Syndrome non-cooperative underwater acoustic communication signal receiving and processing device and method

Abstract

The invention relates to a syndrome non-cooperative underwater acoustic communication signal receiving and processing device and method. The syndrome non-cooperative underwater acoustic communication signal receiving and processing device comprises a computing resource platform and an intelligent data processing platform, the computing resource platform comprises a VPX architecture host and KVM equipment, and an intelligent data processing platform is written into the computing resource platform; the intelligent data processing platform comprises a signal real-time processing module and a training management module; the signal real-time processing module comprises a data processing and forwarding unit, a signal data processing unit and a data display unit; the signal data processing unit comprises an algorithm library and algorithm configuration; the algorithm library comprises a beam forming algorithm, an identification algorithm, a detection algorithm, a demodulation algorithm and a decoding algorithm; the training management module comprises an annotation training unit, a data management unit and a system configuration unit. According to the invention, high-efficiency standardized high-computing-power hardware equipment is designed, a high-computing-power artificial intelligence method is used in combination with the hardware equipment, typical underwater acoustic communication signal acquisition equipment is adapted, and accompanying non-cooperative underwater acoustic communication signal receiving processing is completed.

Description

Technical Field

[0001] The present invention relates to the field of underwater acoustic communication signal receiving and processing, and particularly to an accompanying non-cooperative underwater acoustic communication signal receiving and processing device and method. Background Art

[0002] Currently, there are various limitations in the equipment that can perform real-time underwater acoustic signal processing along with underwater acoustic signal acquisition equipment. Preset underwater acoustic signal acquisition equipment is usually equipped with offline processing equipment. In underwater acoustic signal processing, due to problems such as multipath effect, time-varying effect, narrow available frequency band, and severe signal attenuation in the underwater acoustic channel, the underwater acoustic signal processing is complex. The preset underwater acoustic signal acquisition equipment is limited by its own processing capacity and design architecture and is difficult to process the collected data in real time; For the equipment that can perform real-time acquisition but remote data transmission, the underwater communication rate is much lower than that of wired communication. Common underwater acoustic communication methods use spread spectrum communication technology, etc. Even so, for remote transmission, such as through satellite or shortwave relay, etc., it will still be affected by factors such as transmission bandwidth, transmission distance, and signal interference, thus affecting the real-time performance of the onshore processing equipment.

[0003] For the equipment that can perform real-time acquisition and direct processing along with the boat, although it can obtain the underwater acoustic communication situation in the sea area where the boat is located in real time, determine the azimuth of the underwater sound source target, and obtain the signal time-frequency information, under non-cooperative conditions, the underwater acoustic communication signal may have complex modulation methods, coding formats, etc., and the equipment for direct processing along with the boat may not have the processing algorithms and capabilities for these situations. At the same time, limited by the hardware resources, space, power, etc. on the boat, it is difficult to expand the functions of the equipment.

[0004] In addition, with the development of underwater acoustic signal processing technology, artificial intelligence technology is widely used in signal detection and recognition in complex underwater acoustic channel environments, but it has not been directly used in accompanying underwater acoustic communication signal processing equipment, which limits the processing performance of the equipment. Summary of the Invention

[0005] In order to solve the problem that the existing accompanying underwater acoustic signal acquisition equipment cannot be used for receiving and processing accompanying non-cooperative underwater acoustic communication signals, the present invention proposes an accompanying non-cooperative underwater acoustic communication signal receiving and processing device and method, designs an efficient standardized high-computing-power hardware device, combines high-computing-power artificial intelligence methods with the hardware device, adapts to typical underwater acoustic communication signal acquisition equipment, and completes the receiving and processing of accompanying non-cooperative underwater acoustic communication signals.

[0006] To achieve the above object, the first aspect of the present invention proposes an accompanying non-cooperative underwater acoustic communication signal receiving and processing device, including a computing resource platform and an intelligent data processing platform; The computing resource platform includes a VPX architecture host and a KVM device, and the intelligent data processing platform is written into the computing resource platform; The intelligent data processing platform includes a signal real-time processing module and a training management module. The signal real-time processing module includes a data processing and forwarding unit, a signal data processing unit, and a data display unit; The data processing and forwarding unit includes a transmitting end and a receiving end. The signal data processing unit includes an algorithm library and algorithm configuration. The data display unit includes displaying time-domain diagrams, frequency-spectrum diagrams, waterfall diagrams, constellation diagrams, and azimuth history diagrams, as well as structured data display; The algorithm library includes beamforming algorithms, recognition algorithms, detection algorithms, demodulation algorithms, and decoding algorithms; The training management module includes an annotation training unit, a data management unit, and a system configuration unit; The annotation training unit includes machine pre-annotation, manual annotation, algorithm training, and annotation export. The data management unit includes training data management and annotation data management. The system configuration unit includes platform configuration and log management; The manual annotation includes time-frequency signal annotation and analog acquisition signal annotation.

[0007] Further, the VPX architecture host includes a chassis, a main board, a backplane, a computing board, and a power supply board. The main board, backplane, computing board, and power supply board are electrically connected within the chassis. The VPX architecture host is communicatively connected to the KVM device; The chassis is a square structure with an open side and a hollow interior. One of the opposite sides of the chassis is hollowed out. A machine cover is provided on the chassis. A handle is fixedly provided on the machine cover. The chassis is provided with wiring holes corresponding to the main board, backplane, computing board, and power supply board; A cooling fan and an indicator light are provided within the chassis.

[0008] Further, the backplane is detachably connected to the chassis. The backplane is in a strip structure. Slots are provided on the backplane. A communication module is provided on the backplane. The backplane is connected to a VPX or ATX power supply.

[0009] Further, the computing board is provided with an MXM interface. The computing board is connected to a graphics card device through the MXM interface. The power supply board includes a 6U5HP specification power supply board.

[0010] A second aspect of the present invention proposes an accompanying non-cooperative underwater acoustic communication signal receiving and processing method, including signal processing and algorithm training; The signal processing includes: Step 1: After the original data is read and packetized by the transmitting end, it is transmitted to the receiving end; Step 2: The signal data processing unit performs processing, detection, recognition, demodulation, and decoding operations on the received signal according to the algorithm configuration; Step 3: Finally, the process data and result data are displayed and saved to disk.

[0011] Furthermore, the algorithm training includes: Step 1: After processing the off-line signal data, perform machine pre-labeling through the algorithm model in the algorithm training; Step 2: Manually label, adjust to form a labeled data set, use the labeled data set for algorithm training, and generate a new algorithm model; Step 3: Update the algorithm model and repeat Steps 1-2; Step 4: Obtain the final algorithm model and save it to disk.

[0012] Furthermore, Step 2 of the signal processing specifically includes: Step 2.1: When the data is multi-channel RF data, display the waveform diagram, spectrogram, and waterfall diagram through the data display unit, and perform array signal processing on the data through the beam synthesis algorithm; Step 2.2: When the processed data is single-channel RF data, display the waveform diagram, spectrogram, and waterfall diagram through the data display unit; Step 2.3: Collect the processed data through the detection algorithm, and perform modulation recognition through the recognition algorithm to obtain the signal type; Step 2.4: Perform parameter estimation on the recognized data, demodulate the data parameters according to the demodulation algorithm to obtain the source bit stream data, and display the source bit stream data through the data display unit as a constellation diagram; Step 2.5: Perform channel decoding on the demodulated data through the decoding algorithm to obtain the decoded bit stream data.

[0013] Furthermore, the data processing and forwarding unit, signal data processing unit, and data display unit are implemented using the C / S architecture, and the annotation training unit is implemented using the B / S architecture.

[0014] Through the above technical solutions, the beneficial effects of the present invention are as follows: The present invention realizes the reception and processing of non-cooperative underwater acoustic communication signals in an accompanying manner. A computing resource platform is set up, including a VPX architecture host and a KVM device, which provides hardware support for the intelligent data processing platform. Among them, the VPX architecture host adopts a standardized interface design, which can adapt to typical underwater acoustic communication signal acquisition equipment, and the device has good versatility and compatibility. In addition, the computing power resource insertion device can be flexibly customized according to the reception and processing requirements to improve the scalability of the device processing ability. In addition, in order to better complete the reception and processing of non-cooperative underwater acoustic communication signals in an accompanying manner, an intelligent data processing platform is set up. The intelligent data processing platform includes a signal real-time processing module and a training management module. The data processing flow is modular, and various module algorithms can be added and updated according to the processing ability, which is convenient for optimizing the processing flow and improving the processing effect. At the same time, it supports the application of artificial intelligence algorithms to realize annotation, training, and application, and improve the signal processing performance of the device in a complex underwater acoustic environment. Brief Description of the Drawings

[0015] Figure 1 This is the schematic diagram of the device and method for receiving and processing non - cooperative underwater acoustic communication signals in an accompanying manner according to the present invention.

[0016] Figure 2 This is one of the structural schematic diagrams of the device for receiving and processing non - cooperative underwater acoustic communication signals in an accompanying manner according to the present invention.

[0017] Figure 3 This is the second structural schematic diagram of the device for receiving and processing non - cooperative underwater acoustic communication signals in an accompanying manner according to the present invention.

[0018] Figure 4 This is the third structural schematic diagram of the device for receiving and processing non - cooperative underwater acoustic communication signals in an accompanying manner according to the present invention Figure 5 This is the fourth structural schematic diagram of the device for receiving and processing non - cooperative underwater acoustic communication signals in an accompanying manner according to the present invention.

[0019] Figure 6 This is the fifth structural schematic diagram of the device for receiving and processing non - cooperative underwater acoustic communication signals in an accompanying manner according to the present invention.

[0020] Figure 7 This is the first flowchart of the method for receiving and processing non - cooperative underwater acoustic communication signals in an accompanying manner according to the present invention.

[0021] Figure 8 This is the second flowchart of the method for receiving and processing non - cooperative underwater acoustic communication signals in an accompanying manner according to the present invention.

[0022] Figure 9 This is the schematic diagram of the real - time signal processing interface of the method for receiving and processing non - cooperative underwater acoustic communication signals in an accompanying manner according to the present invention.

[0023] Figure 10 This is the schematic diagram of the channel decoding interface of the method for receiving and processing non - cooperative underwater acoustic communication signals in an accompanying manner according to the present invention.

[0024] Reference Numerals in the Drawings: VPX - architecture host 1, KVM device 2, real - time signal processing module 3, training management module 4, chassis 101, main board 102, backplane 103, computing board 104, power supply board 105, data processing and forwarding unit 301, signal data processing unit 302, data display unit 303, annotation training unit 401, data management unit 402, system configuration unit 403. Detailed Embodiments

[0025] Embodiment 1 As Figures 1-8 shown, a device for receiving and processing non - cooperative underwater acoustic communication signals in an accompanying manner includes a computing resource platform and an intelligent data processing platform; The computing resource platform includes a VPX architecture host 1 and a KVM device 2, and the intelligent data processing platform is written into the computing resource platform; The intelligent data processing platform includes a signal real-time processing module 3 and a training management module 4. The signal real-time processing module 3 includes a data processing and forwarding unit 301, a signal data processing unit 302, and a data display unit 303; The data processing and forwarding unit 301 includes a transmitter and a receiver. The signal data processing unit 302 includes an algorithm library and algorithm configuration. The data display unit 303 includes displaying time-domain diagrams, frequency-spectrum diagrams, waterfall diagrams, constellation diagrams, and azimuth history diagrams, as well as structured data display; The algorithm library includes beamforming algorithms, recognition algorithms, detection algorithms, demodulation algorithms, and decoding algorithms; The training management module 4 includes an annotation training unit 401, a data management unit 402, and a system configuration unit 403; The annotation training unit 401 includes machine pre-annotation, manual annotation, algorithm training, and annotation export. The data management unit 402 includes training data management and annotation data management. The system configuration unit 403 includes platform configuration and log management; The manual annotation includes time-frequency signal annotation and analog acquisition signal annotation.

[0026] The VPX architecture host 1 includes a chassis 101, a main board 102, a backplane 103, a computing board 104, and a power supply board 105. The main board 102, the backplane 103, the computing board 104, and the power supply board 105 are electrically connected within the chassis 101, and the VPX architecture host 1 is communicatively connected to the KVM device 2; The chassis 101 is a square structure with one side open and hollow inside. One of the opposite sides of the chassis 101 is hollowed out. A machine cover is provided on the chassis 101, and a handle is fixedly provided on the machine cover. The chassis 101 is provided with wiring holes corresponding to the main board 102, the backplane 103, the computing board 104, and the power supply board 105; A cooling fan and an indicator light are provided inside the chassis 101.

[0027] In this embodiment, the chassis 101 adopts a 4U 19-inch VPX rack-mounted chassis, which can be installed in a standard 19-inch cabinet. It has electromagnetic compatibility, moisture resistance, salt spray resistance, and mold resistance characteristics, with 4 variable-speed cooling fans and various operation indicator lights. It has rich interfaces, including 1 220V power input interface, 2 1000M adaptive Ethernet interfaces, 2 USB 3.0 interfaces, 2 USB2.0 interfaces, 1 VGA interface, and 1 HDMI interface. The backplane 103 is detachably connected to the chassis 101. The backplane 103 has a strip structure, is provided with slots, is provided with a communication module, and is connected to a VPX or ATX power supply.

[0028] In this embodiment, the backplane 103 is a dedicated 6U6-slot VPX backplane, supporting multiple slots and signal transmissions, such as 1 VPX master slot, 1 VPX GPU slot, etc., supporting 2 USB3.0 signals, 4 HDMI signals, etc., and also supporting VPX and ATX power supplies.

[0029] The computing board 104 is provided with an MXM interface, and the computing board 104 is connected to a graphics card device through the MXM interface. The power board 105 includes a 6U5HP specification power board 105.

[0030] In this embodiment, the main board 102 supports domestic Linux and Windows systems, with an operating temperature range of -20°C to +55°C and a storage temperature range of -40°C to +70°C. It is equipped with an Intel Xeon Gold 5317 processor, having 12 cores and 24 threads, configured with a large-capacity solid-state drive and memory, and supporting the monitoring of multiple hardware parameters.

[0031] The computing board 104 is of the standard VPX 6U5HP specification, equipped with an NVIDIA GTX 3080 MXM graphics card, having 6144 CUDA cores and 16GB 256bit GDDR6 video memory, with a single-precision floating-point operation ability of up to 17.525 TFLOPS. The operating and storage temperature ranges are the same as those of the main control board, and it supports the replacement of multiple MXM interface graphics cards.

[0032] The power board 105 is of the standard 6U5HP specification, supports AC 220V±5% input, has a rated power higher than 720W, a power efficiency ≥85%, supports hot plugging and multiple protection functions, and complies with multiple standards.

[0033] The KVM device 2 adopts a 19-inch standard rack-mounted structure, with a height of 1U, is equipped with a 17.3-inch LCD liquid crystal screen, and has a three-proof design.

[0034] Embodiment 2 Combined with the accompanying non-cooperative underwater acoustic communication signal receiving and processing device in Embodiment 1, the accompanying non-cooperative underwater acoustic communication signal receiving and processing method is described as follows: It includes signal processing and algorithm training; The signal processing includes: Step 1: After the original data is read and packetized by the sending end, it is transmitted to the receiving end; Step 2: The signal data processing unit 302 processes, detects, identifies, demodulates, and decodes the received signal according to the algorithm configuration; Step 3: Finally, save the process data and result data.

[0035] Among them, Step 2 specifically includes: Step 2.1: When the data is multi-channel RF data, display the waveform diagram, spectrogram, and waterfall diagram through the data display unit 303, and perform array signal processing on the data through the beamforming algorithm; Step 2.2: When the processed data is single-channel RF data, display the waveform diagram, spectrogram, and waterfall diagram through the data display unit 303; Step 2.3: Collect the processed data through the detection algorithm, and perform modulation recognition through the recognition algorithm to obtain the signal type; Step 2.4: Perform parameter estimation on the recognized data, demodulate the data parameters according to the demodulation algorithm to obtain the source bitstream data, and display the source bitstream data through the data display unit 303 in the constellation diagram; Step 2.5: Perform channel decoding on the demodulated data through the decoding algorithm to obtain the decoded bitstream data.

[0036] In addition, the algorithm training includes: Step 1: After processing the offline signal data, perform machine pre-annotation through the algorithm model in the algorithm training; Step 2: Perform manual annotation, adjust to form an annotated data set, use the annotated data set for algorithm training, and generate a new algorithm model; Step 3: Update the algorithm model and repeat Steps 1-2; Step 4: Obtain the final algorithm model and save it to disk.

[0037] The data processing and forwarding unit 301, the signal data processing unit 302, and the data display unit 303 are implemented using the C / S architecture, and the annotation training unit 401 is implemented using the B / S architecture.

[0038] As Figure 9 shown. The user selects the modulation method to be processed and clicks the "Start" button. The device performs beamforming on the multi-channel data selected in the channel configuration and detects the signal. The detected signal information is displayed in real time in the signal list in the lower middle part of the interface. The user can also manually adjust parameters such as "symbol rate", "modulation method", "center frequency", and "bandwidth", and perform a "demodulation" operation on the signal. Through this implementation method, the user can flexibly process the real-time collected underwater acoustic signals and obtain the required information.

[0039] As Figure 10As shown. Demodulation and decoding process: The user selects the demodulated and saved data, observes the data bit signal on the interface, obtains the coding parameters and performs decoding, and finally displays and stores the decoding result. This implementation can further extract the effective information in the underwater acoustic signal and provide support for subsequent analysis.

[0040] The above-described embodiments are only preferred embodiments of the present invention and do not limit the scope of implementation of the present invention. Therefore, any equivalent changes or modifications made according to the structure, features, and principles described in the scope of the present invention's patent application shall be included within the scope of the present invention's patent application.

Claims

1. An adjoint non-cooperative underwater acoustic communication signal receiving and processing device, characterized in that, It includes a computing resource platform and an intelligent data processing platform; The computing resource platform includes a VPX architecture host (1) and a KVM device (2), and the intelligent data processing platform is written into the computing resource platform; The intelligent data processing platform includes a signal real-time processing module (3) and a training management module (4), and the signal real-time processing module (3) includes a data processing and forwarding unit (301), a signal data processing unit (302), and a data display unit (303); The data processing and forwarding unit (301) includes a transmitting end and a receiving end, the signal data processing unit (302) includes an algorithm library and algorithm configuration, and the data display unit (303) includes displaying time-domain diagrams, frequency spectrum diagrams, waterfall diagrams, constellation diagrams, and azimuth history diagrams as well as structured data display; The algorithm library includes beamforming algorithms, recognition algorithms, detection algorithms, demodulation algorithms, and decoding algorithms; The training management module (4) includes an annotation training unit (401), a data management unit (402), and a system configuration unit (403); The annotation training unit (401) includes machine pre-annotation, manual annotation, algorithm training, and annotation export, the data management unit (402) includes training data management and annotation data management, and the system configuration unit (403) includes platform configuration and log management; The manual annotation includes time-frequency signal annotation and analog acquisition signal annotation.

2. The syndrome non-cooperative underwater acoustic communication signal receiving and processing device according to claim 1, wherein The VPX architecture host (1) includes a chassis (101), a main board (102), a backplane (103), a computing board (104), and a power supply board (105). The main board (102), the backplane (103), the computing board (104), and the power supply board (105) are electrically connected inside the chassis (101), and the VPX architecture host (1) is communicatively connected to the KVM device (2); The chassis (101) is a square structure with one side open and hollow inside. One side of the chassis (101) is hollowed out, and a machine cover is provided on the chassis (101). A handle is fixedly provided on the machine cover, and wiring holes are provided on the chassis (101) corresponding to the main board (102), the backplane (103), the computing board (104), and the power supply board (105); A cooling fan and an indicator light are provided inside the chassis (101).

3. The syndrome non-cooperative underwater acoustic communication signal receiving and processing device according to claim 2, wherein, The backplane (103) is detachably connected to the chassis (101). The backplane (103) is in a strip structure, with slots provided on the backplane (103). A communication module is provided on the backplane (103), and the backplane (103) is connected to a VPX or ATX power supply.

4. The syndrome non-cooperative underwater acoustic communication signal receiving and processing device according to claim 2, characterized in that, The computing board (104) is provided with an MXM interface, and the computing board (104) is connected to a graphics card device through the MXM interface. The power supply board (105) includes a 6U5HP specification power supply board (105).

5. A method for receiving and processing accompanying non-cooperative underwater acoustic communication signals based on the device according to any one of claims 1 to 4, characterized in that, It includes signal processing and algorithm training; The signal processing includes: Step 1: After the original data is read and packetized by the sending end, it is transmitted to the receiving end; Step 2: The signal data processing unit (302) performs processing, detection, recognition, demodulation, and decoding operations on the received signal according to the algorithm configuration; Step 3: Finally, the process data and result data are displayed and stored in a disk.

6. The syndrome non-cooperative underwater acoustic communication signal receiving and processing method according to claim 5, characterized in that The algorithm training includes: Step 1: After processing the offline signal data, perform machine pre-labeling through the algorithm model in algorithm training; Step 2: Manually label, adjust to form a labeled data set, use the labeled data set for algorithm training, and generate a new algorithm model; Step 3: Update the algorithm model and repeat Steps 1-2; Step 4: Obtain the final algorithm model and save it to disk.

7. The syndrome non-cooperative underwater acoustic communication signal receiving and processing method according to claim 5, characterized in that Signal processing The specific steps of Step 2 include: Step 2.1: When the data is multi-channel RF data, display the waveform diagram, spectrogram, and waterfall diagram through the data display unit (303), and perform array signal processing on the data through the beam synthesis algorithm; Step 2.2: After processing, when the data is single-channel RF data, display the waveform diagram, spectrogram, and waterfall diagram through the data display unit (303); Step 2.3: Collect the processed data through the detection algorithm, and perform modulation recognition through the recognition algorithm to obtain the signal type; Step 2.4: Perform parameter estimation on the recognized data, demodulate the data parameters according to the demodulation algorithm to obtain the source bitstream data, and display the constellation diagram of the source bitstream data through the data display unit (303); Step 2.5: Perform channel decoding on the demodulated data through the decoding algorithm to obtain the decoded bitstream data.

8. The syndrome non-cooperative underwater acoustic communication signal receiving and processing method according to claim 5, characterized in that, The data processing and forwarding unit (301), the signal data processing unit (302), and the data display unit (303) are implemented using the C / S architecture, and the labeling and training unit (401) is implemented using the B / S architecture.

Images