An anti-interference emergency communication signal enhancement device

Abstract

The application provides an anti-interference emergency communication signal enhancement device, which comprises a multi-band antenna array, the multi-band antenna array is connected with a preprocessing circuit, the preprocessing circuit is connected with a machine learning processing chip and a power amplifier, the machine learning processing chip is connected with an environment sensor group, the preprocessing circuit, the machine learning processing chip and the environment sensor group are powered by a dual power supply system, the multi-band antenna array, the preprocessing circuit and the power amplifier are isolated in a high-frequency chamber by a split shielding cavity, and the machine learning processing chip, the dual power supply system and the environment sensor group are isolated in a low-frequency chamber by a split shielding cavity. The machine learning processing chip is used to realize real-time interference filtering parameter adjustment, improve dynamic response capability and anti-interference capability, meanwhile, the split shielding cavity is used to ensure that the high-frequency circuit and the low-frequency circuit work without interference, improve electromagnetic compatibility capability and maintain device operation stability.

Description

Technical Field

[0001] This application relates to the field of wireless signal enhancement devices, and more specifically, to an anti-interference emergency communication signal enhancement device. Background Technology

[0002] Emergency communications are specialized communication systems established to respond to emergencies such as natural disasters and major accidents. They utilize technologies such as radio and satellite communications to solve the problem of information transmission after traditional communication is interrupted. Especially under severe interference conditions, the performance of emergency communication devices directly affects the ability and efficiency of responding to emergencies and accidents.

[0003] Existing technologies suffer from the following technical shortcomings: Disaster sites often experience interference from multiple devices operating on the same frequency band. Some anti-interference devices employ multi-antenna beamforming schemes that can only suppress spatially separated interference sources, failing to address spectral aliasing and resulting in higher bit error rates when signal bandwidths partially overlap. Furthermore, while some current anti-interference methods based on FPGA-based reconfigurable filters support mode switching, they require pre-configured parameters. In unknown interference environments, such as sudden electromagnetic pulses, their response delays are significant, failing to meet the real-time requirements of emergency communication. Additionally, traditional integrated metal cavities exhibit substantial crosstalk noise when high-power RF circuits and digital control circuits coexist, leading to a sharp decline in signal-to-noise ratio and weak electromagnetic compatibility. Summary of the Invention

[0004] The technical problems to be solved by this application are: the inability to solve the spectrum aliasing problem under interference in the same frequency band; the large delay in filter reconstruction response in emergency communication scenarios, which cannot meet the real-time requirements; and the weak electromagnetic compatibility of traditional integrated metal cavity.

[0005] To address the aforementioned problems, this application provides an anti-interference emergency communication signal enhancement device.

[0006] Therefore, the specific technical solution adopted by this utility model is as follows:

[0007] An anti-interference emergency communication signal enhancement device includes a multi-band antenna array, a preprocessing circuit connected to the multi-band antenna array, a machine learning processing chip and a power amplifier connected to the preprocessing circuit, an environmental sensor group connected to the machine learning processing chip, the preprocessing circuit, the machine learning processing chip and the environmental sensor group being powered by a dual power supply system, the multi-band antenna array, the preprocessing circuit and the power amplifier being isolated in a high-frequency chamber by a split-type shielded cavity, and the machine learning processing chip, the dual power supply system and the environmental sensor group being isolated in a low-frequency chamber by a split-type shielded cavity.

[0008] Furthermore, the multi-band antenna array is used to receive multi-standard communication signals. Considering the wide frequency range coverage of emergency communication scenarios, the antenna array covers the 350MHz-2.5GHz frequency band and includes four sets of orthogonal dipole antennas.

[0009] Furthermore, the preprocessing circuit includes a parallel surface acoustic wave filter and an LC filter network, and the switching threshold is controlled in real time by a machine learning processing chip to filter out out-of-band interference signals.

[0010] Furthermore, the machine learning processing chip adopts Xilinx Zynq UltraScale+ MPSoC, which has a built-in algorithm processing module. The algorithm processing module includes a convolutional neural network module, a long short-term memory network module, and a parameter generation module. The convolutional neural network module runs a CNN model to extract the time-frequency domain features of the signal, the long short-term memory network module runs an LSTM model to predict the evolution trend of the interference signal, and the parameter generation module outputs dynamic filter coefficients to the preprocessing circuit to realize the real-time change of filter parameters.

[0011] Furthermore, to enhance electromagnetic compatibility, the split-type shielded cavity separates the high-frequency circuits and low-frequency circuits into independent copper alloy chambers, which are connected by optical fiber channels.

[0012] Furthermore, to ensure reliable power supply, the dual power supply system includes a priority switching circuit that automatically activates the battery when a mains power fluctuation of >15% is detected.

[0013] Furthermore, to further improve the accuracy of filter parameter adjustment, this device also includes an environmental sensor group, including temperature, humidity, and vibration sensors, with data input to a machine learning processing chip for interference compensation.

[0014] Furthermore, to enhance the device's dust and water resistance, all functional modules are encapsulated within an IP67-rated enclosure.

[0015] The technical advantages of this application are as follows:

[0016] The use of machine learning processing chips enables real-time adjustment of interference filtering parameters, improving dynamic response and anti-interference capabilities. This addresses spectral aliasing issues in complex electromagnetic interference environments and facilitates rapid response to sudden electromagnetic pulse interference. Furthermore, a split-type shielded cavity ensures interference-free operation between high-frequency and low-frequency circuits, resolving problems of high crosstalk noise and low signal-to-noise ratio, thus enhancing electromagnetic compatibility and maintaining device operational stability. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall structure of the device.

[0018] Figure 2Internal architecture diagram of a machine learning processing chip.

[0019] Explanation of reference numerals in the attached figures:

[0020] 1. Multi-band antenna array;

[0021] 2. Preprocessing circuit; 21. Parallel surface acoustic wave filter; 22. LC filter network;

[0022] 3. Machine learning processing chip; 31. Convolutional neural network module; 32. Long short-term memory network module; 33. Parameter generation module;

[0023] 4. Power amplifier;

[0024] 5. Dual power supply system; 51. Storage battery; 52. Switching circuit;

[0025] 6. Split-type shielded cavity; 61. High-frequency cavity; 62. Low-frequency cavity;

[0026] 7. Environmental sensor array. Detailed Implementation

[0027] The embodiments of the technical solution of this application will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of this application and are therefore merely examples, and should not be used to limit the scope of protection of this application.

[0028] Example 1

[0029] like Figures 1-2As shown, the anti-interference emergency communication signal enhancement device according to this embodiment includes a multi-band antenna array 1, which is connected to a preprocessing circuit 2. The preprocessing circuit 2 is connected to a machine learning processing chip 3 and a power amplifier 4. The machine learning processing chip 3 is connected to an environmental sensor group 7. The preprocessing circuit 2, the machine learning processing chip 3, and the environmental sensor group 7 are powered by a dual power supply system 5. The multi-band antenna array 1, the preprocessing circuit 2, and the power amplifier 4 are isolated in a high-frequency chamber 61 by a split-type shielded cavity 6, and the machine learning processing chip 3, the dual power supply system 5, and the environmental sensor group 7 are isolated in a low-frequency chamber 62 by the split-type shielded cavity 6. The multi-band antenna array 1 is used to receive multi-mode communication signals, covering the 350MHz-2.5GHz frequency band, and includes four sets of orthogonal dipole antennas. The preprocessing circuit 2 includes a parallel surface acoustic wave filter 21 and an LC filter network 22, and the switching threshold is controlled in real time by the machine learning processing chip 3. The machine learning processing chip 3 uses a Xilinx Zynq UltraScale+ MPSoC and has a built-in algorithm processing module. This module includes a convolutional neural network module 31 running a CNN model, a long short-term memory network module 32 running an LSTM model to predict the evolution trend of interference signals, and a parameter generation module 33 outputting dynamic filter coefficients to the preprocessing circuit 2. The algorithm training dataset contains IQ signal samples from typical electromagnetic interference scenarios. The environmental sensor group 7 includes temperature, humidity, and vibration sensors; the data is input to the machine learning processing chip 3 for interference compensation. The power amplifier 4 outputs the enhanced signal. Through this approach, deploying the machine learning processing chip within the device enables real-time adjustment of filter parameters. The ADC (analog-to-digital converter) samples the output signal from the preprocessing circuit and inputs it to the CNN module 31. The signal's time-frequency features are extracted through convolutional layers, pooling layers compress the data dimension while retaining key features, and normalization layers perform batch normalization to eliminate data offset, collectively achieving signal spectral feature extraction. LSTM module 32 processes the temporal changes of spectral features. Specifically, the forget gate selectively forgets past information using a sigmoid function, the input gate calculates new candidate memories using a sigmoid function and a tanh function, and then performs element-wise operations on the outputs of the forget gate and the input gate to obtain the memory for the current time step. The output gate calculates the hidden state for the current time step using a sigmoid function and a tanh function, and then multiplies it by the output of the output gate to obtain the final output result. By combining the spatial awareness capability of CNN and the temporal series processing capability of LSTM, the prediction of the spatiotemporal features of interference is completed. Parameter generation module 33 generates new filtering parameters based on the interference prediction results.

[0030] like Figure 1As shown, the device internally features a split-type shielded cavity 6, separating the high-frequency and low-frequency circuits into independent copper alloy chambers 61 and 62, connected by an optical fiber channel 63. The dual-power system 5 includes a priority switching circuit 51, which automatically activates the battery 52 when a mains power fluctuation >15% is detected. All modules are encapsulated within an IP67-rated enclosure 8. Through this design, the split-type shielded cavity effectively blocks electromagnetic radiation from the high-frequency circuits from causing electromagnetic interference to the low-frequency circuits through spatial coupling, ensuring normal circuit operation. The dual-power design achieves reliable uninterrupted power supply.

[0031] In summary, by utilizing the above-mentioned technical solution of this utility model, the machine learning processing chip 3 achieves real-time adjustment of interference filtering parameters, improves dynamic response capability and anti-interference capability, solves the problem of spectral aliasing under complex electromagnetic interference environments, and addresses the problem of rapid response to sudden electromagnetic pulse interference. Furthermore, the split-type shielded cavity 6 ensures interference-free operation of high-frequency and low-frequency circuits, solves the problems of high crosstalk noise and low signal-to-noise ratio, improves electromagnetic compatibility, and maintains the operational stability of the device.

[0032] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. An anti-interference emergency communication signal enhancement device, characterized in that, The device includes a multi-band antenna array (1), which is connected to a preprocessing circuit (2). The preprocessing circuit (2) is connected to a machine learning processing chip (3) and a power amplifier (4). The machine learning processing chip (3) is connected to an environmental sensor group (7). The preprocessing circuit (2), the machine learning processing chip (3), and the environmental sensor group (7) are powered by a dual power supply system (5). The multi-band antenna array (1), the preprocessing circuit (2), and the power amplifier (4) are isolated in a high-frequency chamber (61) by a split-type shielded cavity (6). The machine learning processing chip (3), the dual power supply system (5), and the environmental sensor group (7) are isolated in a low-frequency chamber (62) by a split-type shielded cavity (6).

2. The apparatus according to claim 1, characterized in that: The multi-band antenna array (1) is used to receive multi-standard communication signals, covering the 350MHz-2.5GHz frequency band, and includes four sets of orthogonal dipole antennas.

3. The apparatus according to claim 1, characterized in that: The preprocessing circuit (2) includes a surface acoustic wave filter (21) and an LC filter network (22) connected in parallel, and the switching threshold is controlled in real time by the machine learning processing chip (3).

4. The apparatus according to claim 1, characterized in that: The machine learning processing chip (3) adopts Xilinx Zynq UltraScale+ MPSoC and has a built-in algorithm processing module. The algorithm processing module includes a convolutional neural network module (31), a long short-term memory network module (32), and a parameter generation module (33). The convolutional neural network module (31) runs a CNN model to extract the time-frequency domain features of the signal. The long short-term memory network module (32) runs an LSTM model to predict the evolution trend of the interference signal. The parameter generation module (33) outputs dynamic filtering coefficients to the preprocessing circuit (2).

5. The apparatus according to claim 1, characterized in that: The split-type shielded cavity (6) places the high-frequency circuit and the low-frequency circuit in the high-frequency cavity (61) and the low-frequency cavity (62) respectively, and the cavities are connected by an optical fiber channel (63).

6. The apparatus according to claim 1, characterized in that: The dual power supply system (5) includes a priority switching circuit (51) that automatically activates the battery (52) when a mains power fluctuation of >15% is detected.

7. The apparatus according to claim 1, characterized in that: It also includes an environmental sensor group (7), containing temperature, humidity, and vibration sensors, and data input to a machine learning processing chip (3) for interference compensation.

8. The apparatus according to any one of claims 1-7, characterized in that: The entire enclosure is encapsulated within an IP67-rated housing (8).

Images