surface of the cabin body is coated with a bioresistant coating.

[0076] The application also comprises a data fusion processing module 2: the buoy network module 1 is connected through a first data bus 61 and receives buoy network data, wherein the buoy network data comprises underwater noise signals, temperature-salinity-depth and seawater flow rate data, satellite remote sensing data is received through a satellite data interface, and shore-based radar data is accessed through a radar input port; the data fusion processing module 2 performs time-space alignment and dimension normalization on the buoy network data, satellite remote sensing data and shore-based radar data to generate a fusion data package with a unified time-space reference.

[0077] The application also comprises a noise map generation module 3: the output end of the data fusion processing module 2 is connected through a second data bus 62; the noise map generation module 3 performs three-dimensional space interpolation and sound attenuation dynamic correction on the received fusion data package based on a sound propagation physical model to generate a three-dimensional noise distribution map.

[0078] The application also comprises a noise source identification module 4: the output end of the noise map generation module 3 is connected through a third data bus 63; the noise source identification module 4 compares the frequency spectrum features of the received three-dimensional noise distribution map with a soundprint feature library, associates ship AIS information to identify noise source types, and triggers an alarm when the noise exceeds a threshold; the output end of the noise source identification module 4 is connected to a ship database through an AIS data interface.

[0079] The application also comprises an edge response module 5: in communication connection with the noise source identification module 4 and connected to the self-reconfigurable communication circuit 113 of each buoy node 11 through a distributed control line; the edge response module 5 starts a local response strategy according to the noise alarm level, optimizes transmission efficiency through a built-in data compression unit, and uploads key data through a Beidou short message transmitter.

[0080] Referring to Figure 2 , the underwater acoustic sensor array 111, the environmental parameter sensor 112 and the self-reconfigurable communication circuit 113 contained in the buoy node 11 form a signal processing link through the following relationship:

[0081] 1) the output end of the underwater acoustic sensor array 111 is connected to the analog input channel of the analog-to-digital converter 115 through the analog multiplexer 114, and the digital output end of the analog-to-digital converter 115 is connected to the ADC data interface of the node controller 116;

[0082] 2) The environmental parameter sensor 112 is connected to the serial communication interface of the node controller 116 in parallel through the RS-485 bus, and the synchronous clock output end of the environmental parameter sensor 112 is directly connected to the sampling clock pin of the analog-to-digital converter 115;

[0083] 3) The input end of the self-reconfigurable communication circuit 113 receives the fusion data packet sent by the node controller 116 through the UART serial port, the channel switching control end is connected to the GPIO output pin of the signal strength comparator 117, and the signal strength comparator 117 monitors the signal-to-noise ratio of the underwater acoustic and wireless communication channels in real time.

[0084] In the present application, the hardware modules of the buoy node are connected as follows:

[0085] The underwater acoustic sensor array 111 is arranged in a regular tetrahedron topology with a spacing of 120 mm by four piezoelectric ceramic hydrophones (frequency response 10 Hz-20 kHz); the output end of each hydrophone is connected to a ±12V DC bias circuit, and an instrument amplifier is connected in series at the back end; the output of the instrument amplifier is connected to an 8-channel analog multiplexer 114 through a shielded twisted pair; the environmental parameter sensor 112 is integrated with a temperature-salinity-depth sensor and an acoustic Doppler current profiler in the same waterproof cabin, and both are connected to a data acquisition card in parallel through an RS-485 bus; the bus is provided with a synchronous clock generator; the output end of the data acquisition card is connected to the node controller 116 through an SPI interface; the self-reconfigurable communication circuit 113 includes an underwater acoustic communication unit: carrier frequency 20 kHz, including a transmitting transducer and a receiving detection circuit; a wireless radio frequency unit: LoRa module, frequency band 433 MHz, transmitting power 20 dBm; and a channel switching controller: based on the signal strength comparator 117 (threshold -100 dBm) to drive a two-way RF switch to select the path.

[0086] The 4 piezoelectric ceramic hydrophones of the underwater acoustic sensor array 111 are arranged in a regular tetrahedron topology, the output ends of each hydrophone are connected to an instrument amplifier after being powered by a ±12V DC bias circuit, the amplified analog signals are connected to the analog input pin of the analog-to-digital converter 115 through shielded twisted pair lines and an 8-channel analog multiplexer 114; the temperature-salinity-depth sensor of the environmental parameter sensor 112 and the acoustic Doppler current profiler are connected in parallel to the data acquisition card through the RS-485 bus, the data acquisition card is connected to the node controller 116 through the SPI bus, and the synchronous clock generator built-in the data acquisition card outputs a 1kHz square wave to the sampling clock end of the analog-to-digital converter 115; the digital output end of the analog-to-digital converter 115 is connected to the ADC data interface of the node controller 116, and the node controller 116 transmits the acoustic data frame fused with the attitude compensation parameter to the underwater acoustic communication unit of the self-reconfigurable communication circuit 113 through the UART serial port; the signal strength comparator 117 of the self-reconfigurable communication circuit 113 monitors the channel quality of the underwater acoustic communication unit and the wireless radio frequency unit in real time, when the signal-to-noise ratio of the underwater acoustic channel is lower than 10dB, the signal strength comparator 117 drives the two-way RF switch to switch to the LoRa radio frequency unit through the GPIO control line, and the switching instruction is fed back to the power consumption manager through the enable end to turn off the power supply of the idle communication unit.

[0087] In the present application, the temperature-salinity-depth sensor (CTD, Conductivity, Temperature, Depth Sensor) is the core instrument in marine environment monitoring, which mainly synchronously measures the following three basic physical properties of seawater, and the data detected by it is collectively referred to as temperature-salinity-depth.

[0088] Temperature (Temperature):

[0089] Measurement content: In-situ temperature of seawater, usually in Celsius (℃).

[0090] Importance: Temperature is a key parameter affecting marine physical, chemical and biological processes, which directly affects the propagation speed of sound waves in seawater, and is an indispensable input data for building high-precision underwater acoustic models.

[0091] Conductivity (Conductivity):

[0092] Measurement content: Conductivity of seawater, usually in Siemens / meter (S / m).

[0093] Importance: Conductivity is mainly used to calculate another more critical parameter, salinity (Salinity). The unit of salinity is usually practical salinity unit, dimensionless, and salinity also significantly affects the density and sound speed of seawater.

[0094] Pressure:

[0095] Content of measurement: static pressure of seawater at the location of sensor, usually in units of decibar (dbar) or megapascal (MPa).

[0096] Importance: the pressure value directly corresponds to the depth of the sensor, which is the basic coordinate for determining the spatial position of measurement data, conducting three-dimensional sound field reconstruction and profile analysis.

[0097] Referring to Figure 3 , the data fusion processing module 2 comprises a multi-source data receiving submodule 21, a space-time alignment submodule 22 and a normalized output submodule 23.

[0098] The multi-source data receiving submodule 21 is used to receive satellite remote sensing data, shore-based radar data and buoy network data in real time, complete physical layer signal demodulation and protocol analysis, and transmit the original data stream from the output end of the multi-source data receiving submodule 21 to the space-time alignment submodule 22 through a high-speed serial bus.

[0099] The space-time alignment submodule 22 adds a unified timestamp to the multi-source data based on the GPS timing signal, converts the spatial coordinates to the WGS-84 coordinate system and generates space-time synchronization data, the input end of the space-time alignment submodule 22 receives the output data of the multi-source data receiving submodule 21, and the output end of the space-time alignment submodule 22 sends the space-time synchronization data to the normalized output submodule 23 through a PCIe high-speed interface.

[0100] The data fusion between the submodules of the data fusion processing module 2 is realized through the following relationship:

[0101] The multi-source data receiving submodule 21 comprises:

[0102] Satellite data interface: RJ45 port receives satellite remote sensing data, TCP / IP protocol, baud rate 1Gbps;

[0103] Shore-based radar input port: BNC interface receives analog radar signal, frequency band X-band;

[0104] Buoy network data bus: RS-422 differential bus receives buoy node data, baud rate 115.2kbps;

[0105] The space-time alignment submodule 22 comprises:

[0106] FPGA timing controller: connected to the GPS timing module through LVDS differential line, accuracy ±100ns;

[0107] Timestamp alignment circuit: adds UTC time label to each source data, resolution 1μs;

[0108] Spatial grid register: store the geographic coordinate mapping parameters, WGS-84 coordinate system;

[0109] The normalization output submodule 23 comprises:

[0110] Data format converter: convert analog radar signals into digital quantities through ADC;

[0111] Protocol conversion chip: parse satellite TCP / IP data packets into unified RS-485 output;

[0112] Multiplexing output bus: 32-bit parallel bus, clock frequency 100MHz, connected to the noise map generation module 3.

[0113] The RS-422 output end of the multi-source data receiving submodule 21 is connected to the spatial grid register of the space-time alignment submodule 22 through a dual-port RAM; the time stamp alignment circuit of the space-time alignment submodule 22 is connected to the data format converter queue of the normalization output submodule 23 through a PCIe x4 interface; the multiplexing output bus of the normalization output submodule 23 is directly connected to the DMA controller of the noise map generation module 3, and the output end of the normalization output submodule 23 outputs the fusion data packet to the noise map generation module 3 through a parallel data bus.

[0114] In the present application, the fusion data packet is essentially a structured data set with unified space-time reference, standardized format and dimension. It mainly includes the following several categories of data:

[0115] 1. Acoustic monitoring data: original acoustic signals or processed data from the hydrophone sensor array of each buoy node.

[0116] 2. Ocean environmental hydrological data: synchronous measurement results from the environmental parameter sensors of each buoy node, which are the key to correcting the acoustic model. For example, temperature-salinity-depth data, weather data, etc.

[0117] 3. Space-time reference data: provides a unified space-time coordinate for all data, which is the premise of multi-source data fusion. For example, time stamp, latitude and longitude coordinates of the buoy node, latitude and longitude coordinates of the satellite remote sensing data pixel, and geographical position of the radar monitoring target.

[0118] 4. Auxiliary source data: macro-environmental information from satellites and shore-based systems, used to supplement and verify buoy network data. For example, satellite remote sensing data, shore-based radar data.

[0119] 5. Metadata and status information: information describing the state of the data itself and the running state of the system. Such as signal-to-noise ratio, sensor health status, data validity flag bit, etc.

[0120] 6. Target and event identification data: target information from AIS database or radar.

[0121] Referring to Figure 4 The noise map generation module 3 comprises a data buffering submodule 31, a grid processing submodule 32, and a rendering output submodule 33. The data buffering submodule 31 receives and buffers the fusion data packets output by the data fusion processing module 2, performs traffic shaping and priority scheduling, and provides a data stream to the grid processing submodule 32 through a high-speed data transmission channel.

[0122] The grid processing submodule 32 performs spatial interpolation calculation based on a sound propagation physical model and dynamically corrects sound attenuation errors. The input end of the grid processing submodule 32 is connected to the output end of the data buffering submodule 31, and the output end of the grid processing submodule 32 sends noise data to the rendering output submodule 33 through a PCIe high-speed interface. In the present application, the direct object and core object of the spatial interpolation calculation performed by the grid processing submodule are the sound pressure level or noise intensity data measured by each discrete buoy node.

[0123] The noise map generation module 3 achieves noise map generation through a connection relationship between the submodules.

[0124] The data buffering submodule 31 comprises:

[0125] A dual-port RAM memory receives the output of the data fusion processing module through a 32-bit parallel bus;

[0126] A DMA controller manages data transfer to the grid processing submodule, with a bandwidth of ≥3.2Gbps;

[0127] The grid processing submodule 32 comprises:

[0128] An FPGA interpolation processor, model Xilinx XC7VX690T;

[0129] A built-in sound propagation compensation circuit that corrects sound attenuation based on a temperature-salinity profile;

[0130] An integrated spatial interpolator, a Kriging algorithm hardware acceleration core;

[0131] A sound velocity gradient correction circuit connected to the temperature-salinity sensor data input, with a correction range of ±5dB;

[0132] The rendering output submodule 33 comprises:

[0133] A color scale mapping unit that converts noise intensity values to RGB color codes, with a resolution of 24 bits;

[0134] A time axis superimposer that fuses Beidou timing signals, with a time tag accuracy of ±1ms;

[0135] HDMI output interface: transmit noise map to display device, resolution 4K@30Hz.

[0136] The output end of the DMA controller of the data buffering sub-module 31 is connected to the input FPGA interpolation processor of the grid processing sub-module 32 through a 64-bit high-speed bus; the interpolation result output end of the grid processing sub-module 32 is connected to the color index mapping unit of the rendering output sub-module 33 through a PCIe x8 interface; the control signal line of the rendering output sub-module 33 is fed back to the dual-port RAM memory of the data buffering sub-module 31, triggering a data emptying instruction.

[0137] Referring to Figure 5 , the noise source identification module 4 comprises a signal preprocessing sub-module 41, a feature matching sub-module 42 and a decision output sub-module 43.

[0138] The input end of the signal preprocessing sub-module 41 is connected to the data bus of the noise map generation module 3, and the signal preprocessing sub-module 41 performs band selection and dynamic range compression on the input acoustic signal, and outputs a standard spectrum.

[0139] The feature matching sub-module 42 is used for comparing the preprocessed standard spectrum with a voiceprint feature library, calculating a similarity and associating AIS ship information, the input end of the feature matching sub-module 42 receives the output of the signal preprocessing sub-module 41, and the AIS interface of the feature matching sub-module 42 is connected to a ship database through a serial bus.

[0140] The noise source identification module 4 is connected between the sub-modules through the following connection relationship to realize noise source identification:

[0141] The signal preprocessing sub-module 41 comprises:

[0142] Band pass filter array: 4 groups of adjustable filters, frequency band range 10Hz-100kHz, input stage connected to the sound pressure data bus of the noise map generation module;

[0143] Log compression circuit: convert sound pressure linear value to dBμPa unit, dynamic range 120dB;

[0144] Spectrum analysis unit: FFT processor, point number 4096, output amplitude spectrum;

[0145] The feature matching sub-module 42 comprises:

[0146] Voiceprint feature library memory: NOR Flash array, capacity 8GB, storing biological / ship noise spectrum templates;

[0147] Hardware matching engine: FPGA-based parallel correlator, calculating the similarity between input spectrum and template;

[0148] AIS data interface: RJ45 port connecting ship database;

[0149] The decision output submodule 43 comprises:

[0150] Threshold comparator array: 8-way voltage comparator, determining noise type;

[0151] Alarm signal generator: triggering pulse signal when ship noise exceeds 85dB;

[0152] Traceability report generator: outputting noise source position and ship ID to display terminal.

[0153] The output end of the spectrum analysis unit of the signal preprocessing submodule 41 accesses the input buffer of the feature matching submodule 42 through a 32-bit LVDS bus; the output end of the feature matching submodule 42 is connected to the threshold comparator array of the decision output submodule 43 through a PCIe x4 interface; the AIS data interface of the decision output submodule 43 is directly connected to the feature matching submodule 42 through an SPI bus; and the enable end of the alarm signal generator is fed back to the band-pass filter array of the signal preprocessing submodule 41, triggering frequency band switching.

[0154] Referring to Figure 6 , the edge response module 5 comprises a local decision submodule 51, a data compression submodule 52, a Beidou communication submodule 53 and a power management submodule 54.

[0155] The input end of the local decision submodule 51 receives the alarm signal and traceability report of the noise source identification module 4, the local decision submodule 51 determines the noise event level in real time, generates a response instruction and a data packaging strategy; the enable end of the data compression submodule 52 is controlled by the trigger signal of the local decision submodule 51, the data compression submodule 52 is used for performing lossless compression on feature data and packaging Beidou communication protocol frames; the input end of the Beidou communication submodule 53 receives the output of the data compression submodule 52, the Beidou communication submodule 53 is used for transmitting compressed data to a satellite and receiving positioning and timing information; the state input end of the power management submodule 54 monitors the working state of the Beidou communication submodule 53 in real time, the power management submodule 54 dynamically regulates the power supply state of each submodule, and energy efficiency optimization is realized.

[0156] The edge response between the submodules of the edge response module 5 is realized through the following connection relationship:

[0157] The local decision submodule 51 comprises:

[0158] Threshold comparator array: 4-way voltage comparator, determining noise level in real time;

[0159] Event priority encoder: assign response level based on noise type: ship > biology > environment;

[0160] Control register: store response strategy parameters;

[0161] The data compression submodule 52 comprises:

[0162] Feature extraction circuit: DCT-based hardware compression core, compression ratio ≥ 5:1;

[0163] Circular buffer: SRAM memory, cache data to be transmitted;

[0164] Frame encapsulator: add Beidou protocol header, including position / time / noise level;

[0165] The Beidou communication submodule 53 comprises:

[0166] RDSS transmitter: operating frequency band 1610-1626MHz, output power 10W;

[0167] Modulation controller: BPSK modulation chip, symbol rate 8.4kbps;

[0168] Loop antenna: impedance 50Ω, gain ≥ 3dBi;

[0169] The power management submodule 54 comprises:

[0170] Dynamic voltage regulation circuit: switch 3.3V / 5V / 12V output according to load;

[0171] Surge protector: TVS diode array, clamping voltage 15V;

[0172] Acoustic wake-up detector: sensitivity -70dB, standby power consumption ≤0.1W.

[0173] The threshold comparator array of the local decision submodule 51 is directly connected to the start pin of the data compression submodule 52 through a GPIO control line; the output end of the feature extraction circuit of the data compression submodule 52 is connected to the modulation controller of the Beidou communication submodule 53 through an SPI bus; the RDSS transmitter of the Beidou communication submodule 53 is fed back to the dynamic voltage regulation circuit of the power management submodule 54; the acoustic wake-up detector of the power management submodule 54 is connected to the power input of each submodule through a MOSFET switch tube, and the on-resistance thereof is ≤10mΩ.

[0174] In the present application, an environmental adaptability enhancement architecture is also included, which comprises:

[0175] 1) The buoy node 11 shell is provided with a bioresistant electrode array, and a ±5V pulse voltage is applied periodically to inhibit barnacle attachment;

[0176] 2) The water acoustic sensor array 111 is equipped with an acoustic wave-transparent material protective cover with an acoustic impedance matching the characteristics of seawater;

[0177] 3) The data fusion processing module 2 integrates an environmental compensation algorithm, which automatically corrects the sound wave absorption coefficient when the seawater pH value is less than 7.8;

[0178] 4) The self-reconfigurable communication circuit 113 is built-in with a temperature difference compensation circuit, and the working temperature range is expanded to -20℃ to +55℃.

[0179] The working principle of the present application is as follows:

[0180] The system of the present application starts from a monitoring network composed of multiple buoy nodes 11, each node synchronously collects underwater noise signals through a water acoustic sensor array 111, and obtains hydrological data such as temperature, salinity, depth, flow rate, etc. in real time through an environmental parameter sensor 112. The multi-source information collected is dynamically selected by the self-reconfigurable communication circuit 113 built-in the buoy node according to the real-time quality of the water acoustic and wireless channels, and is transmitted to the corresponding processing unit through networking. The data fusion processing module 2 receives heterogeneous data from the buoy network, satellite remote sensing and shore-based radar, first performs physical layer demodulation and protocol analysis, then completes the spatio-temporal alignment of multi-source data based on GPS timing and WGS-84 coordinate system, and finally generates a unified standard fusion data packet through dimensionless processing. The data packet is delivered to the noise map generation module 3, which performs three-dimensional spatial interpolation and sound attenuation dynamic correction on the discrete monitoring point data based on the sound propagation physical model, generating a high-precision three-dimensional noise distribution map. The noise source identification module 4 extracts the spectral features of the noise map, compares them with the acoustic fingerprint feature library, and associates them with the AIS ship database information, to realize the type identification and tracing of the noise source. When the identified noise intensity exceeds the preset threshold, the module immediately triggers an alarm. The edge response module 5 starts the local response strategy according to the alarm level, losslessly compresses and encapsulates the key feature data into Beidou communication protocol frames, and uploads them to the monitoring center through the Beidou short message transmitter, and at the same time adjusts the acquisition and communication strategy of the buoy node 11 through the distributed control line in the reverse direction, to complete the intelligent closed loop from perception, analysis to decision execution.

[0181] Example one:

[0182] The system is deployed in an important shipping channel in the East China Sea. First, 12 buoy nodes are arranged in a regular hexagonal grid. The hydrophone array of a buoy node collects a continuous broadband noise with a center frequency of 187 Hz under the trigger of a synchronous clock signal. The sound pressure level is 23 dB higher than the background noise. The environmental parameter sensor synchronously records the water temperature as 17.8℃ and the salinity as 33.5‰. The self-reconfigurable communication circuit automatically switches to the 433 MHz LoRa backup link due to the decrease of the signal-to-noise ratio of the underwater acoustic channel to 8 dB, and transmits the data packet to the shore-based center station. Subsequently, the data fusion processing module receives the buoy data, AIS database information (showing that there is a container ship passing by at a speed of 18 knots 10 km away) and remote sensing sea surface temperature data. All data are unified to the WGS-84 coordinate system and added with UTC time stamps, and after normalization, the fusion data packet is generated. The noise map generation module generates a three-dimensional map showing the noise anomaly area around the ship based on the sound propagation model, and the highest noise level is 125 dB. Then, the noise source identification module extracts the line spectrum features of the noise, and after comparison with the voiceprint feature library, the spectral features of the 7-blade large skew propeller match the ship model in the AIS information with a matching degree of 92%, confirming that the container ship is the main noise source. Since the noise value exceeds the set threshold (120 dB), the module triggers a secondary alarm. After receiving the alarm, the edge response module starts local decision-making, and after lossless compression of the key spectral features and ship ID information, it is reported to the maritime supervision platform through the Beidou short message transmitter, providing real-time law enforcement basis for the supervision department.

[0183] Example two

[0184] In a certain sea area of the South China Sea, the system network encountered a typhoon. Initially, the severe sea conditions caused by the typhoon caused the attitude of multiple buoy nodes to change dramatically. The environmental adaptability enhancement module was immediately started due to the temporary signal attenuation caused by biological attachment on node No. 4. The anti-bioattachment electrode array on the buoy shell applied a -5V pulse voltage to inhibit barnacle attachment, while the temperature compensation circuit started to work, ensuring the stable operation of the circuit at a high temperature of 29℃. During the process, the anchor line of buoy No. 7 was broken and displaced due to the large waves caused by the typhoon. The adjacent nodes 6, 8 and 9 detected the failure of node No. 7 through the self-reconfiguring communication protocol, and automatically rebuilt the underwater acoustic communication routing path in 28 seconds, bypassing the fault point, transmitting data to the fusion processing center through the backup link, ensuring the integrity of the network topology and the continuity of the data. The data fusion processing module synchronously received the typhoon path data from the weather satellite and processed it as the highest priority data stream. Finally, the noise map generation module dynamically adjusted the sea surface noise source level parameter in the sound propagation model after fusing the typhoon wave noise data. Although the background noise increased significantly due to the typhoon, the grid processing submodule effectively separated the environmental noise from the abnormal signal through the sound ray tracing algorithm, accurately drawing the noise baseline map of the entire sea area during the typhoon passage; The power management submodule of the edge response module monitored the increase in communication load and automatically increased the power supply voltage by 500mV to ensure the transmission power of the Beidou communication submodule under harsh conditions, successfully transmitting the key acoustic and environmental data during the entire typhoon event back to the system, providing valuable data sets for studying the impact of typhoons on the marine acoustic environment.

[0185] The present application constructs an ocean acoustic environment monitoring system integrating intelligent sensing, accurate analysis and active response. The system cooperatively collects multi-source data through a distributed buoy network, generates a high-precision three-dimensional noise map after deep fusion processing, and realizes accurate tracing and classification of noise sources relying on intelligent identification algorithms. Finally, the system dynamically optimizes the monitoring strategy and data transmission through the edge response mechanism, forming a closed-loop control from environmental perception to intelligent decision-making, significantly improving the real-time, reliability and automation level of marine environmental monitoring.

[0186] It should be noted that the above examples are only used to illustrate the technical solutions of the present application and not to limit the technical solutions. Although the applicant has made a detailed description of the present application with reference to the preferred embodiments, those skilled in the art should understand that modifications or equivalent replacements of the technical solutions of the present application cannot deviate from the purpose and scope of the present application, and should be covered in the scope of the claims of the present application.

Claims

1. A marine acoustic environment monitoring system based on noise maps, characterized in that, include: The buoy network module (1) consists of at least 6 buoy nodes (11). Each buoy node (11) serves as a distributed data acquisition unit. It acquires underwater noise signals through a built-in underwater acoustic sensor array (111), obtains temperature, salinity, depth and seawater flow velocity data through an environmental parameter sensor (112), and dynamically switches the transmission link according to the channel quality through a self-reconfigurable communication circuit (113) and transmits multi-source information to the data fusion processing module (2). The data fusion processing module (2) is connected to the buoy network module (1) through the first data bus (61) and receives buoy network data. It receives satellite remote sensing data through the satellite data interface and accesses shore-based radar data through the radar input port. The data fusion processing module (2) performs spatiotemporal alignment and dimension normalization on the buoy network data, satellite remote sensing data and shore-based radar data to generate a fusion data packet with a unified spatiotemporal reference. Noise map generation module (3): Connected to the output of data fusion processing module (2) via second data bus (62), the noise map generation module (3) performs three-dimensional spatial interpolation and dynamic correction of sound attenuation on the received fusion data packet based on the sound propagation physical model to generate a three-dimensional noise distribution map; Noise source identification module (4): Connected to the output of noise map generation module (3) via third data bus (63). The noise source identification module (4) compares the spectral characteristics of the received three-dimensional noise distribution map with the acoustic signature feature library, identifies the noise source type by associating it with ship AIS information, and triggers an alarm when the noise exceeds the threshold. The output of the noise source identification module (4) is connected to the ship database via AIS data interface. Edge response module (5): Communicatively connected to the noise source identification module (4) and connected to the self-reconfigurable communication circuit (113) of each buoy node (11) via distributed control line. The edge response module (5) activates the local response strategy according to the noise alarm level.

2. The marine acoustic environment monitoring system based on noise maps according to claim 1, characterized in that, The underwater acoustic sensor array (111), environmental parameter sensor (112), and self-reconfigurable communication circuit (113) included in the buoy node (11) constitute a signal processing link through the following relationship: 1) The output of the underwater acoustic sensor array (111) is connected to the analog input channel of the analog-to-digital converter (115) through an analog multiplexer (114), and the digital output of the analog-to-digital converter (115) is connected to the ADC data interface of the node controller (116). 2) The environmental parameter sensor (112) is connected in parallel to the serial communication interface of the node controller (116) via an RS-485 bus, and the synchronous clock output terminal of the environmental parameter sensor (112) is directly connected to the sampling clock pin of the analog-to-digital converter (115). 3) The input terminal of the self-reconfigurable communication circuit (113) receives the fused data packet sent by the node controller (116) through the UART serial port. Its channel switching control terminal is connected to the GPIO output pin of the signal strength comparator (117). The signal strength comparator (117) monitors the signal-to-noise ratio of the underwater acoustic and wireless communication channels in real time. The synchronous clock signal of the environmental parameter sensor (112) triggers the analog-to-digital converter (115) to synchronously acquire the underwater acoustic signal; the node controller (116) transmits the data to the self-reconfigurable communication circuit module (113) via the UART serial port; when the signal strength comparator (117) detects that the signal-to-noise ratio of the underwater acoustic channel is <10dB, it switches the self-reconfigurable communication circuit (113) to the wireless radio frequency channel via the GPIO control line.

3. The marine acoustic environment monitoring system based on noise maps according to claim 1, characterized in that, The data fusion processing module (2) includes a multi-source data receiving submodule (21), a spatiotemporal alignment submodule (22), and a normalization output submodule (23); The multi-source data receiving submodule (21) is used to receive satellite remote sensing data, shore-based radar data and buoy network data in real time, and to complete physical layer signal demodulation and protocol parsing. The output of the multi-source data receiving submodule (21) transmits the original data stream to the spatiotemporal alignment submodule (22) through a high-speed serial bus. The spatiotemporal alignment submodule (22) adds a unified timestamp to the multi-source data based on the GPS timing signal, transforms the spatial coordinates to the WGS-84 coordinate system and generates spatiotemporal synchronization data. The input end of the spatiotemporal alignment submodule (22) receives the output data of the multi-source data receiving submodule (21), and the output end of the spatiotemporal alignment submodule (22) sends the spatiotemporal synchronization data to the normalization output submodule (23) via the PCIe high-speed interface. The normalization output submodule (23) is used to convert heterogeneous data formats into a unified transmission protocol and perform dimensional normalization processing. The input end of the normalization output submodule (23) receives the output data of the spatiotemporal alignment submodule (22), and the output end of the normalization output submodule (23) outputs fused data packets to the noise map generation module (3) through a parallel data bus.

4. The marine acoustic environment monitoring system based on noise maps according to claim 1, characterized in that, The noise map generation module (3) includes a data buffer submodule (31), a grid processing submodule (32), and a rendering output submodule (33). The data buffer submodule (31) receives and caches the fusion data packets output by the data fusion processing module (2), performs traffic shaping and priority scheduling, and the output end of the data buffer submodule (31) provides a data stream to the grid processing submodule (32) through a high-speed data transmission channel. The mesh processing submodule (32) performs spatial interpolation calculations based on the physical model of sound propagation and dynamically corrects the sound attenuation error. The input end of the mesh processing submodule (32) is connected to the output end of the data buffer submodule (31), and the output end of the mesh processing submodule (32) sends noise data to the rendering output submodule (33) via the PCIe high-speed interface. The rendering output submodule (33) is used to convert noise data into a visual map and overlay time coordinates and color mark information. The input end of the rendering output submodule (33) receives the output data of the grid processing submodule (32), and the output end of the rendering output submodule (33) outputs a three-dimensional noise distribution map through the video interface.

5. The marine acoustic environment monitoring system based on noise maps according to claim 1, characterized in that, The noise source identification module (4) includes a signal preprocessing submodule (41), a feature matching submodule (42), and a decision output submodule (43); The input terminal of the signal preprocessing submodule (41) is connected to the data bus of the noise map generation module (3). The signal preprocessing submodule (41) performs frequency band selection and dynamic range compression on the input acoustic signal and outputs a standard spectrum. The feature matching submodule (42) is used to compare the preprocessed standard spectrum with the voiceprint feature library, calculate the similarity and associate it with AIS ship information. The input end of the feature matching submodule (42) receives the output of the signal preprocessing submodule (41). The AIS interface of the feature matching submodule (42) is connected to the ship database through a serial bus. The input terminal of the decision output submodule (43) receives the matching result of the feature matching submodule (42). The decision output submodule (43) determines the noise source type according to the spectrum similarity threshold and generates an alarm signal and a source tracing report.

6. The marine acoustic environment monitoring system based on noise maps according to claim 1, characterized in that, The edge response module (5) includes a local decision submodule (51), a data compression submodule (52), a Beidou communication submodule (53), and a power management submodule (54); The input terminal of the local decision submodule (51) receives the alarm signal and source tracing report from the noise source identification module (4). The local decision submodule (51) determines the noise event level in real time and generates response instructions and data encapsulation strategies. The enable terminal of the data compression submodule (52) is controlled by the trigger signal of the local decision submodule (51). The data compression submodule (52) is used to perform lossless compression on the data and encapsulate the Beidou communication protocol frame. The input end of the Beidou communication submodule (53) receives the output of the data compression submodule (52). The Beidou communication submodule (53) is used to transmit compressed data to the satellite and receive positioning and timing information. The power management submodule (54) monitors the working status of the Beidou communication submodule (53) in real time at its status input terminal, and dynamically adjusts the power supply status of each submodule to achieve energy efficiency optimization.

7. The marine acoustic environment monitoring system based on noise maps according to claim 1, characterized in that, The deployment structure of the buoy network module (1) satisfies: 1) The buoy nodes (11) are arranged in a regular hexagonal grid with a node spacing of 5±0.5km; 2) Establish dual-redundant communication links between adjacent nodes: Main link: underwater acoustic communication; Backup link: LoRa radio frequency; 3) Network self-reconstruction protocol: When a node fails, the routing path is automatically reconstructed by the three adjacent nodes, and the recovery time is ≤30 seconds.

8. The marine acoustic environment monitoring system based on noise maps according to claim 1, characterized in that, It also includes an environmentally adaptive enhancement architecture, which comprises: 1) The outer shell of the buoy node (11) is equipped with an anti-bioattachment electrode array, and a ±5V pulse voltage is periodically applied to suppress barnacle attachment; 2) The underwater acoustic sensor array (111) is equipped with an acoustically transparent material protective cover, whose acoustic impedance matches the characteristics of seawater; 3) Data fusion processing module (2) integrates environmental compensation algorithm, automatically corrects the sound absorption coefficient when the seawater pH value is <7.8; 4) The self-reconfigurable communication circuit (113) has a built-in temperature difference compensation circuit, and the operating temperature range is extended to -20℃~+55℃.

9. The marine acoustic environment monitoring system based on noise maps according to claim 2, characterized in that, The buoy node (11) adopts a double-layer pressure-resistant sealed chamber structure: 1) The outer cabin (118) is a titanium alloy mesh skeleton filled with polyurethane foam; 2) The inner cabin (119) is divided into an equipment cabin and a buoyancy regulating cabin. The equipment cabin maintains a constant pressure and temperature environment. 3) The outer surface of the cabin is coated with an anti-bioadhesion coating.

Images