Method and device for suppressing phase noise of oscillator

Abstract

The invention discloses an oscillator phase noise suppression method and device, and belongs to the field of signal processing and noise control, and the method comprises the steps: traversing a phase noise source of an oscillator for noise recognition; performing dynamic tracking based on the plurality of phase noise standard signals; performing noise prediction on an oscillator according to the noise change state data, and filtering the plurality of phase noise standard signals based on a phase noise prediction result; performing phase locking loop according to the noise filtering data, and performing multi-layer noise suppression on the plurality of phase noise standard signals through a multi-stage filtering structure in combination with a noise phase difference value; suppression detection is carried out based on a noise suppression result, a suppression effect is fed back, and a suppression strategy is formulated to carry out self-adaptive suppression on the phase noise of the oscillator. According to the invention, the technical problems of signal instability and spectrum broadening caused by poor phase noise suppression effect of an oscillator in a complex noise source environment in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the fields of signal processing and noise control, and in particular to a method and device for suppressing oscillator phase noise. Background Art

[0002] Signal stability and accuracy are crucial in modern electronic communications and precision measurement systems, particularly in wireless communications, radar, satellite navigation, and high-frequency devices. Oscillators, core components for generating clock and carrier signals, have a significant impact on system performance due to their phase noise. Phase noise can cause spectrum broadening and signal distortion, even affecting the demodulation and decoding processes of the entire system, reducing data transmission rates and signal quality. Therefore, effectively suppressing oscillator phase noise has become a key research topic for improving the reliability of high-precision and high-frequency electronic systems.

[0003] At present, the technologies for suppressing oscillator phase noise mainly include traditional passive and active filtering technologies, as well as phase-locked loop (PLL)-based phase-locked technology. However, these methods have certain limitations when faced with complex noise sources and dynamically changing noise conditions, such as incomplete noise filtering, large system delays, and insufficient dynamic response. As the requirements of electronic systems for high frequency and low noise continue to increase, existing technologies have been unable to meet the needs of phase noise suppression in complex application scenarios. Therefore, there is an urgent need for a method and device for suppressing oscillator phase noise to solve the technical problem in the prior art that the oscillator has poor phase noise suppression effect in a complex noise source environment, resulting in signal instability and spectrum broadening. Summary of the Invention

[0004] The present application provides a method and device for suppressing oscillator phase noise, aiming to solve the technical problem in the prior art that the oscillator has poor phase noise suppression effect in a complex noise source environment, thereby leading to signal instability and spectrum broadening.

[0005] In view of the above problems, the present application provides a method and apparatus for suppressing oscillator phase noise.

[0006] The first aspect disclosed in the present application provides a method for suppressing oscillator phase noise, which includes traversing the phase noise source of the oscillator to perform noise identification, generating multiple phase noise signals, standardizing the multiple phase noise signals, and determining multiple phase noise standard signals; dynamically tracking the multiple phase noise standard signals to construct noise change state data; predicting the noise of the oscillator according to the noise change state data to obtain a phase noise prediction result, filtering the multiple phase noise standard signals based on the phase noise prediction result to determine noise filtering data; performing a phase-locked loop according to the noise filtering data to generate a noise phase difference value, constructing a multi-stage filtering structure, performing multi-layer noise suppression on the multiple phase noise standard signals through the multi-stage filtering structure in combination with the noise phase difference value to obtain a noise suppression result; performing suppression detection based on the noise suppression result, feeding back the suppression effect, and formulating a suppression strategy to adaptively suppress the phase noise of the oscillator.

[0007] Another aspect disclosed in the present application provides an apparatus for suppressing oscillator phase noise, the apparatus comprising a phase noise standard signal determination module for traversing the phase noise source of the oscillator to perform noise identification, generate multiple phase noise signals, perform standardization processing on the multiple phase noise signals, and determine multiple phase noise standard signals; a noise change state data construction module for dynamically tracking based on the multiple phase noise standard signals and constructing noise change state data; a noise filtering data determination module for predicting the noise of the oscillator based on the noise change state data to obtain a phase noise prediction result, filtering the multiple phase noise standard signals based on the phase noise prediction result to determine noise filtering data; a noise suppression result acquisition module for performing a phase-locked loop based on the noise filtering data to generate a noise phase difference value, construct a multi-stage filtering structure, and perform multi-layer noise suppression on the multiple phase noise standard signals through the multi-stage filtering structure in combination with the noise phase difference value to obtain a noise suppression result; and an adaptive suppression module for performing suppression detection based on the noise suppression result, feeding back the suppression effect, and formulating a suppression strategy to adaptively suppress the phase noise of the oscillator.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] By traversing the oscillator's phase noise sources for noise identification, generating multiple phase noise signals, and then performing standardized processing, this technology addresses the existing technical problem of poor phase noise suppression in oscillators in complex noise environments, which leads to signal instability and spectrum broadening. By accurately identifying noise sources and extracting effective features, it is possible to effectively separate and standardize different noise signals, significantly improving phase noise suppression, ensuring signal stability, reducing spectrum broadening, and ultimately optimizing oscillator performance.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A flow chart of a method for suppressing oscillator phase noise is provided for an embodiment of the present application.

[0012] Figure 2 A schematic structural diagram of an oscillator phase noise suppression device is provided for an embodiment of the present application.

[0013] Explanation of the reference numerals: phase noise standard signal determination module 11 , noise change state data construction module 12 , noise filtering data determination module 13 , noise suppression result acquisition module 14 , adaptive suppression module 15 . DETAILED DESCRIPTION

[0014] The overall idea of the technical solution provided by this application is as follows:

[0015] Embodiments of the present application provide a method and apparatus for suppressing oscillator phase noise. By traversing the oscillator's phase noise sources, noise identification is performed and multiple phase noise signals are generated. First, the oscillator's noise sources are identified and an initial noise signal is acquired. Spectral analysis is then used to extract noise features, which are then used for cluster analysis. Finally, the noise signals are separated by category, generating multiple phase noise signals. This process provides an accurate data foundation for subsequent noise processing and suppression.

[0016] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced in conjunction with the drawings in the specification.

[0017] Example 1, as Figure 1 As shown, an embodiment of the present application provides a method for suppressing oscillator phase noise, the method comprising:

[0018] Step S100: traversing the phase noise sources of the oscillator to perform noise identification, generating multiple phase noise signals, performing standardization processing on the multiple phase noise signals, and determining multiple phase noise standard signals.

[0019] Specifically, an oscillator is an electronic component used to generate a periodically varying electrical signal, typically used to generate clock or carrier signals. The stability of an oscillator directly impacts the accuracy and reliability of the signal. Phase noise refers to the phase disturbance caused by a frequency signal deviating from its ideal frequency in the frequency domain. This noise can cause the signal spectrum to broaden and distort. Phase noise typically arises from a variety of noise sources, such as thermal noise, intrinsic device noise, and external interference. A phase noise signal refers to a specific noise signal identified and extracted from a noise source. Each phase noise signal represents a noise characteristic, such as noise components of different frequency bands or intensities. Standardization processing involves adjusting different noise signals to the same standard form by unifying parameters such as signal amplitude, frequency, and time for subsequent analysis. A phase noise standard signal refers to a standardized phase noise signal with a uniform frequency distribution, amplitude range, and time synchronization.

[0020] First, the oscillator's various phase noise sources are iterated through, treating each noise source as a separate signal input for noise identification. During this process, a spectrum analyzer is used to measure the spectrum of the signal from each noise source, thereby generating multiple different phase noise signals.

[0021] Each generated phase noise signal is then normalized. This process involves three main steps: amplitude normalization, frequency normalization, and time alignment normalization. For example, amplitude normalization uses a normalization algorithm to adjust the amplitudes of all noise signals to the same range (e.g., 0 to 1); frequency normalization maps the frequency distribution of each signal to a specified range, giving it uniform frequency characteristics; and time alignment normalization adjusts the signal's time axis to ensure that all noise signals are displayed synchronously within the same time window. These normalization steps are typically performed using signal processing tools, such as MATLAB or Python signal processing libraries. After these processes, multiple phase noise standard signals with uniform characteristics are obtained.

[0022] Through this step, the oscillator's multiple noise sources are identified and corresponding noise signals are generated. Normalization ensures that these noise signals have consistent characteristics in subsequent analysis. This allows the suppression process to more accurately identify and remove noise of different frequency bands and types, improving the effectiveness of phase noise suppression.

[0023] Step S200: Dynamically tracking the multiple phase noise standard signals to construct noise change state data.

[0024] Specifically, dynamic tracking involves real-time monitoring of changes in a phase noise standard signal to identify and record its fluctuation characteristics. Dynamic tracking typically utilizes data updates and change capture algorithms to identify signal variation patterns. Noise variation state data refers to the time-varying state information recorded through dynamic tracking of the phase noise standard signal. This state data includes noise amplitude changes, frequency drift, and temporal fluctuations, reflecting the dynamic characteristics of the noise.

[0025] First, multiple phase noise standard signals are dynamically tracked, meaning their changes over time are monitored to obtain detailed data on the noise's changing state. This is typically achieved using time series analysis tools and dynamic tracking algorithms. For example, using the Python pandas library to record real-time signal fluctuations, combined with the frequency tracking methods of the scipy.signal library, it is possible to continuously monitor the real-time changes in the noise signal at specific frequencies and amplitudes.

[0026] The primary task of dynamic tracking is to build a data stream that receives and processes phase noise signal changes in real time. For example, consider a phase noise signal with an amplitude fluctuating between 0.1 and 1, and a frequency drifting between 100 Hz and 200 Hz. Dynamic tracking can capture the signal's frequency fluctuations and record the amplitude of the oscillation over time. Whenever the noise signal exceeds a specific threshold or deviates from its initial state, the tracking system records this change as part of the noise state, generating a time series of noise data.

[0027] When constructing noise variation state data, a sliding window algorithm is used to improve data accuracy. For example, a 10-second sliding window is used to continuously collect noise signal data, enabling analysis of noise variation patterns within each 10-second interval. This time series data is then fed into a time series model in a machine learning library such as statsmodels or scikit-learn to generate state data containing noise characteristics, reflecting the noise's fluctuation trends and characteristic changes.

[0028] By dynamically tracking and constructing noise change state data, the real-time state changes of the phase noise signal can be obtained, thus providing a key reference for subsequent noise prediction, filtering, and suppression.

[0029] Step S300: performing noise prediction on the oscillator according to the noise change state data to obtain a phase noise prediction result, filtering the multiple phase noise standard signals based on the phase noise prediction result to determine noise filtering data.

[0030] Specifically, phase noise prediction refers to the future noise characteristics inferred by a prediction algorithm based on noise variation data. This helps determine the noise characteristics of the oscillator at a future time. Noise filtering data refers to the data generated after filtering based on the predicted phase noise information. It is used to actually suppress and reduce the oscillator's phase noise.

[0031] First, based on the collected noise variation data, the oscillator's noise trend is predicted to obtain an estimate of future phase noise variations (i.e., the phase noise prediction result). Machine learning time series prediction algorithms, such as the autoregressive integrated moving average (ARIMA) model or long short-term memory (LSTM) network, are typically used to process noise time series data.

[0032] Once the phase noise prediction results are obtained, the next step is to filter multiple phase noise standard signals based on these results. Filtering is the process of suppressing noise fluctuations in a specific frequency band or amplitude, typically performed using a Kalman filter or a low-pass filter. Kalman filtering is a dynamic filtering method particularly suitable for smoothing time series prediction data.

[0033] Using the low-pass filter design in the Python scipy.signal library, we set the filter cutoff frequency (for example, to 2.5kHz) and then perform real-time filtering on the phase noise standard signal. This filtering significantly reduces the noise amplitude above 2.5kHz, resulting in a cleaner signal. The filtered data, called noise-filtered data, contains the signal information after noise suppression.

[0034] Through noise prediction and filtering based on noise variation data, phase noise fluctuation trends can be identified and eliminated in advance. This not only improves the stability of the oscillator signal, but also enhances the system's anti-interference capabilities, making phase noise suppression more accurate and adaptive.

[0035] Step S400: Perform a phase-locked loop based on the noise filtering data to generate a noise phase difference value, construct a multi-stage filtering structure, and perform multi-layer noise suppression on the multiple phase noise standard signals through the multi-stage filtering structure combined with the noise phase difference value to obtain a noise suppression result.

[0036] Specifically, a phase-locked loop (PLL) is a feedback control system that synchronizes the phase of the output signal with the input signal by comparing it with and locking its phase. The PLL is used to detect and adjust the phase difference of the noise signal to achieve phase alignment. The noise phase difference refers to the difference between the phase of the input signal and the phase of the reference signal during the phase locking process. By calculating this phase difference, the fluctuation trend in the noise signal can be detected. A multi-stage filtering structure consists of multiple filters connected sequentially or in parallel. Each filter stage gradually suppresses noise components of different frequencies or phases in the noise signal, ultimately resulting in a smoother and more stable output signal. Multi-layer noise suppression refers to the use of multiple filter stages to suppress noise in layers. Each layer suppresses noise with different characteristics, thereby achieving a more detailed noise reduction effect.

[0037] First, a phase-locked loop (PLL) is applied to the noise-filtered data generated by the oscillator. This loop ensures the phase of the signal is consistent. It first compares the phase of the noise-filtered data with a reference signal to obtain a noise phase difference. For example, when detecting a communication signal, this phase difference can reveal the real-time noise trend and help identify the direction of signal deviation. The core of the PLL consists of a phase detector and a low-pass filter, which together capture phase differences in the signal and adjust them to maintain lock.

[0038] Next, based on the noise phase difference, a multi-stage filtering structure is constructed to achieve hierarchical suppression of the phase noise signal. Specifically, the multi-stage filtering structure can be divided into three filtering stages: low frequency, medium frequency, and high frequency. Each filter stage weakens the noise of different frequency components. For example, the first layer of low-frequency filtering can filter out the disturbance of the low-frequency part through the Butterworth low-pass filter, while the subsequent medium frequency and high frequency filters suppress the noise in other frequency bands respectively. Multiple filters are connected layer by layer or in parallel to ensure that noise components of different frequencies can be effectively suppressed. The scipy.signal library in Python provides tools for designing multi-stage filters, such as the butter function for designing Butterworth filters, and the filtfilt function for bidirectional filtering, thereby obtaining a phase-shift-free filtering effect. On the basis of multi-stage filtering, the filtering results of each stage are corrected by calculating the noise phase difference, thereby enhancing the filtering effect.

[0039] The combination of a phase-locked loop and a multi-stage filtering structure achieves multi-level suppression of oscillator noise signals, effectively improving signal stability and accuracy. The phase-locked loop controls the signal phase, making the noise signal more consistent, while the multi-stage filtering structure ensures effective suppression of noise across different frequency bands.

[0040] Step S500: performing suppression detection based on the noise suppression result, feeding back the suppression effect and formulating a suppression strategy to adaptively suppress the phase noise of the oscillator.

[0041] Specifically, suppression detection refers to analyzing the suppressed signal to verify the noise suppression effect. The effectiveness of noise suppression is confirmed by detecting the difference between the suppressed signal and the original signal. Feedback suppression effect refers to transmitting the suppression detection result (i.e. the effect after noise suppression) back to the control system or filter as a basis for adjustment and optimization. Through the feedback mechanism, the system can be further adjusted and optimized according to the actual effect to improve the noise suppression performance. Suppression strategy refers to formulating corresponding control measures based on the detection results and feedback signals to further optimize the noise suppression process. This includes adjusting filter parameters, changing filtering methods, or enhancing noise suppression in specific frequency bands. Adaptive suppression means that the system automatically adjusts the suppression strategy according to real-time feedback signals and environmental changes to achieve the best noise suppression effect.

[0042] First, the signal, after undergoing multi-stage filtering and noise suppression, is tested to evaluate the effectiveness of noise suppression. This can be done by calculating the signal-to-noise ratio (SNR) improvement, performing spectral analysis, or measuring the difference between the noise amplitude after suppression and the original signal. If the test results indicate poor noise suppression or excessive noise, the system triggers a feedback mechanism. The measured suppression effect is fed back to the control system or filter as a feedback signal. This feedback signal can be used to guide filter adjustments.

[0043] Based on this feedback, new suppression strategies are developed. Specifically, if the noise suppression effect in a particular frequency band is poor, a new strategy will be implemented, such as increasing the filtering strength in that frequency band, changing the filtering algorithm, or optimizing the filter parameters. Furthermore, the suppression strategy can be adjusted based on the system's real-time detection data to better suit the current noise characteristics. Combining feedback and detection results, the system can adaptively adjust the noise suppression strategy to achieve dynamic optimization.

[0044] Through this adaptive suppression process, the system can dynamically adjust its noise suppression strategy based on real-time noise conditions. Compared with static noise suppression solutions, it can more effectively cope with various complex noise environments and ensure that the system maintains high-quality signal output.

[0045] Furthermore, the phase noise source of the oscillator is traversed to perform noise identification and generate multiple phase noise signals. The method includes: identifying the phase noise source of the oscillator, collecting noise according to the phase noise source of the oscillator, and obtaining multiple noise initial signals; performing spectrum analysis based on the multiple noise initial signals, extracting features of the multiple noise initial signals according to the noise spectrum information, and generating multiple noise features; clustering the multiple noise initial signals according to the multiple noise features to determine multiple phase noise classes; and separating and processing the multiple noise initial signals according to the multiple phase noise classes to generate the multiple phase noise signals.

[0046] Specifically, the initial noise signal refers to the raw signal collected from different phase noise sources, typically containing unprocessed noise data. These signals form the basis for further analysis and processing. Spectral analysis involves analyzing the signal's characteristics in the frequency domain to identify the frequency distribution of the noise. Spectral analysis can reveal the primary frequency bands and intensity variations of the noise signal and is typically performed using a spectrum analyzer or FFT technology. Noise signatures refer to the characteristic parameters extracted from each initial noise signal through spectral analysis, including frequency range, amplitude, and phase characteristics. These characteristics describe the properties of the noise signal. Cluster analysis is a data analysis method used to group noise signals with similar characteristics into the same category. Initial noise signals can be classified into different phase noise categories for more targeted processing. Phase noise categories refer to the noise type groups determined after cluster analysis, encompassing specific types of phase noise signals, such as high-frequency, medium- and low-frequency noise. Each phase noise category represents a specific noise characteristic.

[0047] First, identify the various phase noise sources of the oscillator. This involves using noise acquisition equipment (such as a spectrum analyzer or digital oscilloscope) to acquire signals from each noise source, thereby obtaining multiple initial noise signals. For example, thermal noise around the oscillator, electromagnetic interference, and the device's own RF noise are all collected as different noise sources. The signal from each noise source will exhibit different frequency characteristics.

[0048] Next, spectrum analysis is performed on these initial noise signals. By converting the signals from the time domain to the frequency domain, the frequency characteristics of the different noise signals can be extracted. For example, using FFT technology, the amplitude distribution of each noise signal at different frequencies can be obtained. This allows a series of noise characteristics to be extracted, such as high-frequency noise having a larger amplitude and low-frequency noise having a smaller amplitude. The frequency and amplitude characteristics of each noise signal serve as the main parameters of its noise characteristics.

[0049] Next, by clustering these noise features, noise signals with similar characteristics can be grouped into different phase noise classes. For example, the K-means clustering algorithm can be used to group noise signals with similar frequency distributions into the same class, thereby generating multiple phase noise classes. Different types of noise, such as high-frequency noise, medium-frequency noise, and low-frequency noise, can be classified into corresponding noise classes based on their frequency characteristics.

[0050] Finally, the original noise signal is separated according to these phase noise categories, and the signal within each category is extracted, thereby generating multiple independent phase noise signals. For example, the high-frequency noise signal can be separated from the total noise signal to generate an independent high-frequency phase noise signal. This allows for targeted optimization and suppression of different noise signal types in subsequent processing.

[0051] This process enables precise identification and classification of oscillator phase noise sources, breaking down complex noise signals into multiple phase noise classes with similar characteristics. This allows each noise signal to be processed individually, enabling more efficient filtering and suppression of noise of varying frequency bands and characteristics, significantly improving the stability and clarity of the oscillator's output signal.

[0052] Furthermore, the multiple phase noise signals are standardized to determine multiple phase noise standard signals. The method includes: performing amplitude standardization on the multiple phase noise signals according to the multiple noise characteristics to generate a noise amplitude range; performing frequency standardization on the multiple phase noise signals according to the noise spectrum information to generate a noise frequency distribution range; performing time alignment standardization on the multiple phase noise signals according to the signal acquisition timing to generate a synchronization time window; performing consistency verification based on the noise amplitude range, the noise frequency distribution range, and the synchronization time window, and when the verification is passed, determining the multiple phase noise standard signals.

[0053] Specifically, amplitude normalization refers to adjusting the amplitude of phase noise signals to a uniform range for consistency comparison and analysis. Amplitude normalization is typically achieved through normalization or scaling. The noise amplitude range refers to the amplitude range of the noise signal after normalization, for example, adjusting all signal amplitudes to the range of 0 to 1. Frequency normalization refers to adjusting the frequency components of the phase noise signal to a specific frequency range. Frequency normalization often uses frequency normalization or other scaling methods to unify signals from different frequency bands into the same frequency range. The noise frequency distribution range refers to the frequency distribution range of the signal after frequency normalization. Time alignment normalization refers to synchronizing the signal time axis to ensure that different noise signals are aligned within the same time window. Time alignment can be performed using the signal acquisition sequence to ensure that the sampling points of each noise signal are consistent. The synchronization time window refers to the unified time range of multiple noise signals after normalization, ensuring that different signals are analyzed and compared on the same time basis. Consistency verification verifies the success of the normalization process for each phase noise signal, ensuring matching and consistency in amplitude, frequency, and time.

[0054] First, amplitude normalization is performed on multiple phase noise signals. For example, suppose some noise signals have amplitudes ranging from 0.1 to 5, while others have amplitudes ranging from 0 to 2. Using amplitude normalization tools (such as the sklearn.preprocessing library in Python), these amplitudes can be adjusted to the range of 0 to 1, forming a unified noise amplitude range. This amplitude normalization not only ensures amplitude consistency across the noise signals but also facilitates subsequent filtering and suppression.

[0055] Next, frequency normalization is performed based on the frequency distribution characteristics of the noise signal. This step usually requires mapping the frequency range of the signal to a uniform distribution interval. For example, using the frequency conversion function in the scipy.signal library, noise signals of different frequency bands (such as signals in the range of 1MHz to 10MHz and signals in the range of 10Hz to 100Hz) are adjusted to the same frequency range of 0 to 1kHz, thereby forming a consistent noise frequency distribution interval. Based on this, the signal is processed consistently in terms of frequency characteristics, allowing subsequent analysis to more accurately capture noise changes.

[0056] Next, the noise signals are time-aligned and normalized. This standardizes the timing of signal acquisition and aligns the different noise signals to the same synchronized time window. For example, if the acquisition time periods of different noise signals are not exactly the same, this step ensures that the noise signals are displayed in the same time period by aligning the timestamps. Time alignment and normalization can be achieved by resampling the signals to the same time base using interpolation tools in signal processing libraries such as NumPy.

[0057] After completing the aforementioned standardization process, consistency verification is performed to check the matching of the amplitude, frequency, and timing of each phase noise signal. This consistency verification ensures that all signals behave identically under the same amplitude, frequency, and timing standards. Once verified, these signals are confirmed as phase noise standard signals, providing reliable input data for subsequent noise suppression.

[0058] This standardization process aligns multiple phase noise signals with different characteristics to a unified amplitude, frequency, and time scale, making the noise signal characteristics clearer and more consistent. This not only improves the comparability of the noise signals but also provides a high-quality data foundation for subsequent filtering and noise suppression, thereby enhancing the accuracy and effectiveness of phase noise suppression.

[0059] Furthermore, the oscillator is subjected to noise prediction based on the noise change state data to obtain a phase noise prediction result, and the method includes: performing data segmentation based on the noise change state data to generate a noise data training set and a noise data verification set; performing learning based on the noise data training set to generate a phase change trend graph for unsupervised training and constructing a phase noise prediction model; extracting a real-time noise data set based on the multiple phase noise standard signals, synchronizing the real-time noise data set to the phase noise prediction model for alternating prediction to obtain an initial noise matrix; performing data verification on the initial noise matrix using the noise data verification set to obtain a prediction error fluctuation value; and dynamically correcting the initial noise matrix according to the prediction error fluctuation value to obtain the phase noise prediction result.

[0060] Specifically, the noise data training set refers to a set of data sets generated from noise variation state data, which is used in the model training phase to help the model learn the characteristics and trends of noise. The noise data validation set refers to a portion of data segmented from the noise variation state data, which is used to verify the performance of the model and ensure the accuracy of the model's predictions on unseen data. The initial noise matrix refers to the noise data matrix obtained by preliminary calculation of the prediction model from the real-time noise dataset. It represents the preliminary prediction information of phase noise, where the rows of the matrix represent time series and the columns represent the state variables of phase offset and frequency offset. The prediction error fluctuation value refers to the difference between the predicted noise value and the actual noise value, which is used to calibrate the prediction model to improve prediction accuracy.

[0061] First, split the noise variation data into a training set and a validation set. This allows the model to learn noise trends and characteristics during the training phase, while the validation set is used to evaluate model performance. Data splitting can be performed using the train_test_split method in the Python scikit-learn library, which randomly divides the original data into an 80% training set and a 20% validation set. The training set is used for model learning, while the validation set is used for subsequent performance verification.

[0062] Next, unsupervised learning is performed on the training set, using an algorithm to generate a phase variation trend graph and construct a phase noise prediction model. Unsupervised algorithms such as K-means clustering or principal component analysis (PCA) can be used to capture the underlying patterns in the noise data and extract key features. K-means clustering groups the training data so that similar noise features are clustered together; PCA, on the other hand, is used for dimensionality reduction, converting multidimensional noise data into lower-dimensional data that is easier to predict.

[0063] After establishing the phase noise prediction model, real-time noise datasets extracted from multiple phase noise standard signals are fed into the prediction model for alternating predictions to generate an initial noise matrix. For example, assuming that each time point in the real-time noise dataset contains the amplitude and frequency characteristics of the phase noise, by feeding these characteristics into the prediction model, the expected value of the phase noise at that time point can be obtained, thereby constructing an initial noise matrix encompassing multiple time points.

[0064] Next, the initial noise matrix is validated using the validation set to obtain the prediction error fluctuation value. This step evaluates the accuracy of the model by comparing the deviation between the predicted and actual values, using metrics such as mean squared error (MSE) or mean absolute error (MAE). A high error fluctuation value indicates that the model needs correction.

[0065] Finally, the initial noise matrix is dynamically corrected based on the fluctuations in the prediction error to ensure more accurate phase noise predictions. Correction methods typically include techniques such as Kalman filtering or exponentially weighted averaging, which smooth the predictions and suppress short-term fluctuations.

[0066] This step enables accurate prediction of oscillator phase noise. The real-time extraction of noise datasets and the dynamic correction process enable the prediction model to better cope with the randomness and timing characteristics of noise, thereby improving the model's adaptability and prediction accuracy.

[0067] Furthermore, a real-time noise data set is extracted based on the multiple phase noise standard signals, and the real-time noise data set is synchronized to the phase noise prediction model for alternating prediction to obtain an initial noise matrix. The method includes: feature labeling of the multiple phase noise standard signals based on the multiple noise features to determine multiple noise reference benchmarks; comparing and matching the multiple phase noise standard signals according to the multiple noise reference benchmarks to obtain multiple similarity scores, and extracting the real-time noise data set based on the multiple similarity scores; synchronizing the real-time noise data set to the phase noise prediction model for offset prediction to obtain an offset prediction result, and the offset prediction result includes a phase offset prediction result and a frequency offset prediction result; feeding back the phase offset prediction result and the frequency offset prediction result to the phase noise prediction model for alternating calculation and update to obtain state variables of phase offset and state variables of frequency offset; arranging the state variables of phase offset and the state variables of frequency offset according to the signal acquisition timing to construct the initial noise matrix.

[0068] Specifically, the noise reference refers to a set of benchmark values determined based on noise characteristics, used for comparison with real-time data to determine the characteristics and similarity of real-time noise signals. The similarity score is the score obtained by comparing real-time noise characteristics with the noise reference. It is used to quantify signal similarity and facilitate the extraction of highly relevant real-time noise data. The offset prediction result refers to information containing phase and frequency offsets. The prediction model calculates the noise trend based on these offsets. State variables are variables that represent the system state in the prediction model. They describe the phase and frequency states of the noise signal over time and are used for further noise prediction and suppression.

[0069] First, we annotate the noise characteristics of multiple phase noise standard signals and determine multiple noise reference benchmarks. For example, we can use statistics such as mean and variance as benchmarks for subsequent similarity matching. This annotation process can yield multiple benchmarks, such as the "typical offset value" or "frequency fluctuation range" of phase noise.

[0070] Next, each phase noise signal is compared and matched against the reference benchmarks based on these benchmarks to generate a similarity score, which is then used to extract the real-time noise dataset. For example, if the reference benchmark shows a more significant noise signal within a specific frequency range, the real-time data closest to the benchmark is assigned a higher similarity score and selected for inclusion in the real-time noise dataset. Similarity can be measured using algorithms such as Euclidean distance or cosine similarity. For example, the cosine_similarity function in Python can help calculate similarity scores.

[0071] The extracted real-time noise data is then synchronized with the phase noise prediction model to perform offset prediction and obtain the offset prediction results. During this prediction process, the model calculates the offset trends of phase and frequency.

[0072] Based on the obtained phase and frequency offsets, these results are fed back into the phase noise prediction model for alternating computational updates, thereby forming new state variables. This feedback process, through multiple alternating updates, gradually approximates the true noise trend. Specifically, methods such as Kalman filters or recurrent neural networks (RNNs) can be used to predict and update time series data to improve noise prediction accuracy.

[0073] Finally, all state variables are arranged according to the time sequence of signal acquisition to generate an initial noise matrix. This matrix, composed of state variables arranged in chronological order, fully reflects the temporal evolution of noise, laying the foundation for subsequent filtering and suppression. The initial noise matrix not only contains information about the current noise state but also includes predicted trends, providing data support for subsequent suppression operations.

[0074] This step enables more accurate data extraction and trend capture in oscillator phase noise prediction. The application of noise benchmarks and similarity scores makes the extracted real-time noise dataset more representative and relevant. The alternating updating of offset prediction results and state variables improves adaptability and prediction accuracy for phase and frequency offsets.

[0075] Furthermore, a phase-locked loop is performed according to the noise filtering data to generate a noise phase difference and construct a multi-stage filtering structure. The method includes: setting a locked target signal, performing phase comparison between the noise filtering signal and the locked target signal according to the noise filtering data to obtain a noise phase difference; performing an oscillator adjustment on the noise phase difference through a feedback loop of the phase-locked loop to generate a phase difference signal; performing dynamic filtering adjustment based on the phase difference signal in combination with Kalman filtering to construct a first-stage filter; dynamically adjusting the weights of the first-stage filter according to the noise phase difference to construct a second-stage filter; performing long-term and short-term timing analysis based on the second-stage filter, and constructing a third-stage filter according to the analysis results; and associating and integrating the first-stage filter, the second-stage filter, and the third-stage filter to construct the multi-stage filtering structure.

[0076] Specifically, the locked target signal is the reference signal used in a phase-locked loop for comparison with the noise signal. The target signal is typically a stable signal with a known frequency and phase. As the desired output in the phase-locked loop, the phase of the noise signal is compared with this target signal and adjusted accordingly. The noise phase difference is the difference between the target and noise signal phases, representing the phase deviation between the noise and target signals. By calculating this phase difference, the signal can be effectively adjusted to eliminate the phase error. The Kalman filter is a dynamic filtering method based on a linear system model that denoises signals by estimating the system state. The Kalman filter uses historical system data to infer the current state, thereby smoothing and predicting dynamic systems. It is widely used in fields such as signal processing and navigation. The primary, secondary, and tertiary filters represent multiple filtering stages, each processing noise signals with different characteristics. The primary filter performs preliminary signal processing, the secondary filter further adjusts the filtered signal, and the tertiary filter is used for more complex time series signal analysis. Long-short-term time series analysis (LSTM) is a neural network-based time series data prediction method that can learn long-term dependencies in data. LSTM is often used to model time series data. It can capture long-term and short-term information and help make accurate noise predictions.

[0077] First, a target signal (a known, stable signal) must be set as the baseline for feedback adjustments. During signal processing, the noise-filtered signal is compared with the target signal in phase, and the noise phase difference is calculated. For example, in a communications system, the target signal is a stable carrier signal, while the noise signal is a disturbance signal with phase drift. Calculating the phase difference reveals the signal deviation caused by the noise.

[0078] Next, through a phase-locked loop, a feedback control mechanism is used to feed the noise phase difference into the system for oscillator adjustment, minimizing the phase error and generating a phase difference signal. This process is implemented through a phase detector and loop filter, allowing the oscillator's phase to be adjusted in real time to align with the target signal.

[0079] Next, a Kalman filter is used to dynamically filter and adjust the generated phase difference signal, constructing a first-stage filter. The Kalman filter uses historical data and the current noise phase difference to predict future states, thereby smoothing out random fluctuations in the noise signal. This provides an effective noise removal method, reducing errors caused by transient noise.

[0080] The output signal of the first-stage filter then enters the second-stage filter, where the signal weights are dynamically adjusted. This stage primarily relies on changes in the noise phase difference, combined with the filter's adaptive characteristics, to dynamically adjust the filter parameters for a smoother signal. The second-stage filter operates by adjusting the weights based on real-time feedback signals, enabling it to better cope with noise of varying frequencies and amplitudes.

[0081] During further processing, a third-stage filter is constructed based on the output of the second-stage filter and combined with a long-short-term time series analysis (LSTM). LSTM can capture long-term dependencies in time series data and is particularly suitable for processing noisy data with temporal correlation. By analyzing the temporal changes in noise, the third-stage filter can predict future noise trends and perform corresponding filtering to achieve more precise noise suppression.

[0082] Finally, the first, second, and third filters are linked and integrated to form a multi-stage filtering structure. Through the coordinated work of each level of filtering, noise signals are suppressed in multiple dimensions and at multiple levels. This multi-stage filtering structure can effectively filter out noise across various frequency bands, especially those that vary over time.

[0083] By implementing this multi-stage filtering structure, the oscillator's phase noise can be effectively processed and suppressed within different frequency bands and time windows. Dynamic adjustments using the Kalman filter smooth the signal and reduce errors caused by transient interference, while the use of the LSTM captures both long-term and short-term dependencies in the time series, effectively predicting and processing complex noise patterns. Ultimately, the combined processing of the multi-stage filters significantly suppresses the oscillator's output phase noise, thereby improving system stability and accuracy.

[0084] Furthermore, multi-layer noise suppression is performed on the multiple phase noise standard signals through the multi-stage filtering structure in combination with the noise phase difference value to obtain a noise suppression result. The method includes: inputting the multiple phase noise standard signals into the first-stage filter for high-frequency noise processing according to the noise phase difference value to generate a first suppression result, and the first suppression result includes a first suppression effect; synchronizing the first suppression result to the second-stage filter for medium and low-frequency noise processing, updating the first suppression result, and obtaining a second suppression result, and the second suppression result includes a second suppression effect; synchronizing the second suppression result to the third-stage filter for long-term noise processing, updating the second suppression result, and obtaining a third suppression result, and the third suppression result includes a third suppression effect; integrating the first suppression result, the second suppression result, and the third suppression result according to the first suppression effect, the second suppression effect, and the second suppression effect to obtain the noise suppression result.

[0085] Specifically, high-frequency noise processing refers to suppressing noise in the high-frequency portion of a signal. High-frequency noise often originates from rapidly changing interference in the system. Addressing this noise can improve signal stability, especially in high-frequency applications (such as RF signals). Mid- and low-frequency noise processing refers to suppressing mid- and low-frequency noise in a signal. This noise originates from sources such as power supply fluctuations or vibration within the device itself. Its low frequency often has a long-term impact on system stability. Long-term noise processing refers to suppressing slowly varying noise in a signal. This noise has a long time period and is caused by environmental changes or long-term signal drift. This type of noise is often difficult to eliminate quickly and requires long-term tracking and adjustment to smooth it out. The first, second, and third suppression results are the signal results after processing through different filters (primary, secondary, and tertiary filters). Each filter stage suppresses noise in different frequency bands, creating multiple suppression effects.

[0086] First, multiple phase noise standard signals are fed into a first-stage filter using noise phase differences to remove high-frequency noise. This filter removes higher-frequency noise components from the signal, typically caused by external disturbances or equipment noise.

[0087] The first suppression result is then synchronously input into a secondary filter for mid- and low-frequency noise processing. The secondary filter eliminates mid- and low-frequency noise, which typically arises from factors such as power supply fluctuations and temperature changes caused by long-term device operation. Based on the characteristics of the input signal, the secondary filter dynamically adjusts the filtering process, updates the signal, and generates a second suppression result. This result includes the signal after the mid- and low-frequency noise is removed, as well as the secondary suppression effect after noise removal.

[0088] On this basis, the second suppression result is synchronously input into the third-stage filter for long-term noise processing. The third-stage filter suppresses slowly changing noise in the signal by analyzing the long-term changes of the signal.

[0089] Finally, the first, second, and third suppression results are integrated based on their respective suppression effects (first, second, and third suppression effects) to obtain the final noise suppression result. This process ensures that noise from high to low frequencies and from short to long durations is effectively suppressed by weighted integration of the suppression effects of each filter level, ultimately outputting a noise-minimized signal.

[0090] This multi-stage filtering structure effectively suppresses noise with different frequency bands and time characteristics, improving overall system performance. By classifying noise into different frequency bands and time series characteristics and processing them layer by layer, the system can accurately eliminate various noise components and provide a high-quality, noise-minimized output signal.

[0091] Furthermore, after performing multi-layer noise suppression on the multiple phase noise standard signals through the multi-stage filtering structure in combination with the noise phase difference, the method includes: performing layer-by-layer suppression feedback on the first-stage filter, the second-stage filter, and the third-stage filter through the feedback loop using the first suppression effect, the second suppression effect, and the second suppression effect to obtain a first-stage optimized filter, a second-stage optimized filter, and a third-stage optimized filter; updating and optimizing the multi-stage filtering structure according to the first-stage optimized filter, the second-stage optimized filter, and the third-stage optimized filter to obtain a multi-stage optimized filtering structure.

[0092] Specifically, a feedback loop refers to a portion of a system's output signal being returned to its input, typically used to adjust system behavior or improve system performance. In noise suppression methods, feedback loops are used to provide feedback on the processing results of each filter stage, allowing the filter to be adjusted based on actual performance and further optimizing noise suppression capabilities. A multi-stage optimized filtering structure is a filtering system composed of multiple optimized filters, including primary, secondary, and tertiary filters. By integrating these optimized filters into a single system, different types of noise can be more comprehensively processed, achieving efficient noise suppression.

[0093] After multi-layer noise suppression—that is, after the initial noise suppression by the primary, secondary, and tertiary filters—the system transmits feedback on the suppression effects of the first, second, and third filters through a feedback loop. These feedback signals reflect the processing effectiveness of each filter layer and provide a basis for improvement. The primary, secondary, and tertiary filters adjust their operating conditions based on the feedback signals. For example, if the primary filter performs poorly at high-frequency noise, the feedback signal will prompt it to improve its suppression of high-frequency noise. Similarly, the secondary and tertiary filters will optimize their handling of low-frequency and long-duration noise based on the feedback signal. After this feedback optimization process, the resulting filters are optimized, including the primary, secondary, and tertiary filters. These optimized filters are more accurate in processing noise across different frequency bands and can adapt to environmental changes.

[0094] Finally, the system updates and optimizes the original multi-stage filter structure based on the optimized filter, forming a multi-stage optimized filter structure. This optimized structure can more effectively suppress different types of noise, ensuring higher quality of the final output signal.

[0095] By optimizing the filter layer by layer through a feedback loop, the accuracy and effectiveness of noise suppression can be significantly improved. The optimized filter can accurately suppress different types of noise, making it particularly suitable for complex signal environments and applications requiring high precision.

[0096] In summary, the method for suppressing oscillator phase noise provided by the embodiments of the present application has the following technical effects:

[0097] 1. By traversing the oscillator's phase noise sources for noise identification and generating multiple phase noise signals, the characteristics of different noise sources can be accurately captured and extracted. This method improves the accuracy of noise source identification, provides a reliable data foundation for subsequent noise processing and suppression, and optimizes the oscillator's phase noise suppression effect.

[0098] 2. By dynamically tracking and predicting noise variation data, the oscillator's future phase noise trends can be accurately predicted. This method utilizes real-time noise data sets for alternating prediction and error correction, effectively improving the accuracy of phase noise prediction and providing a more accurate basis for noise suppression.

[0099] 3. By constructing a multi-stage filtering structure and combining noise phase difference for noise suppression, the oscillator's phase noise suppression capability is effectively improved. This method uses a phase-locked loop and Kalman filter for dynamic adjustment, ensuring efficient noise suppression across different frequency bands, significantly improving the system's signal quality.

[0100] Embodiment 2 is based on the same inventive concept as the method for suppressing oscillator phase noise in the above embodiment. Figure 2 As shown, an embodiment of the present application provides a device for suppressing oscillator phase noise, the device comprising:

[0101] A phase noise standard signal determination module 11 is used to traverse the phase noise source of the oscillator to identify noise, generate multiple phase noise signals, standardize the multiple phase noise signals, and determine multiple phase noise standard signals; a noise change state data construction module 12 is used to perform dynamic tracking based on the multiple phase noise standard signals and construct noise change state data; a noise filtering data determination module 13 is used to predict the noise of the oscillator according to the noise change state data, obtain a phase noise prediction result, filter the multiple phase noise standard signals based on the phase noise prediction result, and determine noise filtering data; a noise suppression result acquisition module 14 is used to perform a phase-locked loop according to the noise filtering data, generate a noise phase difference, construct a multi-stage filtering structure, and perform multi-layer noise suppression on the multiple phase noise standard signals through the multi-stage filtering structure in combination with the noise phase difference to obtain a noise suppression result; an adaptive suppression module 15 is used to perform suppression detection based on the noise suppression result, feedback the suppression effect, and formulate a suppression strategy to adaptively suppress the phase noise of the oscillator.

[0102] Furthermore, the phase noise standard signal determination module 11 is further configured to perform the following steps:

[0103] Identify the phase noise source of the oscillator, collect noise according to the phase noise source of the oscillator, and obtain multiple noise initial signals; perform spectrum analysis based on the multiple noise initial signals, extract features of the multiple noise initial signals based on noise spectrum information, and generate multiple noise features; perform cluster analysis on the multiple noise initial signals according to the multiple noise features to determine multiple phase noise classes; separate and process the multiple noise initial signals according to the multiple phase noise classes to generate the multiple phase noise signals.

[0104] Furthermore, the phase noise standard signal determination module 11 is further configured to perform the following steps:

[0105] The multiple phase noise signals are amplitude-standardized according to the multiple noise characteristics to generate a noise amplitude range; the multiple phase noise signals are frequency-standardized according to the noise spectrum information to generate a noise frequency distribution range; the multiple phase noise signals are time-aligned and standardized according to the signal acquisition timing to generate a synchronization time window; consistency verification is performed based on the noise amplitude range, the noise frequency distribution range, and the synchronization time window, and when the verification is passed, the multiple phase noise standard signals are determined.

[0106] Furthermore, the noise filtering data determination module 13 is further configured to perform the following steps:

[0107] Data segmentation is performed based on the noise change state data to generate a noise data training set and a noise data verification set; learning is performed based on the noise data training set to generate a phase change trend graph for unsupervised training and to construct a phase noise prediction model; a real-time noise data set is extracted based on the multiple phase noise standard signals, and the real-time noise data set is synchronized to the phase noise prediction model for alternating prediction to obtain an initial noise matrix; data verification is performed on the initial noise matrix using the noise data verification set to obtain a prediction error fluctuation value; the initial noise matrix is dynamically corrected according to the prediction error fluctuation value to obtain the phase noise prediction result.

[0108] Furthermore, the noise filtering data determination module 13 is further configured to perform the following steps:

[0109] Based on the multiple noise features, the multiple phase noise standard signals are feature-labeled to determine multiple noise reference benchmarks; the multiple phase noise standard signals are compared and matched according to the multiple noise reference benchmarks to obtain multiple similarity scores, and a real-time noise data set is extracted according to the multiple similarity scores; the real-time noise data set is synchronized to the phase noise prediction model for offset prediction to obtain an offset prediction result, and the offset prediction result includes a phase offset prediction result and a frequency offset prediction result; the phase offset prediction result and the frequency offset prediction result are fed back to the phase noise prediction model for alternating calculation and update to obtain state variables of phase offset and state variables of frequency offset; the state variables of phase offset and the state variables of frequency offset are arranged according to the signal acquisition timing to construct the initial noise matrix.

[0110] Furthermore, the noise suppression result obtaining module 14 is further configured to perform the following steps:

[0111] A locked target signal is set, and a phase comparison is performed between the noise filtered signal and the locked target signal according to the noise filtering data to obtain a noise phase difference; an oscillator adjustment is performed on the noise phase difference through a feedback loop of a phase-locked loop to generate a phase difference signal; dynamic filtering adjustment is performed based on the phase difference signal in combination with Kalman filtering to construct a first-stage filter; dynamic adjustment of weights is performed according to the noise phase difference based on the first-stage filter to construct a second-stage filter; long-term and short-term timing analysis is performed based on the second-stage filter, and a third-stage filter is constructed according to the analysis results; the first-stage filter, the second-stage filter, and the third-stage filter are associated and integrated to construct the multi-stage filtering structure.

[0112] Furthermore, the noise suppression result obtaining module 14 is further configured to perform the following steps:

[0113] According to the noise phase difference value, the multiple phase noise standard signals are input into the first-stage filter for high-frequency noise processing to generate a first suppression result, and the first suppression result includes a first suppression effect; the first suppression result is synchronized to the second-stage filter for medium and low-frequency noise processing, and the first suppression result is updated to obtain a second suppression result, and the second suppression result includes a second suppression effect; the second suppression result is synchronized to the third-stage filter for long-term noise processing, and the second suppression result is updated to obtain a third suppression result, and the third suppression result includes a third suppression effect; the first suppression result, the second suppression result, and the third suppression result are integrated according to the first suppression effect, the second suppression effect, and the second suppression effect to obtain the noise suppression result.

[0114] Furthermore, the system is further configured to perform the following steps:

[0115] The first suppression effect, the second suppression effect, and the second suppression effect are used to perform layer-by-layer suppression feedback on the first-stage filter, the second-stage filter, and the third-stage filter through the feedback loop to obtain a first-stage optimized filter, a second-stage optimized filter, and a third-stage optimized filter; and the multi-stage filtering structure is updated and optimized according to the first-stage optimized filter, the second-stage optimized filter, and the third-stage optimized filter to obtain a multi-stage optimized filtering structure.

[0116] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0117] Furthermore, the terms "first" or "second" as described above not only represent an order relationship but also represent specific concepts and / or refer to the selectability of multiple elements, either individually or in combination. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, if such modifications and variations fall within the scope of this application and its equivalents, this application is intended to include such modifications and variations.

Claims

1. A method for suppressing oscillator phase noise, characterized in that: The method comprises: traversing the phase noise sources of the oscillator to perform noise identification, generating multiple phase noise signals, performing standardization processing on the multiple phase noise signals, and determining multiple phase noise standard signals; Perform dynamic tracking based on the multiple phase noise standard signals to construct noise change state data; Performing noise prediction on the oscillator according to the noise change state data to obtain a phase noise prediction result, filtering the multiple phase noise standard signals based on the phase noise prediction result to determine noise filtering data; Performing a phase-locked loop according to the noise filtering data to generate a noise phase difference value, constructing a multi-stage filtering structure, and performing multi-layer noise suppression on the multiple phase noise standard signals through the multi-stage filtering structure in combination with the noise phase difference value to obtain a noise suppression result; Based on the noise suppression result, suppression detection is performed, the suppression effect is fed back, and a suppression strategy is formulated to adaptively suppress the phase noise of the oscillator.

2. The method for suppressing oscillator phase noise according to claim 1, wherein: Phase noise sources of an oscillator are traversed to perform noise identification and generate multiple phase noise signals, the method comprising: Identifying a phase noise source of the oscillator, collecting noise according to the phase noise source of the oscillator, and obtaining a plurality of initial noise signals; Performing spectrum analysis on the multiple noise initial signals, extracting features from the multiple noise initial signals according to noise spectrum information, and generating multiple noise features; Performing cluster analysis on the multiple noise initial signals according to the multiple noise features to determine multiple phase noise classes; The multiple noise initial signals are separated and processed according to the multiple phase noise types to generate the multiple phase noise signals.

3. The method for suppressing oscillator phase noise according to claim 2, wherein: The method includes performing standardization processing on the multiple phase noise signals to determine multiple phase noise standard signals: Normalizing the amplitudes of the multiple phase noise signals according to the multiple noise characteristics to generate noise amplitude intervals; Normalizing the frequencies of the multiple phase noise signals according to the noise spectrum information to generate a noise frequency distribution interval; Performing time alignment and standardization on the multiple phase noise signals according to the signal acquisition timing to generate a synchronization time window; A consistency verification is performed based on the noise amplitude interval, the noise frequency distribution interval, and the synchronization time window, and when the verification passes, the multiple phase noise standard signals are determined.

4. The method for suppressing oscillator phase noise according to claim 3, wherein: The oscillator is subjected to noise prediction according to the noise change state data to obtain a phase noise prediction result, the method comprising: Performing data segmentation based on the noise change state data to generate a noise data training set and a noise data verification set; Learning is performed based on the noise data training set to generate a phase change trend graph for unsupervised training and to construct a phase noise prediction model; Extracting a real-time noise dataset based on the multiple phase noise standard signals, synchronizing the real-time noise dataset to the phase noise prediction model for alternating prediction, and obtaining an initial noise matrix; Using the noise data verification set to perform data verification on the initial noise matrix to obtain a prediction error fluctuation value; The initial noise matrix is dynamically corrected according to the prediction error fluctuation value to obtain the phase noise prediction result.

5. The method for suppressing oscillator phase noise according to claim 4, wherein: Extracting a real-time noise dataset based on the multiple phase noise standard signals, synchronizing the real-time noise dataset to the phase noise prediction model for alternating prediction, and obtaining an initial noise matrix, the method comprising: Characterizing the multiple phase noise standard signals based on the multiple noise characteristics to determine multiple noise reference benchmarks; performing comparison and matching of the plurality of phase noise standard signals according to the plurality of noise reference benchmarks to obtain a plurality of similarity scores, and extracting a real-time noise dataset based on the plurality of similarity scores; Synchronizing the real-time noise data set to the phase noise prediction model to perform offset prediction to obtain an offset prediction result, wherein the offset prediction result includes a phase offset prediction result and a frequency offset prediction result; Feedback of the phase offset prediction result and the frequency offset prediction result to the phase noise prediction model is performed for alternating calculation and update to obtain a state variable of the phase offset and a state variable of the frequency offset; The state variables of the phase shift and the state variables of the frequency shift are arranged according to the signal acquisition timing to construct the initial noise matrix.

6. The method for suppressing oscillator phase noise according to claim 1, wherein: A phase-locked loop is performed according to the noise filtering data to generate a noise phase difference value and construct a multi-stage filtering structure, the method comprising: Setting a locked target signal, performing a phase comparison between the noise filtered signal and the locked target signal according to the noise filtering data, and obtaining a noise phase difference; Performing an oscillator adjustment on the noise phase difference through a feedback loop of a phase-locked loop to generate a phase difference signal; Perform dynamic filtering adjustment based on the phase difference signal in combination with Kalman filtering to construct a first-level filter; Dynamically adjust the weights of the first-stage filter according to the noise phase difference to construct a second-stage filter; Performing long-term and short-term time series analysis based on the two-stage filter, and constructing a three-stage filter according to the analysis results; The first-stage filter, the second-stage filter, and the third-stage filter are associated and integrated to construct the multi-stage filtering structure.

7. The method for suppressing oscillator phase noise according to claim 6, wherein: The method includes: performing multi-layer noise suppression on the multiple phase noise standard signals by combining the multi-stage filtering structure with the noise phase difference value to obtain a noise suppression result. Inputting the multiple phase noise standard signals into the first-stage filter for high-frequency noise processing according to the noise phase difference value to generate a first suppression result, wherein the first suppression result includes a first suppression effect; Synchronizing the first suppression result to the secondary filter for low- and medium-frequency noise processing, updating the first suppression result to obtain a second suppression result, where the second suppression result includes a second suppression effect; Synchronizing the second suppression result to the three-stage filter for long-time noise processing, updating the second suppression result to obtain a third suppression result, wherein the third suppression result includes a third suppression effect; The first suppression result, the second suppression result, and the third suppression result are integrated according to the first suppression effect, the second suppression effect, and the third suppression effect to obtain the noise suppression result.

8. The method for suppressing oscillator phase noise according to claim 7, wherein: After performing multi-layer noise suppression on the multiple phase noise standard signals by combining the multi-stage filtering structure with the noise phase difference value, the method includes: The first suppression effect, the second suppression effect, and the second suppression effect are used to perform layer-by-layer suppression feedback on the first-stage filter, the second-stage filter, and the third-stage filter through the feedback loop to obtain a first-stage optimized filter, a second-stage optimized filter, and a third-stage optimized filter; The multi-stage filtering structure is updated and optimized according to the first-stage optimized filter, the second-stage optimized filter, and the third-stage optimized filter to obtain a multi-stage optimized filtering structure.

9. A device for suppressing oscillator phase noise, characterized in that: For executing the method according to any one of claims 1 to 8, the apparatus comprises: a phase noise standard signal determination module, configured to traverse the phase noise sources of the oscillator to perform noise identification, generate multiple phase noise signals, perform standardization processing on the multiple phase noise signals, and determine multiple phase noise standard signals; A noise change state data construction module, configured to perform dynamic tracking based on the multiple phase noise standard signals to construct noise change state data; a noise filtering data determining module, configured to perform noise prediction on the oscillator according to the noise change state data, obtain a phase noise prediction result, filter the multiple phase noise standard signals based on the phase noise prediction result, and determine noise filtering data; a noise suppression result acquisition module, configured to perform a phase-locked loop according to the noise filtering data, generate a noise phase difference value, construct a multi-stage filtering structure, and perform multi-layer noise suppression on the multiple phase noise standard signals through the multi-stage filtering structure in combination with the noise phase difference value to obtain a noise suppression result; The adaptive suppression module is used to perform suppression detection based on the noise suppression result, feedback the suppression effect and formulate a suppression strategy to adaptively suppress the phase noise of the oscillator.

Images