A dynamic noise reduction circuit for multidimensional signals and its working method

Abstract

This invention discloses a dynamic denoising circuit for multidimensional signals and its operating method. The circuit includes: a multidimensional signal selection unit for selecting multidimensional signals for dynamic denoising; a dual-port signal data buffer for storing multidimensional signals; a multiply-accumulate denoising unit for performing Gaussian denoising on the stored multidimensional signals; a single-port denoising coefficient storage unit for storing denoising coefficients for retrieval by the multiply-accumulate denoising unit, wherein the denoising coefficients are the optimal denoising coefficients determined by an adaptive denoising coefficient adjustment unit learning the characteristics of multidimensional signals under different conditions; and an adaptive denoising coefficient adjustment unit for adjusting the denoising coefficients based on the denoising results. By implementing the circuit of this invention, the size and power consumption of the denoising circuit can be reduced, enabling the multidimensional signal detection device to achieve miniaturization and low power consumption, ensuring that the circuit achieves optimal denoising effect under different working environments, and improving the accuracy and reliability of detection.

Description

Technical Field

[0001] This invention relates to the field of live-line detection technology in power systems, and in particular to a dynamic noise reduction circuit for multidimensional signals and its operating method. Background Technology

[0002] In the field of live-line testing of power systems, infrared thermal imaging temperature measurement technology, along with partial discharge detection technologies including transient ground voltage detection, ultrasonic detection, high-frequency detection, and ultra-high-frequency detection, have been widely applied. Infrared thermal imaging temperature measurement technology can detect and identify equipment faults with abnormal temperatures in power systems, while partial discharge detection technology can detect partial discharge faults in equipment such as transformers, combined electrical appliances, and switchgear.

[0003] Currently, infrared thermal imaging temperature measurement and detection equipment and partial discharge detection equipment are independent devices, and their large size and inconvenience make them difficult to carry, resulting in heavy detection tasks, long detection cycles, and impacting the efficiency of operation and maintenance tasks and management. Therefore, integrating infrared thermal imaging detection, transient ground voltage detection, ultrasonic detection, and other conventional detection technologies has become an urgent need for power system operation and maintenance. Furthermore, simply combining multiple devices mechanically to achieve the integration of various detection functions only provides a multi-dimensional signal detection device in form, without addressing issues such as miniaturization, low power consumption, and long battery life. Moreover, current automatic noise reduction circuits use pre-set noise reduction coefficients, which may not achieve optimal noise reduction results under different operating environments.

[0004] Therefore, it is necessary to design a new circuit to reduce the size and power consumption of the denoising circuit, so as to achieve miniaturization and low power consumption of the multidimensional signal detection device. Furthermore, the circuit should automatically adjust the denoising coefficient according to the characteristics of the input signal to ensure that the circuit can achieve the optimal denoising effect under different working environments, thereby improving the accuracy and reliability of detection. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a dynamic noise reduction circuit for multidimensional signals and its working method.

[0006] To solve the above-mentioned technical problems, the objective of this invention is achieved through the following technical solution: A dynamic denoising circuit for multidimensional signals is provided, characterized in that it includes: a denoising mode control unit, a multidimensional signal selection unit, a signal data buffer dual-port, an adaptive denoising coefficient adjustment unit, and a multiply-accumulate denoising unit; wherein, the multidimensional signal selection unit is used to select the multidimensional signal for dynamic denoising; the signal data buffer dual-port is used to store the multidimensional signal; the multiply-accumulate denoising unit is used to perform Gaussian denoising processing on the multidimensional signal stored in the signal data buffer dual-port; the denoising mode control unit is connected to the multidimensional signal selection unit, the signal data buffer dual-port, and the multiply-accumulate denoising unit respectively;

[0007] It also includes a single port for storing denoising coefficients; wherein, the single port for storing denoising coefficients is used to store denoising coefficients for retrieval by the multiply-accumulate denoising unit, and the denoising coefficients are the optimal denoising coefficients determined by the adaptive denoising coefficient adjustment unit learning the multidimensional signal features under different conditions; the adaptive denoising coefficient adjustment unit is used to adjust the denoising coefficients in the single port for storing denoising coefficients according to the denoising result.

[0008] The further technical solution includes an enhanced signal preprocessing unit, which is used to filter, amplify, and correct the original signal.

[0009] The further technical solution is as follows: the enhanced signal preprocessing unit includes a programmable gain amplifier, a programmable filter, an automatic correction circuit, a real-time signal quality monitoring module, and a microcontroller. The programmable gain amplifier, programmable filter, automatic correction circuit, and real-time signal quality monitoring module are respectively connected to the microcontroller; the microcontroller is connected to the multi-dimensional signal selection unit.

[0010] The further technical solution is as follows: the multiply-accumulate denoising unit includes a multiply-accumulate circuit and a filter circuit connected in sequence.

[0011] Furthermore, the technical problem to be solved by this invention is to provide a method for operating a dynamic noise reduction circuit for multidimensional signals, comprising:

[0012] The multidimensional signal selection unit selects the multidimensional signal for dynamic denoising.

[0013] The signal data buffer stores the multidimensional signal in a dual-port manner.

[0014] The multiply-accumulate denoising unit retrieves denoising coefficients from the single port of the denoising coefficient storage and performs Gaussian denoising on the multidimensional signal stored in the dual port of the signal data buffer.

[0015] The adaptive denoising coefficient adjustment unit adjusts the denoising coefficients stored in the single port based on the denoising result.

[0016] The further technical solution is to determine whether the multidimensional signal is a calibrated infrared signal;

[0017] If the multidimensional signal is a calibrated infrared signal, then two rows of buffered infrared signals are read out sequentially from the two sets of signal data buffer dual ports.

[0018] The Gaussian denoising filter coefficients are read from the single port of the denoising coefficient storage to obtain the denoising coefficients;

[0019] The multiply-accumulate denoising unit uses 9 multipliers and 8 adders, and performs two-dimensional convolution operation on the infrared signal using denoising coefficients to obtain the Gaussian denoised signal.

[0020] A further technical solution is as follows: after determining whether the multidimensional signal is a corrected infrared signal, the method further includes:

[0021] If the multidimensional signal is not a calibrated infrared signal, then determine whether the multidimensional signal is a transient voltage signal;

[0022] If the multidimensional signal is a transient voltage signal, then the transient voltage signal is read out sequentially from the two sets of signal data buffer dual ports;

[0023] The L-order low-pass denoising filter coefficients are read from the denoising coefficient storage single port to obtain the denoising coefficients;

[0024] The multiply-accumulate denoising unit uses denoising coefficients to perform a one-dimensional convolution operation on the transient ground voltage signal to obtain the Gaussian denoised signal.

[0025] A further technical solution is as follows: after determining whether the multidimensional signal is a transient voltage signal, the method further includes:

[0026] If the multidimensional signal is not a transient voltage signal, then the ultrasonic signal is read out sequentially from the two sets of signal data buffer dual ports;

[0027] The L-order bandpass denoising filter coefficients are read from the denoising coefficient storage single port to obtain the denoising coefficients;

[0028] The multiply-accumulate denoising unit performs a one-dimensional convolution operation on the ultrasonic signal using denoising coefficients to obtain the Gaussian-denoised signal.

[0029] A further technical solution is as follows: before the multi-dimensional signal selection unit selects the multi-dimensional signal for dynamic denoising, it further includes:

[0030] The noise reduction mode control unit configures the noise reduction mode;

[0031] The noise reduction mode control unit configures the noise reduction coefficient.

[0032] The further technical solution is as follows: the adaptive denoising coefficient adjustment unit adjusts the denoising coefficient in the denoising coefficient storage port according to the denoising result, including:

[0033] The adaptive denoising coefficient adjustment unit evaluates the quality of the denoising result and fine-tunes the currently used denoising coefficient based on the quality evaluation result.

[0034] The beneficial effects of this invention compared with the prior art are as follows: This invention utilizes multidimensional signal selection and signal data buffering, implements Gaussian denoising processing through a multiply-accumulate denoising unit, and further adjusts the adaptive denoising coefficient by learning the multidimensional signal characteristics to determine the optimal denoising coefficient. The overall denoising process is managed by the denoising mode control unit, thereby reducing the size and power consumption of the denoising circuit. This enables the multidimensional signal detection device to achieve miniaturization and low power consumption. Moreover, it automatically adjusts the denoising coefficient according to the characteristics of the input signal, ensuring that the circuit can achieve the optimal denoising effect under different working environments, thus improving the accuracy and reliability of detection.

[0035] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 A schematic block diagram of a dynamic noise reduction circuit for multidimensional signals provided in an embodiment of the present invention;

[0038] Figure 2 A flowchart illustrating the operation of a dynamic noise reduction circuit for multidimensional signals provided in an embodiment of the present invention;

[0039] Figure 3 A schematic diagram of a sub-flow of a method for operating a dynamic noise reduction circuit for multidimensional signals provided in an embodiment of the present invention;

[0040] Figure 4 A flowchart illustrating the operation of a dynamic noise reduction circuit for multidimensional signals, provided in another embodiment of the present invention;

[0041] Explanation of the markings in the image:

[0042] 10. Noise reduction mode control unit; 20. Multidimensional signal selection unit; 30. Signal data buffer dual-port; 40. Multiply-accumulate noise reduction unit; 50. Noise reduction coefficient storage single-port; 60. Main control unit; 70. Adaptive noise reduction coefficient adjustment unit; 80. Enhanced signal preprocessing unit. Detailed Implementation

[0043] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0045] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0046] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0047] Please see Figure 1 , Figure 1This is a schematic block diagram of a dynamic denoising circuit for multidimensional signals provided in an embodiment of the present invention. This circuit can perform time-division multiplexing detection of infrared, transient ground voltage, and ultrasonic signals from power equipment. The acquired infrared, transient ground voltage, and ultrasonic multidimensional signals are processed in a time-division multiplexing manner within the same signal processing circuit. During multidimensional signal processing, firstly, a denoising mode and denoising coefficients are configured. Then, one of the infrared, transient ground voltage, and ultrasonic signals is selected and buffered into a dual-port 30MB RAM. Next, the buffered signal data and denoising coefficients are read. Based on the denoising mode, a multiply-accumulate circuit is invoked as needed, and the denoising circuit structure is reconstructed for time-division denoising of the multidimensional signals. Finally, the denoised infrared, ultrasonic, and transient ground voltage signals are output for further signal processing by subsequent circuits. This achieves the goal of reducing the size and power consumption of the denoising circuit, making miniaturization and low-power design of multidimensional signal detection equipment possible.

[0048] Furthermore, the optimal denoising coefficient is determined by learning the multi-dimensional signal characteristics under different conditions using the adaptive denoising coefficient adjustment unit and stored in the denoising coefficient storage single port. Moreover, the adaptive denoising coefficient adjustment unit will also adjust the corresponding denoising coefficient according to the result after each denoising to ensure that the circuit can achieve the best denoising effect under different working environments, thereby improving the accuracy and reliability of detection.

[0049] Please see Figure 1 The aforementioned dynamic denoising circuit for multidimensional signals includes: a denoising mode control unit 10, a multidimensional signal selection unit 20, a signal data buffer dual-port 30, and a multiply-accumulate denoising unit 40; wherein, the multidimensional signal selection unit 20 is used to select the multidimensional signal for dynamic denoising; the signal data buffer dual-port 30 is used to store the multidimensional signal; the multiply-accumulate denoising unit 40 is used to perform Gaussian denoising processing on the multidimensional signal stored in the signal data buffer dual-port 30; the denoising mode control unit 10 is connected to the multidimensional signal selection unit 20, the signal data buffer dual-port 30, and the multiply-accumulate denoising unit 40 respectively;

[0050] It also includes a denoising coefficient storage port 50; wherein, the denoising coefficient storage port 50 is used to store denoising coefficients for retrieval by the multiply-accumulate denoising unit, and the denoising coefficients are the optimal denoising coefficients determined by the adaptive denoising coefficient adjustment unit 70 learning the multidimensional signal features of different situations; the adaptive denoising coefficient adjustment unit 70 is used to adjust the denoising coefficients in the denoising coefficient storage port according to the denoising result.

[0051] The noise reduction coefficient storage port 50 and the noise reduction mode control unit 10 are respectively connected to the main control unit 60 and are used to control the operation of the noise reduction mode control unit 10 and the noise reduction coefficient storage port 50.

[0052] The circuit in this embodiment performs dynamic denoising on the acquired infrared thermal radiation, transient ground voltage, ultrasonic waves and other multidimensional signals through a dynamic denoising circuit. This allows for effective time-division multiplexing of the signal processing storage circuit and multiply-accumulate operation circuit, greatly reducing the circuit size and power consumption of multidimensional signal denoising processing. It also enables the integration of multiple charged detection functions into a single miniaturized portable device.

[0053] The adaptive denoising coefficient adjustment unit 70 uses machine learning algorithms to adjust the denoising coefficient in real time according to the characteristics of the input signal, so as to ensure the best denoising effect under different working environments, thereby improving the accuracy and reliability of detection.

[0054] In one embodiment, the aforementioned noise reduction coefficient storage single-port 50 includes two single-port 50 RAMs.

[0055] In one embodiment, the signal data buffer dual-port 30 described above includes two dual-port 30 RAMs.

[0056] In one embodiment, the multiply-accumulate denoising unit 40 described above includes a multiply-accumulate circuit and a filter circuit connected in sequence.

[0057] In this embodiment, the raw infrared array data (IR) with N×M resolution output from the infrared detector is acquired via an ADC converter. ori (x, y), outputting the corrected infrared array data IR after non-uniformity correction processing. nuc (x, y) represents the calibrated infrared signal. Here, N represents the number of columns in the infrared array, M represents the number of rows in the infrared array, x represents the horizontal axis of the infrared array data, and y represents the vertical axis of the infrared array data. IR nuc (x, y) represents the intensity value of the infrared array data after non-uniform correction at pixel coordinates (x, y). nuc (x, y) is kerneled by a 3×3 Gaussian kernel function. Perform Gaussian filtering for noise reduction, and reduce IR nuc (x, y) convolved with a Gaussian kernel function, i.e., Gaussian denoised infrared data The multiply-add denoising unit 40 needs to cache two rows of data through two sets of dual-port 30 RAM with a storage depth of N, and implement Gaussian denoising processing of infrared data through a circuit of 9 multipliers and 8 adders.

[0058] The transient ground voltage signal is processed by an analog conditioning circuit, including an analog filter circuit, a logarithmic amplifier circuit, and a peak detection circuit. The output circuit produces the most important peak information from the transient ground voltage signal. After sampling by an ADC, it undergoes L-order FIR low-pass denoising processing in the digital domain to remove high-frequency noise from the transient ground voltage detection data, thus improving the accuracy of transient ground voltage detection. The input sequence x of this L-order even-order FIR low-pass filter... tev (n) and transfer function h tev (n) convolution, outputting denoised transient ground voltage detection data: This denoising filter, using a semi-parallel filter design approach, first divides the denoising coefficients into two groups. and The data is stored in two separate single-port 50MB RAMs, and then two... A deep dual-port 30MB RAM is used for data buffering. Data from the dual-port 30MB RAM and denoising coefficients from the single-port 50MB RAM are read again via address control. Filtering operations are performed using two multipliers and two adders in a time-division multiplexing manner. Finally, the transient ground voltage detection data y after low-pass denoising is output. tev (n).

[0059] The ultrasonic signal, after being amplified by an analog amplifier and filtered by an analog low-pass filter with a cutoff frequency of 200kHz, is then sampled by an ADC with a sampling frequency of 400kHz to obtain a digital ultrasonic signal. Applying an L-order FIR bandpass filter to the digital ultrasonic signal extracts a denoised ultrasonic signal with a center frequency of 40kHz and a frequency range of 20kHz to 60kHz. The input sequence x of this L-order FIR bandpass denoising filter... us (n) and transfer function h us (n) convolution, outputting the bandpass-de-noiseed ultrasonic signal: This denoising filter uses a semi-parallel filter design approach to divide the denoising coefficients into two groups. and Stored into two sets of single-port 50MB RAM. Two... A deep dual-port 30MB RAM is used for data buffering. Data from the dual-port 30MB RAM and bandpass denoising coefficients from the single-port 50MB RAM are read via address control. Filtering operations are performed using two multipliers and two adders in a time-division multiplexing manner. Finally, the bandpass-filtered ultrasonic signal y is output. us (n).

[0060] In addition, the workflow of the adaptive noise reduction coefficient adjustment unit is as follows:

[0061] First, a suitable machine learning algorithm is selected, such as neural networks, support vector machines (SVM), or decision trees, based on the complexity of the signal features and the amount of data. Key features are extracted from the input signal (i.e., the multidimensional signal) through methods such as spectral analysis, time-domain analysis, and energy distribution analysis. These features directly affect the subsequent denoising effect. The model is then trained using a machine learning algorithm with labeled datasets, enabling it to learn the complex relationship between signal features and the optimal denoising coefficients. A large amount of data is required during training to ensure the model's accuracy and generalization ability. This results in a trained model that can be used to extract the optimal denoising coefficients corresponding to multidimensional signal features under different conditions.

[0062] Secondly, a memory or data structure, namely a single-port denoising coefficient storage unit (Port 50), is set up to store the optimal denoising coefficients under different conditions. These coefficients are dynamically updated and adjusted based on the output of the machine learning algorithm.

[0063] Next, during runtime, appropriate denoising coefficients are retrieved from the denoising coefficient storage single-port 50 based on the characteristics of the current input signal, i.e., the multidimensional information. These coefficients are obtained from previous training and learning and can be adjusted according to real-time requirements.

[0064] Finally, the quality of the denoised signal is evaluated, such as calculating the signal-to-noise ratio (SNR) and analyzing the energy within a specific frequency range. These evaluations help the system understand the current processing performance and make decisions for further adjustments. Based on real-time feedback evaluations, it is determined whether the currently used denoising coefficients need to be fine-tuned. Small adjustments can improve the denoising effect to adapt to constantly changing working environments or signal conditions. Specifically, by calculating the power or energy ratio of the signal to the noise, a quantitative indicator of signal quality can be obtained. A common formula is SNR = 10 * log10(P... signal / P noise ), where P signal It is the signal power, P noise It is noise power.

[0065] The system uses spectrum analysis tools or techniques such as Fourier transform to analyze the energy distribution of the signal within a specific frequency range. This helps determine the presence of noise or interference in the frequency domain; the quality assessment results are then fed back to the denoising mode control unit. This is achieved through data transmission and processing modules, ensuring the timeliness and accuracy of the feedback information. Based on the real-time quality assessment results, the system determines whether the currently used denoising coefficients are still effective. If the signal quality is poor or the operating environment changes, the system can trigger a fine-tuning process for the denoising coefficients. Utilizing previously learned multi-dimensional signal characteristics and data stored in the optimal denoising coefficient storage port, the system can use an adaptive algorithm to adjust the denoising coefficients. This algorithm can perform more precise denoising processing on the signal under current operating conditions based on the real-time feedback quality assessment results; fine-tuning typically uses small increments or decrements to avoid over-adjustment that could cause signal distortion or reduce processing effectiveness. This ensures that the system can adapt to dynamic changes in the operating environment and signal conditions, maintaining high denoising performance and signal quality.

[0066] By implementing the above steps, the dynamic noise reduction circuit can intelligently adjust and optimize the noise reduction effect based on the real-time quality assessment results, so as to ensure that it can provide stable and high-quality signal processing under different working conditions and signal environments.

[0067] These steps form a tightly linked closed-loop system: from identifying and learning signal features, to dynamically adjusting the denoising coefficients based on the learning results, and then optimizing the denoising effect based on real-time feedback. This design not only improves the performance and stability of the signal processing system, but also ensures optimal denoising results under various operating conditions, thereby effectively enhancing the overall system efficiency and responsiveness.

[0068] In one embodiment, the above-mentioned dynamic noise reduction circuit for multidimensional signals further includes an enhanced signal preprocessing unit 80, which is used to filter, amplify, and correct the original signal.

[0069] Specifically, the enhanced signal preprocessing unit 80 includes a programmable gain amplifier, a programmable filter, an automatic correction circuit, a real-time signal quality monitoring module, and a microcontroller. The programmable gain amplifier, programmable filter, automatic correction circuit, and real-time signal quality monitoring module are respectively connected to the microcontroller; the microcontroller is connected to the multidimensional signal selection unit.

[0070] In this embodiment, the programmable gain amplifier is model AD8367, which features a high gain range and precise gain control. It provides extensive dynamic range adjustment in the signal processing front-end, making it suitable for pre-processing in multi-dimensional signal selection units.

[0071] The programmable filter, model AD9361, contains a high-performance digital filter that can immediately filter received signals to meet different frequency and bandwidth requirements.

[0072] The automatic calibration circuit employs a high-precision, low-noise digital-to-analog converter, model AD5791, to generate accurate calibration signals. Combined with a microcontroller or FPGA, it enables precise offset and gain correction of the signal, eliminating system errors.

[0073] The real-time signal quality monitoring module is a high-speed, high-resolution analog-to-digital converter, model AD9467. It can acquire and monitor signal quality parameters such as signal-to-noise ratio and distortion in real time. These parameters can be analyzed in real time by an embedded processor, thereby dynamically adjusting the parameters of the preprocessing module.

[0074] The microcontroller, model Microchip PIC32MX, features a rich set of peripherals and communication interfaces, enabling it to manage and control the entire preprocessing module. It can efficiently exchange and coordinate data with various components such as PGA, PF, DAC, and ADC.

[0075] The integration and optimization of these components can form a highly efficient enhanced signal preprocessing unit 80, which can perform precise filtering, amplification and correction of the original signal before the multi-dimensional signal selection unit, thereby significantly improving the effect of subsequent noise reduction processing and the overall system performance.

[0076] In another embodiment, the aforementioned dynamic denoising circuit for multidimensional signals further includes a denoising strategy configuration unit, which is connected to the denoising mode control unit 10. This unit configures corresponding denoising parameters and algorithm strategies based on the signal type, noise characteristics, and user-defined denoising targets. Simultaneously, the denoising mode control unit 10 uses a machine learning model to predict the optimal denoising scheme.

[0077] Specifically, the first step is to classify the input signal and analyze its noise characteristics. This involves spectral analysis of sensor data, extraction of time-domain features, and estimation of the noise power spectrum. These analyses help the system understand the basic characteristics of the signal and the statistical properties of the noise, providing a foundation for subsequent denoising decisions.

[0078] Next, the desired noise reduction can be set through the system interface or preset parameters, such as reducing noise levels while preserving signal details. These objectives will serve as the basis for optimizing the system's noise reduction algorithm and parameter selection.

[0079] Secondly, based on signal analysis and user-defined objectives, the denoising strategy configuration unit is responsible for selecting appropriate denoising algorithms and parameter configurations. This includes selecting processing methods for different frequency bands or time domains, setting filter types and parameters, and adjusting the intensity and timeliness of denoising.

[0080] Next, the denoising mode control unit integrates a machine learning model, whose goal is to predict the optimal denoising scheme based on real-time signal characteristics and previous learning experience. This model can be a supervised learning-based classifier or regressor, an unsupervised learning clustering or anomaly detection model, or even a reinforcement learning agent.

[0081] Model training is typically based on historical datasets, including processing results under various signal types and noisy scenarios. During real-time execution, the model outputs a recommended optimal denoising strategy based on the characteristics of the current signal and the prediction objective. This may involve selecting a specific algorithm, adjusting parameters, or dynamically adjusting the priority and intensity of denoising.

[0082] Finally, after applying the denoising algorithm, the quality of the processed signal is monitored in real time. This feedback includes changes in the signal-to-noise ratio, measurements of signal distortion, or user-defined quality metrics. If the system detects poor denoising performance or changes in the signal environment, it can automatically adjust the denoising parameters or algorithm to optimize the processing results.

[0083] Through these steps and techniques, the dynamic denoising system can efficiently handle complex signal processing scenarios, ensuring excellent processing performance and user satisfaction under various noise interferences and changes in signal characteristics.

[0084] In another embodiment, the aforementioned dynamic denoising circuit for multidimensional signals further includes a high-performance data buffer and processing array. This array reads signal data and denoising coefficients according to instructions from the denoising strategy configuration unit and executes the selected denoising algorithm. During processing, the denoising effect is monitored in real time, and the denoising parameters are dynamically adjusted as needed.

[0085] In this embodiment, appropriate denoising algorithms and parameters are selected based on the system's real-time requirements and user settings. The high-performance data caching and processing array includes a software controller that dynamically generates denoising configuration instructions based on a preset algorithm library and parameter table, or an intelligent model (such as a machine learning model), according to real-time signal characteristics and denoising effect requirements.

[0086] The high-performance data caching and processing array internally contains a denoising algorithm library and execution unit. These algorithms can cover frequency domain processing (such as FFT, filter design), time domain processing (such as convolution, mean filtering), and complex signal processing algorithms (such as wavelet transform, deep learning models). According to the instructions of the denoising strategy configuration unit, the selected denoising algorithm is invoked by the execution unit and applied to the input signal in the data stream.

[0087] During signal processing, the adaptive denoising coefficient adjustment unit 70 monitors the denoising effect in real time. This includes calculating the signal-to-noise ratio (SNR) of the signal, evaluating the spectral characteristics of the denoised signal, or other relevant quality indicators. The monitoring results are fed back to the denoising strategy configuration unit and used to dynamically adjust the denoising parameters.

[0088] Based on real-time monitoring feedback, the adaptive denoising coefficient adjustment unit 70, combined with a high-performance data cache and processing array, can dynamically adjust the denoising parameters for the current application. This involves adjusting the filter cutoff frequency, increasing or decreasing the denoising intensity, and modifying the processing window size of the algorithm. These adjustments can be implemented through a software controller, ensuring the system can quickly respond to changes in the signal environment and adjustments to user needs. In other words, the adaptive denoising coefficient adjustment unit 70 adjusts the denoising coefficient, while the high-performance data cache and processing array adjusts other parameters.

[0089] Therefore, the high-performance data caching and processing array can effectively denoise various signal types and complex noise environments while ensuring processing speed and efficiency. This system architecture is not only suitable for applications with high real-time processing requirements, but also allows for flexible adjustment and optimization of denoising strategies according to different application needs.

[0090] The aforementioned dynamic denoising circuit for multidimensional signals selects one of the multidimensional signals, such as infrared, transient ground voltage, and ultrasonic, and writes it into the signal data buffer dual-port 30 by configuring the denoising mode and denoising coefficient. Then, it reads the buffered signal data and denoising coefficient, calls the multiply-accumulate denoising unit 40 as needed according to the denoising mode, and reconstructs the denoising circuit structure to perform time-division denoising processing of multidimensional signals. Finally, it outputs the denoised infrared, ultrasonic, and transient ground voltage signals, thereby reducing the size and power consumption of the denoising circuit and enabling the multidimensional signal detection equipment to achieve miniaturization and low power consumption.

[0091] In one embodiment, please refer to Figure 2 The above-mentioned method for operating a dynamic noise reduction circuit for multidimensional signals includes steps S110 to S140.

[0092] S110, Multi-dimensional signal selection unit 20 selects the multi-dimensional signal for dynamic denoising;

[0093] S120, signal data buffer dual-port 30 stores the multidimensional signal;

[0094] S130, the multiply-accumulate denoising unit 40 performs Gaussian denoising processing on the multidimensional signal stored in the signal data buffer dual-port 30;

[0095] In one embodiment, please refer to Figure 3 The above-mentioned step S130 may include steps S130a to S130k.

[0096] S130a. Determine whether the multidimensional signal is a calibrated infrared signal;

[0097] S130b If the multidimensional signal is a calibrated infrared signal, then read out two rows of buffered infrared signals sequentially from the two sets of signal data buffer dual ports 30.

[0098] S130c: Read the Gaussian denoising filter coefficients from the denoising coefficient storage single port 50 to obtain the denoising coefficients;

[0099] S130d, the multiply-accumulate denoising unit 40 performs a two-dimensional convolution operation on the infrared signal using denoising coefficients to obtain the signal after Gaussian denoising.

[0100] Specifically, the multiply-accumulate denoising unit 40 uses 9 multipliers and 8 adders, and performs two-dimensional convolution operation on the infrared signal using denoising coefficients to obtain the signal after Gaussian denoising.

[0101] S130e. If the multidimensional signal is not a calibrated infrared signal, then determine whether the multidimensional signal is a transient voltage signal.

[0102] S130f If the multidimensional signal is a transient voltage signal, then the transient voltage signal is read out sequentially from the two sets of signal data buffer dual ports 30;

[0103] S130g: Read the L-order low-pass denoising filter coefficients from the denoising coefficient storage single port 50 to obtain the denoising coefficients;

[0104] S130h, the multiply-accumulate denoising unit 40 uses denoising coefficients to perform a one-dimensional convolution operation on the transient ground voltage signal to obtain the signal after Gaussian denoising.

[0105] Specifically, the multiply-add denoising unit 40 performs filtering operations by time-division multiplexing two multipliers and two adders and using denoising coefficients to perform one-dimensional convolution operations on the transient ground voltage signal to obtain the Gaussian denoised signal.

[0106] S130i. If the multidimensional signal is not a transient voltage signal, the ultrasonic signal is read out sequentially from the two sets of signal data buffer dual-port 30.

[0107] S130j: Read the L-order bandpass denoising filter coefficients from the denoising coefficient storage single port 50 to obtain the denoising coefficients;

[0108] S130k, the multiply-accumulate denoising unit 40 uses denoising coefficients to perform one-dimensional convolution operation on the ultrasonic signal to obtain the signal after Gaussian denoising.

[0109] Specifically, the multiply-add denoising unit 40 performs a one-dimensional convolution operation on the transient ground voltage signal by time-division multiplexing two multipliers and two adders and using L-order bandpass denoising filter coefficients to achieve filtering operation, so as to obtain the signal after Gaussian denoising.

[0110] In this embodiment, the denoising mode control unit 10 selects the input signal for denoising, choosing one of infrared, transient ground voltage, or ultrasonic signals to write data into the dual-port 30 RAM. When the dynamic denoising circuit performs infrared signal denoising processing, two sets of dual-port 30 RAMs with a storage depth of N are configured, with each set of dual-port 30 RAMs buffering an adjacent row of infrared data; when the dynamic denoising circuit performs transient ground voltage or ultrasonic signal denoising processing, two sets of dual-port 30 RAMs with a storage depth of N are configured. The dual-port 30 RAM is used to cache half-depth signal processing data in each dual-port 30 RAM.

[0111] The noise reduction mode control unit 10 controls the read address generator of the dual-port 30 RAM, and controls the read port data output of the dual-port 30 RAM. When the dynamic noise reduction circuit performs infrared signal noise reduction processing, it sequentially reads two lines of infrared buffer data from the two sets of N-depth dual-port 30 RAM. When the dynamic noise reduction circuit performs transient ground voltage or ultrasonic signal noise reduction processing, it reads data from the two sets of N-depth dual-port 30 RAM. The deep dual-port 30-bit RAM sequentially reads two sets of transient ground voltage or ultrasonic buffer data.

[0112] The denoising mode control unit 10 controls the read address generator of the single-port 50 RAM to sequentially read the denoising coefficients of the corresponding denoising mode. When the dynamic denoising circuit performs infrared signal denoising processing, it reads the Gaussian denoising filter coefficients from the single-port 50 RAM. When the dynamic denoising circuit performs transient ground voltage signal denoising processing, it reads the L-order low-pass denoising filter coefficients from the single-port 50 RAM. When the dynamic denoising circuit performs ultrasonic signal denoising processing, it reads the L-order band-pass denoising filter coefficients from the single-port 50 RAM.

[0113] The denoising mode control unit 10 is configured with a multiply-accumulate circuit and a filter network to reconstruct the denoising circuit. When the dynamic denoising circuit performs infrared signal denoising processing, the multiply-accumulate circuit and the filter network are configured to perform two-dimensional convolution operations. When the dynamic denoising circuit performs transient voltage signal denoising, the multiply-accumulate circuit and the filter network are configured to perform one-dimensional convolution operations: When the dynamic denoising circuit performs ultrasonic signal denoising, the multiply-accumulate circuit and the filter network are configured to perform one-dimensional convolution operations:

[0114] The dynamic denoising circuit in this embodiment uses two sets of dual-port 30MB RAM with a depth of MAX(N, L / 2), two sets of single-port 50MB RAM with a depth of L / 2, nine multiplication circuits, and eight addition circuits to complete multidimensional denoising processing of infrared, transient ground voltage, and ultrasonic signals. In contrast, a separate, parallel three-channel denoising circuit requires 2×N + 4×L / 2 sets of dual-port 30MB RAM, 2×2×L / 2 sets of single-port 50MB RAM, 13 multiplication circuits, and 12 addition circuits to complete the corresponding denoising and filtering processing. Therefore, this invention effectively reduces the circuit size and power consumption of the denoising circuit, which is crucial for the integration, miniaturization, and low-power design of multidimensional signal detection products.

[0115] S140, The adaptive denoising coefficient adjustment unit 70 adjusts the denoising coefficient stored in the denoising coefficient storage port 50 according to the denoising result.

[0116] In this embodiment, the adaptive denoising coefficient adjustment unit 70 performs a quality assessment on the denoising result and fine-tunes the currently used denoising coefficient based on the quality assessment result.

[0117] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned dynamic noise reduction circuit for multidimensional signals can be referred to the corresponding description in the aforementioned device embodiments. For the sake of convenience and brevity, it will not be repeated here.

[0118] Figure 4 This is a schematic flowchart illustrating the operation of a dynamic noise reduction circuit for multidimensional signals according to another embodiment of the present invention. Figure 4 As shown, the working method of a dynamic noise reduction circuit for multidimensional signals in this embodiment includes steps S210-S260. Steps S230-S260 are similar to steps S110-S140 in the above embodiment and will not be described again here. The following details the additional steps S210-S220 in this embodiment.

[0119] S210, Noise Reduction Mode Control Unit 10 configures the noise reduction mode;

[0120] S220, the noise reduction mode control unit 10 is configured with a noise reduction coefficient.

[0121] The denoising mode refers to the operation mode performed by the multiply-accumulate denoising unit 40. The denoising coefficients include Gaussian denoising filter coefficients, L-order low-pass denoising filter coefficients, and L-order band-pass denoising filter coefficients.

[0122] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned dynamic noise reduction circuit for multidimensional signals can be referred to the corresponding description in the aforementioned device embodiments. For the sake of convenience and brevity, it will not be repeated here.

[0123] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A dynamic noise reduction circuit for multidimensional signals, characterized in that, include: The system comprises a denoising mode control unit, a multidimensional signal selection unit, a signal data buffer dual-port, an adaptive denoising coefficient adjustment unit, and a multiply-accumulate denoising unit; wherein, the multidimensional signal selection unit is used to select the multidimensional signal for dynamic denoising; the signal data buffer dual-port is used to store the multidimensional signal; the multiply-accumulate denoising unit is used to perform Gaussian denoising processing on the multidimensional signal stored in the signal data buffer dual-port; and the denoising mode control unit is connected to the multidimensional signal selection unit, the signal data buffer dual-port, and the multiply-accumulate denoising unit respectively. It also includes a single port for storing denoising coefficients; wherein, the single port for storing denoising coefficients is used to store denoising coefficients for retrieval by the multiply-accumulate denoising unit, and the denoising coefficients are the optimal denoising coefficients determined by the adaptive denoising coefficient adjustment unit learning the multidimensional signal features under different conditions; the adaptive denoising coefficient adjustment unit is used to adjust the denoising coefficients in the single port for storing denoising coefficients according to the denoising result. It also includes an enhanced signal preprocessing unit, which is used to filter, amplify and correct the original signal; The enhanced signal preprocessing unit includes a programmable gain amplifier, a programmable filter, an automatic correction circuit, a real-time signal quality monitoring module, and a microcontroller. The programmable gain amplifier, programmable filter, automatic correction circuit, and real-time signal quality monitoring module are respectively connected to the microcontroller; the microcontroller is connected to the multi-dimensional signal selection unit.

2. The dynamic noise reduction circuit for multidimensional signals according to claim 1, characterized in that, The multiply-accumulate noise reduction unit includes a multiply-accumulate circuit and a filter circuit connected in sequence.

3. A method for operating a dynamic noise reduction circuit for multidimensional signals as described in any one of claims 1 to 2, characterized in that, include: The multidimensional signal selection unit selects the multidimensional signal for dynamic denoising. The signal data buffer stores the multidimensional signal in a dual-port manner. The multiply-accumulate denoising unit retrieves denoising coefficients from the single port of the denoising coefficient storage and performs Gaussian denoising on the multidimensional signal stored in the dual port of the signal data buffer. The adaptive denoising coefficient adjustment unit adjusts the denoising coefficients stored in the single port based on the denoising result.

4. The operating method of the dynamic noise reduction circuit for multidimensional signals according to claim 3, characterized in that, The multiply-accumulate denoising unit performs Gaussian denoising processing on the multidimensional signal stored in the dual-port signal data buffer, including: Determine whether the multidimensional signal is a calibrated infrared signal; If the multidimensional signal is a calibrated infrared signal, then two rows of buffered infrared signals are read out sequentially from the two sets of signal data buffer dual ports. The Gaussian denoising filter coefficients are read from the single port of the denoising coefficient storage to obtain the denoising coefficients; The multiply-accumulate denoising unit uses 9 multipliers and 8 adders, and performs two-dimensional convolution operation on the infrared signal using denoising coefficients to obtain the Gaussian denoised signal.

5. The operating method of the dynamic noise reduction circuit for multidimensional signals according to claim 4, characterized in that, After determining whether the multidimensional signal is a corrected infrared signal, the method further includes: If the multidimensional signal is not a calibrated infrared signal, then determine whether the multidimensional signal is a transient voltage signal; If the multidimensional signal is a transient voltage signal, then the transient voltage signal is read out sequentially from the two sets of signal data buffer dual ports; The L-order low-pass denoising filter coefficients are read from the denoising coefficient storage single port to obtain the denoising coefficients; The multiply-accumulate denoising unit uses denoising coefficients to perform a one-dimensional convolution operation on the transient ground voltage signal to obtain the Gaussian denoised signal.

6. The operating method of the dynamic noise reduction circuit for multidimensional signals according to claim 5, characterized in that, After determining whether the multidimensional signal is a transient voltage signal, the method further includes: If the multidimensional signal is not a transient voltage signal, then the ultrasonic signal is read out sequentially from the two sets of signal data buffer dual ports; The L-order bandpass denoising filter coefficients are read from the denoising coefficient storage single port to obtain the denoising coefficients; The multiply-accumulate denoising unit performs a one-dimensional convolution operation on the ultrasonic signal using denoising coefficients to obtain the Gaussian-denoised signal.

7. The operating method of the dynamic noise reduction circuit for multidimensional signals according to claim 3, characterized in that, Before the multidimensional signal selection unit selects the multidimensional signal for dynamic denoising, it also includes: The noise reduction mode control unit configures the noise reduction mode; The noise reduction mode control unit configures the noise reduction coefficient.

8. The operating method of the dynamic noise reduction circuit for multidimensional signals according to claim 4, characterized in that, The step of adjusting the denoising coefficients in the denoising coefficient storage port by the adaptive denoising coefficient adjustment unit according to the denoising result includes: The adaptive denoising coefficient adjustment unit evaluates the quality of the denoising result and fine-tunes the currently used denoising coefficient based on the quality evaluation result.

Images