ICG signal denoising method and device

Abstract

The embodiment of the present application provides an ICG signal denoising method and device, and relates to the technical field of medical apparatuses.The above method comprises the following steps: obtaining a first ICG signal to be denoised, and obtaining a first time domain reference signal and a frequency domain reference signal; determining noise signals in the first ICG signal and the first time domain reference signal based on the frequency domain reference signal, removing the noise signals in the first ICG signal as a second ICG signal, and removing the noise signals in the first time domain reference signal as a second time domain reference signal; and determining target effective signals in the second ICG signal based on the second time domain reference signal, so as to realize ICG signal denoising. The scheme provided by the embodiment can accurately perform ICG signal denoising.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, and in particular to a method and apparatus for denoising ICG signals. Background Technology

[0002] Impedance cardiography (ICG) signals can be used for non-invasive, real-time, and continuous measurement of human hemodynamic parameters. However, because ICG signals are relatively weak and easily affected by noise, which impacts signal stability and feature extraction, a denoising scheme for ICG signals is urgently needed. Summary of the Invention

[0003] The purpose of this invention is to provide a method and apparatus for denoising ICG signals to accurately remove noise interference from ICG signals. The specific technical solution is as follows: In a first aspect, embodiments of the present invention provide an ICG signal denoising method, the method comprising: A first ICG (Impedance Cardiogra) signal to be denoised is obtained, and a first time-domain reference signal and a frequency-domain reference signal are obtained. The first time-domain reference signal is a signal that is correlated with the time-domain variation trend of the first ICG signal, and the frequency-domain reference signal is a signal that is correlated with the frequency-domain variation trend of the first ICG signal and the first time-domain reference signal. Based on the frequency domain reference signal, noise signals in the first ICG signal and the first time domain reference signal are determined respectively. The noise signals in the first ICG signal are removed to obtain the second ICG signal, and the noise signals in the first time-frequency reference signal are removed to obtain the second time domain reference signal. Based on the second time-domain reference signal, the target valid signal in the second ICG signal is determined in order to achieve ICG signal denoising.

[0004] In one embodiment of the present invention, determining the target valid signal in the second ICG signal based on the second time-domain reference signal includes: Based on the second time-domain reference signal and the second ICG signal, the initial valid signal in the second ICG signal is determined; Based on the initial valid signal and each signal component contained in the second ICG signal, the target valid signal in the second ICG signal is determined.

[0005] In one embodiment of the present invention, determining the target valid signal in the second ICG signal based on each signal component contained in the initial valid signal and the second ICG signal includes: Cluster each signal component contained in the second ICG signal to obtain a clustered set of each signal component; Based on the signal components contained in each signal component set and the initial valid signal, determine the valid signal of each signal component set; Based on the effective signals of the adjacent signal component sets of each signal component set, the effective signals of each signal component set are updated, and based on the updated effective signals, the target effective signal in the second ICG signal is determined.

[0006] In one embodiment of the present invention, the above-described clustering of each signal component contained in the second ICG signal to obtain a clustered set of each signal component includes: Based on the frequency segmentation of each signal component contained in the second ICG signal, the critical signal component that characterizes the boundary between the effective signal and the noise signal is determined in the signal component, and the effective signal set and the noise signal set are determined based on the critical signal component; The process of updating the effective signals of each signal component set based on the effective signals of its adjacent signal component sets, and determining the target effective signal in the second ICG signal based on the updated effective signals, includes: Based on the effective signal of the noise signal set, update the effective signal of the effective signal set, and determine the updated effective signal as the target effective signal in the second ICG signal.

[0007] In one embodiment of the present invention, determining the effective signal of each signal component set based on the signal components contained in each signal component set and the initial effective signal includes: For each signal component set, iterative reconstruction is performed according to the frequency band order of each signal component contained in the signal component set. In each iteration, the signal component targeted by the current iteration and the reconstructed signal obtained in the previous iteration are reconstructed, and the reconstructed signal obtained after each iteration is determined as the candidate signal. Calculate the correlation between the multiple candidate signals corresponding to each signal component set and the initial valid signal. Based on the correlation, determine the valid signal among the multiple candidate signals corresponding to each signal component set.

[0008] In one embodiment of the present invention, the above-mentioned determination of noise signals in the first ICG signal and the first time-domain reference signal based on the frequency domain reference signal includes: Calculate the first correlation between the first ICG signal and the frequency domain reference signal, the second correlation between the first time domain reference signal and the frequency domain reference signal, and the third correlation between the first ICG signal and the first time domain reference signal; Based on the third relevance, the first relevance and the second relevance are adjusted respectively; Based on the adjusted first correlation, the noise signal in the first ICG signal is determined, and based on the adjusted second correlation, the noise signal in the first time-domain reference signal is determined.

[0009] Secondly, embodiments of the present invention provide an ICG signal denoising device, the device comprising: The signal acquisition module is used to acquire a first ICG signal to be denoised, and to acquire a first time-domain reference signal and a frequency-domain reference signal, wherein the first time-domain reference signal is a signal that is correlated with the time-domain variation trend of the first ICG signal, and the frequency-domain reference signal is a signal that is correlated with the frequency-domain variation trend of the first ICG signal and the first time-domain reference signal. The noise signal determination module is used to determine the noise signal in the first ICG signal and the first time-domain reference signal based on the frequency domain reference signal, remove the noise signal in the first ICG signal and use it as the second ICG signal, and remove the noise signal in the first time-frequency reference signal and use it as the second time-domain reference signal. The signal denoising module is used to determine the target valid signal in the second ICG signal based on the second time-domain reference signal, so as to achieve ICG signal denoising.

[0010] In one embodiment of the present invention, the signal denoising module includes: The first signal determination submodule is used to determine the initial valid signal in the second ICG signal based on the second time-domain reference signal and the second ICG signal; The second signal determination submodule is used to determine the target valid signal in the second ICG signal based on the initial valid signal and each signal component contained in the second ICG signal.

[0011] In one embodiment of the present invention, the second signal determination submodule includes: The signal clustering unit is used to cluster each signal component contained in the second ICG signal to obtain a clustered set of each signal component. A signal determination unit is used to determine the effective signal of each signal component set based on the signal components contained in each signal component set and the initial effective signal. The signal update unit is used to update the effective signal of each signal component set based on the effective signals of the adjacent signal component sets of each signal component set, and to determine the target effective signal in the second ICG signal based on the updated effective signal.

[0012] In one embodiment of the present invention, the signal clustering unit described above is specifically used to determine the critical signal component that characterizes the boundary between effective signal and noise signal in the signal component based on the frequency segmentation of each signal component contained in the second ICG signal, and to determine the effective signal set and the noise signal set based on the critical signal component; The signal update unit is specifically used to update the effective signal of the effective signal set based on the effective signal of the noise signal set, and determine the updated effective signal as the target effective signal in the second ICG signal.

[0013] In one embodiment of the present invention, the signal determination unit is specifically configured to perform iterative reconstruction for each signal component set according to the frequency band order of each signal component contained in the signal component set. In each iteration, the signal component targeted in the current iteration and the reconstructed signal obtained in the previous iteration are reconstructed, and the reconstructed signal obtained after each iteration is determined as a candidate signal. The correlation between the multiple candidate signals corresponding to each signal component set and the initial valid signal is calculated, and based on the correlation, the valid signal among the multiple candidate signals corresponding to each signal component set is determined.

[0014] In one embodiment of the present invention, the noise signal determination module includes: The correlation calculation submodule is used to calculate the first correlation between the first ICG signal and the frequency domain reference signal, the second correlation between the first time domain reference signal and the frequency domain reference signal, and the third correlation between the first ICG signal and the first time domain reference signal. The relevance adjustment submodule is used to adjust the first relevance and the second relevance respectively based on the third relevance. The noise signal determination submodule is used to determine the noise signal in the first ICG signal based on the adjusted first correlation, and to determine the noise signal in the first time-domain reference signal based on the adjusted second correlation.

[0015] Thirdly, embodiments of the present invention provide an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; When a processor executes a program stored in memory, it implements the steps of the method described in the first aspect above.

[0016] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method described in the first aspect above.

[0017] As can be seen from the above, the solution applied in this application employs a two-stage signal denoising process. The first stage uses a frequency domain reference signal to determine the noise signal in the first ICG signal and the first time domain reference signal. The second stage follows the first stage, using the denoised second time domain reference signal to determine the valid signal in the denoised second ICG signal, thereby achieving signal denoising. Furthermore, the reference signals used in the denoising process are the first time domain reference signal and the frequency domain reference signal. Since the first time domain reference signal is correlated with the time-domain variation trend of the first ICG signal, and the frequency domain reference signal is correlated with the frequency-domain variation trend of the first ICG signal and the first time domain reference signal, denoising of the first ICG signal can be comprehensively performed from both time-domain and frequency-domain correlation dimensions. Therefore, combining the above two aspects, the solution of this embodiment can accurately denoise the ICG signal.

[0018] Furthermore, in the above two-stage signal denoising process, the first stage is to identify the noise signal in the first ICG signal and perform identification on the noise signal, so that the second ICG signal obtained has eliminated most of the noise signal; the second stage is to identify the target valid signal in the second ICG signal and perform identification on the valid signal, without having to identify the noise signal again, reducing the limitations of noise signal identification, thereby further improving the denoising effect.

[0019] In addition, in the second stage of the two-stage signal denoising process, the second time-domain reference signal is also obtained by denoising the frequency-domain reference signal. In this way, the effectiveness of the second time-domain reference signal is relatively high. By using the second time-domain reference signal, the effective target signal can be determined more accurately, which significantly improves the signal denoising effect.

[0020] Of course, implementing any product or method of the present invention does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0022] Figure 1 A flowchart illustrating the first ICG signal denoising method provided in this application embodiment; Figure 2 A flowchart illustrating the second ICG signal denoising method provided in this application embodiment; Figure 3A flowchart illustrating the third ICG signal denoising method provided in this application embodiment; Figure 4 A flowchart illustrating the fourth ICG signal denoising method provided in this application embodiment; Figure 5 This is a schematic diagram of the structure of the first ICG signal denoising device provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of the second ICG signal denoising device provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of the structure of the third ICG signal denoising device provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of the structure of the fourth ICG signal denoising device provided in an embodiment of the present invention; Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0023] 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 embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of the present invention.

[0024] First, the application scenarios of the embodiments of this application will be described.

[0025] The application scenario of this application embodiment can be: a scenario of measuring hemodynamic parameters based on ICG signals, including stroke volume, cardiac output, cardiac ejection time, mean arterial pressure, etc.

[0026] Because ICG signals are susceptible to noise interference, they need to be accurately denoised to better measure hemodynamic parameters.

[0027] See Figure 1 , Figure 1 This is a flowchart illustrating the first ICG signal denoising method provided in the embodiments of this application. The method includes the following steps S101-S103.

[0028] Step S101: Obtain the first ICG signal to be denoised, and obtain the first time-domain reference signal and the frequency-domain reference signal.

[0029] The first ICG signal mentioned above refers to the ICG signal that needs to be denoised.

[0030] The aforementioned first time-domain reference signal is a signal that is correlated with the time-domain variation trend of the first ICG signal. This first time-domain reference signal can be a PPG (Photoplethysmography) signal, a heart sound signal, etc.

[0031] The aforementioned frequency domain reference signal is a signal whose frequency domain variation trend is correlated with that of the first ICG signal and the first time domain reference signal. Since the frequency domain can accurately reflect signal energy, the frequency domain reference signal is correlated with the first ICG signal and the first time domain reference signal from the perspective of signal energy. The frequency domain reference signal can be a heart rate signal, etc.

[0032] The aforementioned first time-domain reference signal and frequency-domain reference signal can be signals of the same cardiac cycle as the first ICG signal, signals of adjacent cardiac cycles to the first ICG signal, or signals of multiple cardiac cycles containing the cardiac cycle of the first ICG signal.

[0033] The aforementioned signals can be signals acquired in real time or signals stored in a data storage device. Based on this, When obtaining the aforementioned signals, the signals can be acquired in real time or read from a data storage device. This application does not limit the scope of the acquisition.

[0034] Step S102: Based on the frequency domain reference signal, determine the noise signals in the first ICG signal and the first time domain reference signal respectively, remove the noise signals in the first ICG signal to obtain the second ICG signal, and remove the noise signals in the first time-frequency reference signal to obtain the second time domain reference signal.

[0035] Since the noise signal in the first ICG signal and the first time-domain reference signal is determined based on the frequency-domain reference signal, and the frequency-domain reference signal is a signal that is correlated with the first ICG signal and the first time-domain reference signal from the perspective of signal energy, while the noise signal has a significant impact on the signal energy, the noise signal determined by the frequency-domain reference signal can be accurately determined from the perspective of signal energy.

[0036] When determining the noise signal, the following two implementation methods can be used: In the first embodiment, for each frequency band of the first ICG signal and the first time-domain reference signal, a first correlation degree between the signal component of that frequency band and the frequency-domain reference signal can be calculated, and based on the first correlation degree, it can be determined whether the signal component of each frequency band is a noise signal.

[0037] A signal is composed of signal components in different frequency bands. In this embodiment, the first ICG signal and the first time-domain reference signal can be divided into frequency bands to obtain the signal components of each frequency band.

[0038] The aforementioned first correlation degree characterizes the correlation between the signal component in each frequency band and the frequency domain reference signal. A higher correlation degree indicates a higher probability that the signal component in that frequency band is noise, while a lower correlation degree indicates a lower probability that the signal component in that frequency band is noise. The first correlation degree can be calculated using any existing correlation analysis technique, such as CCA or ICA analysis.

[0039] When determining whether a signal component in each frequency band is a noise signal, signal components with a correlation greater than a preset correlation threshold can be identified as noise signals. Alternatively, a preset number of signal components with the highest correlation can be identified as noise signals.

[0040] Step S103: Based on the second time-domain reference signal, determine the target effective signal in the second ICG signal to achieve ICG signal denoising.

[0041] The target valid signal can be understood as the truly valid signal in the second ICG signal, thereby indirectly eliminating other interference signals in the second ICG signal. In this way, due to the two-stage denoising method, the denoising effect of the ICG signal is good.

[0042] When determining the target effective signal, the second time-domain reference signal can be used as a reference standard to calculate the correlation between the second time-domain reference signal component and the second ICG signal component in each frequency band. Based on the correlation, the effective signal component in each frequency band is determined, and the effective signal component in each frequency band is reconstructed to obtain the target effective signal.

[0043] When calculating the above correlation, ICA analysis can be used to analyze the correlation between the two signal components.

[0044] When determining valid signal components, signal components with a correlation greater than a preset correlation threshold can be identified as valid signal components. Alternatively, a preset number of signal components with the highest correlation can be identified as valid signal components.

[0045] Other methods for determining the effective signal of a target can be found in the following sections. Figure 2 The corresponding implementation examples will not be described in detail here.

[0046] As can be seen from the above, the solution of this embodiment employs a two-stage signal denoising process. The first stage uses a frequency domain reference signal to determine the noise signal in the first ICG signal and the first time domain reference signal. The second stage follows the first stage, using the denoised second time domain reference signal to determine the valid signal in the denoised second ICG signal, thereby achieving signal denoising. Furthermore, the reference signals used in the denoising process are the first time domain reference signal and the frequency domain reference signal. Since the first time domain reference signal is correlated with the time-domain variation trend of the first ICG signal, and the frequency domain reference signal is correlated with the frequency-domain variation trend of the first ICG signal and the first time domain reference signal, denoising of the first ICG signal can be comprehensively performed from both time-domain and frequency-domain correlation dimensions. Therefore, combining the above two aspects, the solution of this embodiment can accurately denoise the ICG signal.

[0047] Furthermore, in the above two-stage signal denoising process, the first stage is to identify the noise signal in the first ICG signal and perform identification on the noise signal, so that the second ICG signal obtained has eliminated most of the noise signal; the second stage is to identify the target valid signal in the second ICG signal and perform identification on the valid signal, without having to identify the noise signal again, reducing the limitations of noise signal identification, thereby further improving the denoising effect.

[0048] In addition, in the second stage of the two-stage signal denoising process, the second time-domain reference signal is also obtained by denoising the frequency-domain reference signal. In this way, the effectiveness of the second time-domain reference signal is relatively high. By using the second time-domain reference signal, the effective target signal can be determined more accurately, which significantly improves the signal denoising effect.

[0049] In the foregoing Figure 1 In step S103 of the corresponding embodiment, in addition to determining the target valid signal using the mentioned method, it can also be implemented using the following steps S203-S204. Based on this, in one embodiment of this application, see... Figure 2 , Figure 2 This is a flowchart illustrating a second ICG signal denoising method provided in an embodiment of this application. The method includes the following steps S201-S204.

[0050] Step S201: Obtain the first ICG signal to be denoised, and obtain the first time-domain reference signal and the frequency-domain reference signal.

[0051] The aforementioned first time-domain reference signal is a signal that is correlated with the time-domain variation trend of the first ICG signal, and the frequency-domain reference signal is a signal that is correlated with the frequency-domain variation trend of the first ICG signal and the first time-domain reference signal.

[0052] Step S202: Based on the frequency domain reference signal, determine the noise signals in the first ICG signal and the first time domain reference signal respectively, remove the noise signals in the first ICG signal to obtain the second ICG signal, and remove the noise signals in the first time-frequency reference signal to obtain the second time domain reference signal.

[0053] The steps S201-S202 described above are the same as the steps S101-S102 described above, and will not be repeated here.

[0054] Step S203: Based on the second time-domain reference signal and the second ICG signal, determine the initial valid signal in the second ICG signal.

[0055] When determining the initial valid signal, the second time-domain reference signal can be used as a reference standard to calculate the correlation between the second time-domain reference signal component and the second ICG signal component of each frequency band. Based on the correlation, the valid signal component of each frequency band is determined, and the valid signal component of each frequency band is determined as the initial valid signal.

[0056] When calculating the above correlation, ICA analysis can be used to analyze the correlation between the two signal components.

[0057] When determining valid signal components, signal components with a correlation greater than a preset correlation threshold can be identified as valid signal components. Alternatively, a preset number of signal components with the highest correlation can be identified as valid signal components.

[0058] Step S204: Based on the initial valid signal and each signal component contained in the second ICG signal, determine the target valid signal in the second ICG signal to achieve ICG signal denoising.

[0059] Since the target valid signal is determined further based on the initial valid signal, the valid signal in the second ICG signal can be fully extracted.

[0060] In determining the target valid signal, in one embodiment, the correlation between the initial valid signal and each signal component contained in the second ICG signal can be calculated, the signal component with the highest correlation can be determined, the initial valid signal and the signal component can be reconstructed, and the reconstructed signal can be determined as the target valid signal.

[0061] Other implementation methods for determining the valid signal of the target can be found in the following sections. Figure 3 The corresponding implementation examples will not be described in detail here.

[0062] As can be seen from the above, in this embodiment, the initial valid signal in the second ICG signal is first determined using the second time-domain reference signal. Then, starting from within the second ICG signal, the target valid signal is determined using the initial valid signal and each signal component. In this way, the determined target valid signal considers both the second time-domain reference signal and the internal signal components of the second ICG signal, comprehensively determining it from different dimensions. This ensures that the determined target valid signal is highly accurate as a valid signal within the second ICG signal, thereby further improving the denoising effect.

[0063] In the foregoing Figure 2 In step S204 of the corresponding embodiment, in addition to determining the target valid signal using the mentioned method, it can also be implemented using the following steps S304-S306. Based on this, in one embodiment of this application, see... Figure 3 , Figure 3 This is a flowchart illustrating a third ICG signal denoising method provided in an embodiment of this application. The method includes the following steps S301-S306.

[0064] Step S301: Obtain the first ICG signal to be denoised, and obtain the first time-domain reference signal and the frequency-domain reference signal.

[0065] The aforementioned first time-domain reference signal is a signal that is correlated with the time-domain variation trend of the first ICG signal, and the frequency-domain reference signal is a signal that is correlated with the frequency-domain variation trend of the first ICG signal and the first time-domain reference signal.

[0066] Step S302: Based on the frequency domain reference signal, determine the noise signals in the first ICG signal and the first time domain reference signal respectively, remove the noise signals in the first ICG signal to obtain the second ICG signal, and remove the noise signals in the first time-frequency reference signal to obtain the second time domain reference signal.

[0067] Step S303: Based on the second time-domain reference signal and the second ICG signal, determine the initial valid signal in the second ICG signal.

[0068] The steps S301-S303 described above are the same as the steps S201-S203 described above, and will not be repeated here.

[0069] Step S304: Cluster each signal component contained in the second ICG signal to obtain a clustered set of each signal component.

[0070] In one implementation, each signal component can be clustered based on the frequency band to which it belongs. For example, the set of signal components from multiple frequency bands within a preset length frequency range can be determined as a signal component set in descending order of frequency band size.

[0071] In another implementation, the critical signal component that represents the boundary between the effective signal and the noise signal can be determined based on the frequency segmentation of each signal component contained in the second ICG signal. Based on the critical signal component, the effective signal set and the noise signal set are determined as the clustered signal component set.

[0072] The signal components contained in the above effective signal set are highly likely to be effective signals of the second ICG signal; the signal components contained in the above noise signal set are less likely to be effective signals of the second ICG signal.

[0073] When determining the effective signal set and the noise signal set, the second ICG signal can be decomposed using CEEMD to obtain multiple IMF components. The FastICA algorithm can then be used to perform correlation analysis on each IMF component to obtain multiple effective signals and multiple noise signals. The set of multiple effective signals is determined as the effective signal set, and the set of multiple noise signals is determined as the noise signal set.

[0074] Step S305: Determine the effective signal of each signal component set based on the signal components contained in each signal component set and the initial effective signal.

[0075] In determining the valid signal for each signal component set, in the first embodiment, the correlation between each signal component in the signal component set and the initial valid signal can be calculated. Based on the correlation, the valid signal is determined from each signal component in the signal component set. For example, signal components with a correlation greater than a preset correlation threshold can be determined as valid signals.

[0076] In the second implementation, for each signal component set, iterative reconstruction can be performed according to the frequency band order of each signal component contained in the signal component set. In each iteration, the signal component targeted in the current iteration and the reconstructed signal obtained in the previous iteration are reconstructed, and the reconstructed signal obtained after each iteration is determined as the candidate signal. The correlation between the multiple candidate signals corresponding to each signal component set and the initial valid signal is calculated, and based on the correlation, the valid signal among the multiple candidate signals corresponding to each signal component set is determined.

[0077] The frequency band order mentioned above can be either from largest to smallest or from smallest to largest. To illustrate iterative reconstruction, consider this example: the signal component set contains signal components S1, S2, S3, S4, and S5. Arrange these signal components in descending order of frequency band size as S5, S4, S3, S2, and S1. Following this order, the first iteration reconstructs S5 and S4 to obtain the reconstructed signal Rb1. The second iteration reconstructs S3 and Rb1 to obtain the reconstructed signal Rb2. In this manner, the reconstructed signals (Rb1, Rb2, Rb3, Rb4) obtained from each iteration are identified as multiple candidate signals corresponding to the signal component set.

[0078] As can be seen, the reconstructed signal obtained in each iteration reflects the local fine-grained information of the signal component set, thus laying the foundation for the extraction of effective signals.

[0079] For each signal component set, a correlation analysis algorithm is used to calculate the correlation between each candidate signal contained in the signal component set and the initial valid signal. Candidate signals with a correlation greater than a preset correlation threshold can be identified as valid signals of the signal component set.

[0080] Since iterative reconstruction is performed on each signal component contained in the signal component set, a reconstructed signal reflecting multiple local fine-grained information of the signal component set is obtained. By considering multiple local signal information of the signal component set, a more accurate and effective signal can be obtained.

[0081] Step S306: Based on the effective signals of the adjacent signal component sets of each signal component set, update the effective signal of each signal component set. Based on the updated effective signal, determine the target effective signal in the second ICG signal to achieve ICG signal denoising.

[0082] The effective signal of each signal component set can be updated sequentially according to the frequency range corresponding to each signal component set from smallest to largest. The effective signal of the last updated signal component set is then determined as the target effective signal in the second ICG signal. In this case, adjacent signal component sets can be the signal component sets preceding the current signal component set.

[0083] In one implementation, when updating the effective signals of a signal component set, the correlation between the effective signals of adjacent signal component sets and the effective signals of the current signal component set can be calculated. Based on this correlation, the updated effective signals of the current signal component set can be determined; for example, signals with a correlation greater than a preset correlation threshold can be used as the updated effective signals. In another implementation, the effective signals of adjacent signal component sets and the effective signals of the current signal component set can be fused, and the fused signal can be determined as the updated effective signal.

[0084] In the second embodiment, the determination of the effective signal set and the noise signal set is mentioned in the aforementioned step S304. In this case, when determining the target effective signal, the effective signal of the effective signal set can be updated based on the effective signal of the noise signal set, and the updated effective signal can be determined as the target effective signal in the second ICG signal.

[0085] Since the updated effective signal is determined based on the effective signal of the noise signal set, the updated effective signal takes into account both the signals within the effective signal set and the effective signals of the noise signal set, making the updated effective signal more comprehensive. This results in high accuracy of the determined target effective signal being the effective signal of the second ICG signal.

[0086] In one implementation, when updating the valid signal of the valid signal set, the valid signal of the noise signal set and the valid signal of the valid signal set can be reconstructed, and the reconstructed signal can be determined as the updated valid signal.

[0087] As can be seen from the above, by clustering each signal component in the second ICG to obtain each signal component set, the effective signal of each signal component set is determined using the initial effective signal. Then, based on the effective signals of the adjacent signal component sets, the effective signal of the current signal component set is updated. In this way, the updated effective signal not only considers the information of the initial effective signal, but also the signal information of the adjacent signal component sets, thereby further improving the probability that the obtained target effective signal is the effective signal of the second ICG signal, thus optimizing the denoising effect.

[0088] In the foregoing Figure 1 / Figure 2 / Figure 3 In the corresponding embodiments, the steps all involve determining the noise signal in the first ICG signal and the first time-domain reference signal based on the frequency domain reference signal. Besides the methods mentioned above, it can also be implemented according to the following steps S402-S404. Based on this, in one embodiment of this application, see... Figure 4 , Figure 4This is a flowchart illustrating the fourth ICG signal denoising method provided in the embodiments of this application. The method includes the following steps S401-S405.

[0089] It should be noted that, Figure 4 Therefore Figure 1 Based on. This application also includes [other applications]. Figure 2 / Figure 3 This is an embodiment of determining the noise signal based on the basic embodiment, and will not be described again to avoid repetition.

[0090] Step S401: Obtain the first ICG signal to be denoised, and obtain the first time-domain reference signal and the frequency-domain reference signal.

[0091] The aforementioned first time-domain reference signal is a signal that is correlated with the time-domain variation trend of the first ICG signal, and the frequency-domain reference signal is a signal that is correlated with the frequency-domain variation trend of the first ICG signal and the first time-domain reference signal.

[0092] Step S402: Calculate the first correlation between the first ICG signal and the frequency domain reference signal, the second correlation between the first time domain reference signal and the frequency domain reference signal, and the third correlation between the first ICG signal and the first time domain reference signal.

[0093] Each correlation score represents the correlation between the two signals used. When calculating each correlation score, the two signals can be segmented to obtain segmented signal components, and correlation analysis can be performed on the two types of signal components. Specifically, any existing correlation analysis technique can be used for calculation, such as CCA or ICA analysis techniques; this application is not limited to this.

[0094] Step S403: Based on the third relevance, adjust the first relevance and the second relevance respectively.

[0095] When adjusting the first and second relevance, the third relevance can be normalized. The difference between 1 and the normalized result is calculated, and the difference is used as a coefficient. The product between the first relevance and the coefficient is calculated to obtain the adjusted first relevance. The product between the second relevance and the coefficient is then calculated to obtain the adjusted second relevance.

[0096] Step S404: Based on the adjusted first correlation, determine the noise signal in the first ICG signal, and based on the adjusted second correlation, determine the noise signal in the first time-domain reference signal.

[0097] Signals with a first correlation degree less than a preset correlation degree threshold can be used as noise signals in the first ICG signal, and signals with a second correlation degree greater than a preset correlation degree threshold can be used as noise signals in the first time-domain reference signal.

[0098] Step S405: Based on the second time-domain reference signal, determine the target effective signal in the second ICG signal to achieve ICG signal denoising.

[0099] Steps S401 and S405 are the same as steps S101 and S103 mentioned above. They will not be repeated here.

[0100] As can be seen, by utilizing the third correlation between the first ICG signal and the first time-domain reference signal, the first and second correlations are further adjusted. Since the first correlation reflects the signal energy of the first ICG signal from the frequency domain perspective, the second correlation reflects the signal energy of the first time-domain reference signal from the frequency domain perspective, and the third correlation reflects the signal characteristics of the first time-domain reference signal and the first ICG signal from the time domain perspective, adjusting the first and second correlations based on the third correlation considers not only the signal frequency domain situation but also the signal time domain situation. Combining these two dimensions comprehensively determines the noise signal, thereby improving the accuracy of the noise signal.

[0101] Corresponding to the above-described ICG signal denoising method, this embodiment of the invention also provides an ICG signal denoising device.

[0102] See Figure 5 , Figure 5 This is a schematic diagram of the structure of the first ICG signal denoising device provided in an embodiment of the present invention. The device includes 501-503.

[0103] The signal acquisition module 501 is used to acquire the first ICG signal to be denoised, and to acquire the first time-domain reference signal and the frequency-domain reference signal. The first time-domain reference signal is a signal that is related to the time-domain variation trend of the first ICG signal, and the frequency-domain reference signal is a signal that is related to the frequency-domain variation trend of the first ICG signal and the first time-domain reference signal. The noise signal determination module 502 is used to determine the noise signal in the first ICG signal and the first time-domain reference signal based on the frequency domain reference signal, remove the noise signal in the first ICG signal and use it as the second ICG signal, and remove the noise signal in the first time-frequency reference signal and use it as the second time-domain reference signal. The signal denoising module 503 is used to determine the target valid signal in the second ICG signal based on the second time-domain reference signal, so as to achieve ICG signal denoising.

[0104] As can be seen from the above, the solution of this embodiment employs a two-stage signal denoising process. The first stage uses a frequency domain reference signal to determine the noise signal in the first ICG signal and the first time domain reference signal. The second stage follows the first stage, using the denoised second time domain reference signal to determine the valid signal in the denoised second ICG signal, thereby achieving signal denoising. Furthermore, the reference signals used in the denoising process are the first time domain reference signal and the frequency domain reference signal. Since the first time domain reference signal is correlated with the time-domain variation trend of the first ICG signal, and the frequency domain reference signal is correlated with the frequency-domain variation trend of the first ICG signal and the first time domain reference signal, denoising of the first ICG signal can be comprehensively performed from both time-domain and frequency-domain correlation dimensions. Therefore, combining the above two aspects, the solution of this embodiment can accurately denoise the ICG signal.

[0105] Furthermore, in the above two-stage signal denoising process, the first stage is to identify the noise signal in the first ICG signal and perform identification on the noise signal, so that the second ICG signal obtained has eliminated most of the noise signal; the second stage is to identify the target valid signal in the second ICG signal and perform identification on the valid signal, without having to identify the noise signal again, reducing the limitations of noise signal identification, thereby further improving the denoising effect.

[0106] In addition, in the second stage of the two-stage signal denoising process, the second time-domain reference signal is also obtained by denoising the frequency-domain reference signal. In this way, the effectiveness of the second time-domain reference signal is relatively high. By using the second time-domain reference signal, the effective target signal can be determined more accurately, which significantly improves the signal denoising effect.

[0107] See Figure 6 , Figure 6 This is a schematic diagram of the structure of a second ICG signal denoising device provided in an embodiment of the present invention. The device includes 601-604.

[0108] The signal acquisition module 601 is used to acquire a first ICG signal to be denoised, and to acquire a first time-domain reference signal and a frequency-domain reference signal, wherein the first time-domain reference signal is a signal that is related to the time-domain variation trend of the first ICG signal, and the frequency-domain reference signal is a signal that is related to the frequency-domain variation trend of the first ICG signal and the first time-domain reference signal. The noise signal determination module 602 is used to determine the noise signal in the first ICG signal and the first time-domain reference signal based on the frequency domain reference signal, remove the noise signal in the first ICG signal and use it as the second ICG signal, and remove the noise signal in the first time-frequency reference signal and use it as the second time-domain reference signal. The above 601-602 are the same as the aforementioned 501-502.

[0109] The first signal determination submodule 603 is used to determine the initial valid signal in the second ICG signal based on the second time-domain reference signal and the second ICG signal; The second signal determination submodule 604 is used to determine the target valid signal in the second ICG signal based on the initial valid signal and each signal component contained in the second ICG signal.

[0110] As can be seen from the above, in this embodiment, the initial valid signal in the second ICG signal is first determined using the second time-domain reference signal. Then, starting from within the second ICG signal, the target valid signal is determined using the initial valid signal and each signal component. In this way, the determined target valid signal considers both the second time-domain reference signal and the internal signal components of the second ICG signal, comprehensively determining it from different dimensions. This ensures that the determined target valid signal is highly accurate as a valid signal within the second ICG signal, thereby further improving the denoising effect.

[0111] See Figure 7 , Figure 7 This is a schematic diagram of the structure of a third ICG signal denoising device provided in an embodiment of the present invention. The device includes 701-706.

[0112] The signal acquisition module 701 is used to acquire a first ICG signal to be denoised, and to acquire a first time-domain reference signal and a frequency-domain reference signal. The first time-domain reference signal is a signal that is related to the time-domain variation trend of the first ICG signal, and the frequency-domain reference signal is a signal that is related to the frequency-domain variation trend of the first ICG signal and the first time-domain reference signal. The noise signal determination module 702 is used to determine the noise signal in the first ICG signal and the first time-domain reference signal based on the frequency domain reference signal, remove the noise signal in the first ICG signal and use it as the second ICG signal, and remove the noise signal in the first time-frequency reference signal and use it as the second time-domain reference signal. The first signal determination submodule 703 is used to determine the initial valid signal in the second ICG signal based on the second time-domain reference signal and the second ICG signal; The above 701-703 are the same as the aforementioned 601-603.

[0113] The signal clustering unit 704 is used to cluster each signal component contained in the second ICG signal to obtain a clustered set of each signal component. The signal determination unit 705 is used to determine the effective signal of each signal component set based on the signal components contained in each signal component set and the initial effective signal. The signal update unit 706 is used to update the effective signal of each signal component set based on the effective signals of the adjacent signal component sets of each signal component set, and to determine the target effective signal in the second ICG signal based on the updated effective signal.

[0114] As can be seen from the above, by clustering each signal component in the second ICG to obtain each signal component set, the effective signal of each signal component set is determined using the initial effective signal. Then, based on the effective signals of the adjacent signal component sets, the effective signal of the current signal component set is updated. In this way, the updated effective signal not only considers the information of the initial effective signal, but also the signal information of the adjacent signal component sets, thereby further improving the probability that the obtained target effective signal is the effective signal of the second ICG signal, thus optimizing the denoising effect.

[0115] In one embodiment of the present invention, the signal clustering unit 704 described above is specifically used to determine the critical signal component that characterizes the boundary between effective signal and noise signal in the signal component based on the frequency segmentation of each signal component contained in the second ICG signal, and to determine the effective signal set and the noise signal set based on the critical signal component; The signal update unit 706 is specifically used to update the effective signal of the effective signal set based on the effective signal of the noise signal set, and determine the updated effective signal as the target effective signal in the second ICG signal.

[0116] Since the updated effective signal is determined based on the effective signal of the noise signal set, the updated effective signal takes into account both the signals within the effective signal set and the effective signals of the noise signal set, making the updated effective signal more comprehensive. This results in high accuracy of the determined target effective signal being the effective signal of the second ICG signal.

[0117] In one embodiment of the present invention, the signal determination unit 705 is specifically configured to perform iterative reconstruction for each signal component set according to the frequency band order of each signal component contained in the signal component set. In each iteration, the signal component targeted in the current iteration and the reconstructed signal obtained in the previous iteration are reconstructed, and the reconstructed signal obtained after each iteration is determined as a candidate signal. The correlation between the multiple candidate signals corresponding to each signal component set and the initial valid signal is calculated, and based on the correlation, the valid signal among the multiple candidate signals corresponding to each signal component set is determined.

[0118] Since iterative reconstruction is performed on each signal component contained in the signal component set, a reconstructed signal reflecting multiple local fine-grained information of the signal component set is obtained. By considering multiple local signal information of the signal component set, a more accurate and effective signal can be obtained.

[0119] See Figure 8 , Figure 8 This is a schematic diagram of the structure of the fourth ICG signal denoising device provided in the embodiment of the present invention. The device includes 801-805.

[0120] The signal acquisition module 801 is used to acquire a first ICG signal to be denoised, and to acquire a first time-domain reference signal and a frequency-domain reference signal. The first time-domain reference signal is a signal that is related to the time-domain variation trend of the first ICG signal, and the frequency-domain reference signal is a signal that is related to the frequency-domain variation trend of the first ICG signal and the first time-domain reference signal. The correlation calculation submodule 802 is used to calculate the first correlation between the first ICG signal and the frequency domain reference signal, the second correlation between the first time domain reference signal and the frequency domain reference signal, and the third correlation between the first ICG signal and the first time domain reference signal. The relevance adjustment submodule 803 is used to adjust the first relevance and the second relevance respectively based on the third relevance; The noise signal determination submodule 804 is used to determine the noise signal in the first ICG signal based on the adjusted first correlation, and to determine the noise signal in the first time-domain reference signal based on the adjusted second correlation.

[0121] The signal denoising module 805 is used to determine the target valid signal in the second ICG signal based on the second time-domain reference signal, so as to achieve ICG signal denoising.

[0122] The above 801 and 805 are the same as the aforementioned 501 and 503.

[0123] As can be seen, by utilizing the third correlation between the first ICG signal and the first time-domain reference signal, the first and second correlations are further adjusted. Since the first correlation reflects the signal energy of the first ICG signal from the frequency domain perspective, the second correlation reflects the signal energy of the first time-domain reference signal from the frequency domain perspective, and the third correlation reflects the signal characteristics of the first time-domain reference signal and the first ICG signal from the time domain perspective, adjusting the first and second correlations based on the third correlation considers not only the signal frequency domain situation but also the signal time domain situation. Combining these two dimensions comprehensively determines the noise signal, thereby improving the accuracy of the noise signal.

[0124] Corresponding to the above-described ICG signal denoising method, this embodiment of the invention also provides an electronic device.

[0125] See Figure 9 , Figure 9 A schematic diagram of an electronic device provided in an embodiment of the present invention includes a processor 901, a communication interface 902, a memory 903, and a communication bus 904, wherein the processor 901, the communication interface 902, and the memory 903 communicate with each other through the communication bus 904. Memory 903 is used to store computer programs; The processor 901 is used to execute the program stored in the memory 903 to implement the ICG signal denoising method provided in the embodiments of the present invention.

[0126] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0127] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0128] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0129] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0130] In another embodiment of the present invention, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the ICG signal denoising method provided in the embodiment of the present invention.

[0131] In another embodiment of the present invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to implement the ICG signal denoising method provided in the embodiment of the present invention.

[0132] As can be seen from the above, the solution of this embodiment employs a two-stage signal denoising process. The first stage uses a frequency domain reference signal to determine the noise signal in the first ICG signal and the first time domain reference signal. The second stage follows the first stage, using the denoised second time domain reference signal to determine the valid signal in the denoised second ICG signal, thereby achieving signal denoising. Furthermore, the reference signals used in the denoising process are the first time domain reference signal and the frequency domain reference signal. Since the first time domain reference signal is correlated with the time-domain variation trend of the first ICG signal, and the frequency domain reference signal is correlated with the frequency-domain variation trend of the first ICG signal and the first time domain reference signal, denoising of the first ICG signal can be comprehensively performed from both time-domain and frequency-domain correlation dimensions. Therefore, combining the above two aspects, the solution of this embodiment can accurately denoise the ICG signal.

[0133] Furthermore, in the above two-stage signal denoising process, the first stage is to identify the noise signal in the first ICG signal and perform identification on the noise signal, so that the second ICG signal obtained has eliminated most of the noise signal; the second stage is to identify the target valid signal in the second ICG signal and perform identification on the valid signal, without having to identify the noise signal again, reducing the limitations of noise signal identification, thereby further improving the denoising effect.

[0134] In addition, in the second stage of the two-stage signal denoising process, the second time-domain reference signal is also obtained by denoising the frequency-domain reference signal. In this way, the effectiveness of the second time-domain reference signal is relatively high. By using the second time-domain reference signal, the effective target signal can be determined more accurately, which significantly improves the signal denoising effect.

[0135] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0136] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0137] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments for apparatus, electronic devices, and computer-readable storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0138] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A method for denoising ICG signals, characterized in that, The method includes: A first ICG signal to be denoised is obtained, and a first time-domain reference signal and a frequency-domain reference signal are obtained. The first time-domain reference signal is a signal that is related to the time-domain variation trend of the first ICG signal, and the frequency-domain reference signal is a signal that is related to the frequency-domain variation trend of the first ICG signal and the first time-domain reference signal. Based on the frequency domain reference signal, noise signals in the first ICG signal and the first time domain reference signal are determined respectively. The noise signals in the first ICG signal are removed to obtain the second ICG signal, and the noise signals in the first time-frequency reference signal are removed to obtain the second time domain reference signal. Based on the second time-domain reference signal, the target valid signal in the second ICG signal is determined in order to achieve ICG signal denoising; The step of determining the target valid signal in the second ICG signal based on the second time-domain reference signal includes: Based on the second time-domain reference signal and the second ICG signal, the initial valid signal in the second ICG signal is determined; Based on the initial valid signal and each signal component contained in the second ICG signal, the target valid signal in the second ICG signal is determined; The step of determining the noise signal in the first ICG signal and the first time-domain reference signal based on the frequency domain reference signal includes: Calculate the first correlation between the first ICG signal and the frequency domain reference signal, the second correlation between the first time domain reference signal and the frequency domain reference signal, and the third correlation between the first ICG signal and the first time domain reference signal; Based on the third relevance, the first relevance and the second relevance are adjusted respectively; Based on the adjusted first correlation, the noise signal in the first ICG signal is determined, and based on the adjusted second correlation, the noise signal in the first time-domain reference signal is determined.

2. The method according to claim 1, characterized in that, The step of determining the target valid signal in the second ICG signal based on each signal component contained in the initial valid signal and the second ICG signal includes: Cluster each signal component contained in the second ICG signal to obtain a clustered set of each signal component; Based on the signal components contained in each signal component set and the initial valid signal, determine the valid signal of each signal component set; Based on the effective signals of the adjacent signal component sets of each signal component set, the effective signals of each signal component set are updated, and based on the updated effective signals, the target effective signal in the second ICG signal is determined.

3. The method according to claim 2, characterized in that, The step of clustering each signal component contained in the second ICG signal to obtain a clustered set of each signal component includes: Based on the frequency segmentation of each signal component contained in the second ICG signal, the critical signal component that characterizes the boundary between the effective signal and the noise signal is determined in the signal component, and the effective signal set and the noise signal set are determined based on the critical signal component; The process of updating the effective signals of each signal component set based on the effective signals of its adjacent signal component sets, and determining the target effective signal in the second ICG signal based on the updated effective signals, includes: Based on the effective signal of the noise signal set, update the effective signal of the effective signal set, and determine the updated effective signal as the target effective signal in the second ICG signal.

4. The method according to claim 2 or 3, characterized in that, Determining the effective signal of each signal component set based on the signal components contained in each signal component set and the initial effective signal includes: For each signal component set, iterative reconstruction is performed according to the frequency band order of each signal component contained in the signal component set. In each iteration, the signal component targeted by the current iteration and the reconstructed signal obtained in the previous iteration are reconstructed, and the reconstructed signal obtained after each iteration is determined as the candidate signal. Calculate the correlation between the multiple candidate signals corresponding to each signal component set and the initial valid signal. Based on the correlation, determine the valid signal among the multiple candidate signals corresponding to each signal component set.

5. An ICG signal denoising device, characterized in that, The device includes: The signal acquisition module is used to acquire a first ICG signal to be denoised, and to acquire a first time-domain reference signal and a frequency-domain reference signal, wherein the first time-domain reference signal is a signal that is correlated with the time-domain variation trend of the first ICG signal, and the frequency-domain reference signal is a signal that is correlated with the frequency-domain variation trend of the first ICG signal and the first time-domain reference signal. The noise signal determination module is used to determine the noise signal in the first ICG signal and the first time-domain reference signal based on the frequency domain reference signal, remove the noise signal in the first ICG signal and use it as the second ICG signal, and remove the noise signal in the first time-frequency reference signal and use it as the second time-domain reference signal. The signal denoising module is used to determine the target valid signal in the second ICG signal based on the second time-domain reference signal, so as to achieve ICG signal denoising; The signal denoising module includes: The first signal determination submodule is used to determine the initial valid signal in the second ICG signal based on the second time-domain reference signal and the second ICG signal; The second signal determination submodule is used to determine the target valid signal in the second ICG signal based on the initial valid signal and each signal component contained in the second ICG signal; The noise signal determination module includes: The correlation calculation submodule is used to calculate the first correlation between the first ICG signal and the frequency domain reference signal, the second correlation between the first time domain reference signal and the frequency domain reference signal, and the third correlation between the first ICG signal and the first time domain reference signal. The relevance adjustment submodule is used to adjust the first relevance and the second relevance respectively based on the third relevance. The noise signal determination submodule is used to determine the noise signal in the first ICG signal based on the adjusted first correlation, and to determine the noise signal in the first time-domain reference signal based on the adjusted second correlation.

6. The apparatus according to claim 5, characterized in that, The second signal determination submodule includes: The signal clustering unit is used to cluster each signal component contained in the second ICG signal to obtain a clustered set of each signal component. A signal determination unit is used to determine the effective signal of each signal component set based on the signal components contained in each signal component set and the initial effective signal. The signal update unit is used to update the effective signal of each signal component set based on the effective signals of the adjacent signal component sets of each signal component set, and to determine the target effective signal in the second ICG signal based on the updated effective signal.

7. The apparatus according to claim 6, characterized in that, The signal clustering unit is specifically used to determine the critical signal component that represents the boundary between effective signal and noise signal in the signal component based on the frequency segmentation of each signal component contained in the second ICG signal, and to determine the effective signal set and the noise signal set based on the critical signal component. The signal update unit is specifically used to update the effective signal of the effective signal set based on the effective signal of the noise signal set, and determine the updated effective signal as the target effective signal in the second ICG signal.

8. The apparatus according to claim 6 or 7, characterized in that, The signal determination unit is specifically used to iteratively reconstruct each signal component set according to the frequency band order of each signal component contained in the signal component set. In each iteration, the signal component targeted in the current iteration and the reconstructed signal obtained in the previous iteration are reconstructed, and the reconstructed signal obtained after each iteration is determined as the candidate signal. The correlation between the multiple candidate signals corresponding to each signal component set and the initial valid signal is calculated, and based on the correlation, the valid signal among the multiple candidate signals corresponding to each signal component set is determined.

Images