Signal bandwidth estimation method and device based on wavelet reconstruction

Abstract

The present disclosure specifically relates to the field of mobile communication technology, and specifically to a signal bandwidth estimation method and device based on wavelet reconstruction, a computer-readable medium, and a terminal device. The method includes: performing segmented processing on a signal to be processed to obtain multiple segmented signals, preprocessing the multiple segmented signals to obtain amplitude spectrum data after noise reduction; performing wavelet low-frequency reconstruction processing on the amplitude spectrum data to extract the corresponding envelope signal; determining the boundary of the envelope signal to estimate the bandwidth of the signal to be processed based on the boundary of the envelope signal. The method disclosed in the present disclosure can be applied to a variety of different signals, which not only reduces the complexity of the algorithm, but also weakens the influence of signal randomness on the estimation accuracy; it has strong robustness.

Description

Technical Field

[0001] The present disclosure relates to the field of mobile communication technology, and in particular to a signal bandwidth estimation method based on wavelet reconstruction, a signal bandwidth estimation device based on wavelet reconstruction, a computer-readable medium, and a terminal device. Background Art

[0002] Signal bandwidth refers to the width of the signal spectrum from low frequency to high frequency. In signal processing, bandwidth estimation is a classic parameter estimation problem and an important part of the signal processing discipline. Accurately estimating the signal bandwidth can provide great help for subsequent signal processing. In non-cooperative communications, bandwidth estimation can be applied to signal sorting. According to the estimated bandwidth of the signal, a suitable filter is designed to separate multiple signals in a single channel. In the fields related to signal monitoring and modulation recognition, bandwidth estimation also has a wide range of application backgrounds and important application value.

[0003] In the related art, bandwidth estimation methods include: autocorrelation method, root mean square method, energy concentration method, maximum entropy method and Welch method, etc. Among them, the autocorrelation method is a -3dB bandwidth defined based on the half-power point range of the power spectrum. This method is based on linear spectrum estimation and has a large error; the bandwidth estimated by the root mean square method is the bandwidth in the sense of mean square. There is no definite relationship between the bandwidth obtained for signals of different spectral types and the theoretical -3dB bandwidth, so this method is not universal; and the premise of the energy concentration method to estimate the bandwidth is that the center frequency needs to be estimated first, which will cause error superposition and is suitable for Gaussian or cubic power spectra; the maximum entropy method is suitable for scenes with high signal-to-noise ratio, and the effect is poor under low signal-to-noise ratio. The above methods are simple in principle and easy to implement, but the estimation accuracy is poor, and the estimation errors for signals of different spectral types are different. Moreover, the above traditional methods are all based on the study of -3dB bandwidth, which may cause partial information loss during signal sorting. In addition, the existing technology also includes methods based on power spectrum curve fitting, power spectrum wavelet transform and power spectrum distribution function geometric analysis to estimate bandwidth and achieve high accuracy; however, such algorithms are highly complex and difficult to implement.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The present invention provides a signal bandwidth estimation method based on wavelet reconstruction, a signal bandwidth estimation device based on wavelet reconstruction, a computer-readable medium, and a terminal device, which can effectively overcome the defects of low estimation accuracy and poor adaptability in the prior art.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0007] According to a first aspect of the present disclosure, a signal bandwidth estimation method based on wavelet reconstruction is provided, the method comprising:

[0008] Performing segment processing on the signal to be processed to obtain a plurality of segment signals, and preprocessing the plurality of segment signals to obtain amplitude spectrum data after noise reduction;

[0009] Performing wavelet low-frequency reconstruction processing on the amplitude spectrum data to extract the corresponding envelope signal;

[0010] The boundary of the envelope signal is determined to estimate the bandwidth of the signal to be processed according to the boundary of the envelope signal.

[0011] In some exemplary embodiments, the preprocessing of the plurality of segmented signals to obtain noise-reduced amplitude spectrum data includes:

[0012] Performing cross-correlation operation on multiple segmented signals according to the signal segmentation order to obtain a noise reduction signal;

[0013] Performing discrete Fourier transform on the noise reduction signal to obtain corresponding spectrum data;

[0014] The frequency spectrum data is converted into corresponding amplitude spectrum data.

[0015] In some exemplary embodiments, performing wavelet low-frequency reconstruction processing on the amplitude spectrum data to extract the corresponding envelope signal includes:

[0016] Based on a wavelet basis of a predetermined order and a wavelet decomposition layer number, the amplitude spectrum data is recursively decomposed using a low-frequency filter to obtain target low-frequency data;

[0017] Recursive reconstruction is performed based on the target low-frequency data according to the determined number of wavelet decomposition layers to obtain the envelope signal.

[0018] In some exemplary embodiments, recursively decomposing the amplitude spectrum data using a low-frequency filter to obtain target low-frequency data includes:

[0019] Determine the filter coefficients according to a wavelet basis of a predetermined order;

[0020] At the n+1th layer, a convolution operation is performed on the decomposition result and the filter coefficient of the nth layer, the convolution operation result is downsampled, and the downsampled result is used as the decomposition result of the n+1th layer; and, the process is repeated to a preset number of wavelet decomposition layers to obtain the target low-frequency coefficient; wherein n is a positive integer, and the amplitude spectrum data is configured as an input parameter of the first layer.

[0021] In some exemplary embodiments, recursively reconstructing the target low-frequency data according to the determined number of wavelet decomposition layers to obtain the envelope signal includes:

[0022] The high-frequency information of the nth layer is set to zero, the low-frequency data of the nth layer is up-sampled, a convolution operation is performed based on the up-sampling result and the filter coefficient, and the convolution operation result is configured as the low-frequency information of the n-1th layer; this step is repeated to the top layer to complete the recursive reconstruction and obtain the envelope signal; wherein n is a positive integer; the target low-frequency data is configured as the low-frequency data of the last layer.

[0023] In some exemplary embodiments, determining the boundary of the envelope signal to estimate the bandwidth of the signal to be processed according to the boundary of the envelope signal includes:

[0024] Normalizing the envelope signal, and performing a first forward difference operation on the normalized result to obtain a first difference result;

[0025] Performing segment processing on the first differential result, and performing a second forward differential operation on the segment processing result to obtain a second differential result;

[0026] Using a preset threshold to filter the second difference result for a maximum value to obtain a maximum value set;

[0027] Determine a left boundary and a right boundary according to the maximum value set;

[0028] Determine a first minimum value index based on the left boundary and configure it as an upper frequency band of the spectrum, and determine a second minimum value index based on the right boundary and configure it as a lower frequency band of the spectrum;

[0029] The bandwidth of the signal to be processed is estimated according to the upper frequency band of the frequency spectrum and the lower frequency band of the frequency spectrum.

[0030] In some exemplary embodiments, the bandwidth of the signal to be processed is the zero-crossing full bandwidth of the signal to be processed.

[0031] According to a second aspect of the present disclosure, a signal bandwidth estimation device based on wavelet reconstruction of a terminal device is provided, comprising:

[0032] A signal preprocessing module is used to perform segment processing on the signal to be processed to obtain multiple segment signals, and preprocess the multiple segment signals to obtain amplitude spectrum data after noise reduction;

[0033] A signal reconstruction module, used for performing wavelet low-frequency reconstruction processing on the amplitude spectrum data to extract the corresponding envelope signal;

[0034] The bandwidth estimation module is used to determine the boundary of the envelope signal to estimate the bandwidth of the signal to be processed according to the boundary of the envelope signal.

[0035] According to a third aspect of the present disclosure, a computer-readable medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the signal bandwidth estimation method based on wavelet reconstruction is implemented.

[0036] According to a fourth aspect of the present disclosure, a terminal device is provided, including:

[0037] Processor; and

[0038] A memory, configured to store executable instructions of the processor;

[0039] Wherein, the processor is configured to implement the above-mentioned signal bandwidth estimation method based on wavelet reconstruction by executing the executable instructions.

[0040] A signal bandwidth estimation method based on wavelet reconstruction provided by an embodiment of the present disclosure realizes signal noise reduction and reduces the influence of noise by segmenting and preprocessing the processed signal; and extracts the corresponding envelope signal by performing wavelet low-frequency reconstruction processing on the amplitude spectrum data, and finds the boundary of the signal spectrum for the envelope signal, and finally realizes the estimation of the signal bandwidth. The method disclosed in the present disclosure can be applied to a variety of different signals, which not only reduces the complexity of the algorithm, but also weakens the influence of signal randomness on the estimation accuracy; it has strong robustness.

[0041] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0043] Figure 1A schematic diagram schematically illustrates a signal bandwidth estimation method based on wavelet reconstruction in an exemplary embodiment of the present disclosure;

[0044] Figure 2 A schematic diagram schematically illustrates a principle diagram of a signal wavelet decomposition method according to an exemplary embodiment of the present disclosure;

[0045] Figure 3 Schematically shows an amplitude diagram of a signal of an exemplary embodiment of the present disclosure;

[0046] Figure 4 A schematic diagram schematically shows a data amplitude spectrum after segmented correlation in an exemplary embodiment of the present disclosure;

[0047] Figure 5 Schematically showing the results of reconstruction of different wavelet bases in an exemplary embodiment of the present disclosure;

[0048] Figure 6 A schematic diagram showing relative errors of bandwidth estimation under different orders of dbN wavelet basis in an exemplary embodiment of the present disclosure is shown schematically;

[0049] Figure 7 A schematic diagram schematically showing the relative error of the db4 wavelet basis at different decomposition levels in an exemplary embodiment of the present disclosure;

[0050] Figure 8 A schematic diagram schematically showing the decomposition and reconstruction results of 8 layers in an exemplary embodiment of the present disclosure;

[0051] Fig. 9 A schematic diagram schematically showing a 10-layer decomposition and reconstruction result in an exemplary embodiment of the present disclosure;

[0052] Fig.10 Schematically showing the results of error analysis of different algorithms in the exemplary embodiments of the present disclosure;

[0053] Fig.11 A schematic diagram of a flow chart of a signal bandwidth estimation method based on wavelet reconstruction in an exemplary embodiment of the present disclosure is schematically shown;

[0054] Fig.12 A schematic diagram schematically shows the composition of a signal bandwidth estimation device based on wavelet reconstruction in an exemplary embodiment of the present disclosure;

[0055] Fig.13 The following is a schematic diagram schematically showing the composition of a terminal device in an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0056] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0057] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0058] In view of the shortcomings and deficiencies of the prior art, this exemplary embodiment provides a signal bandwidth estimation method based on wavelet reconstruction, which can be applied to smart terminal devices such as mobile phones and tablet computers, or can also be applied to network side devices such as base stations. Figure 1 As shown in , the above-mentioned signal bandwidth estimation method based on wavelet reconstruction may include:

[0059] Step S11, performing segment processing on the signal to be processed to obtain a plurality of segment signals, and preprocessing the plurality of segment signals to obtain amplitude spectrum data after noise reduction;

[0060] Step S12, performing wavelet low-frequency reconstruction processing on the amplitude spectrum data to extract the corresponding envelope signal;

[0061] Step S13: determining the boundary of the envelope signal to estimate the bandwidth of the signal to be processed according to the boundary of the envelope signal.

[0062] The signal bandwidth estimation method based on wavelet reconstruction provided in this example implementation achieves signal denoising and reduces the influence of noise by segmenting and preprocessing the signal to be processed; and extracts the corresponding envelope signal by performing wavelet low-frequency reconstruction processing on the amplitude spectrum data, and finds the boundary of the signal spectrum for the envelope signal, and finally estimates the signal bandwidth. The method disclosed in this disclosure can be applied to a variety of different signals, which not only reduces the complexity of the algorithm, but also weakens the influence of signal randomness on the estimation accuracy; it has strong robustness.

[0063] In the following, each step of the signal bandwidth estimation method based on wavelet reconstruction in this example implementation will be described in more detail with reference to the accompanying drawings and embodiments.

[0064] In step S11, the signal to be processed is segmented to obtain a plurality of segmented signals, and the plurality of segmented signals are pre-processed to obtain amplitude spectrum data after noise reduction.

[0065] In this example implementation, the above step S11 may include:

[0066] Step S111, performing cross-correlation operation on a plurality of segmented signals according to the signal segmentation order to obtain a noise reduction signal;

[0067] Step S112, performing discrete Fourier transform on the noise reduction signal to obtain corresponding spectrum data;

[0068] Step S113, converting the frequency spectrum data into corresponding amplitude spectrum data.

[0069] For example, the above-mentioned signal to be processed may be a radio frequency signal containing noise. For example, the signal to be processed may be divided into three segments of equal length; for example, the noisy signal x(n) may be divided into three segments of equal length, namely x1(n), x2(n), and x3(n). Of course, in some exemplary embodiments of the present disclosure, the signal to be processed may also be divided into four, five or other segments of equal length to obtain a plurality of segmented signals.

[0070] For the continuous multiple segmented signals after division, the first segmented signal sequence and the second segmented signal sequence can be cross-correlated in the order of segmentation to obtain a first cross-correlation result; then the first cross-correlation result and the third segmented signal sequence are cross-correlated to obtain a second cross-correlation result; and so on, the continuous multiple segmented signal sequences are cross-correlated to obtain a noise reduction signal. For example, taking the signal to be processed as divided into three segmented signals as an example; the first segmented signal sequence and the second segmented signal sequence are cross-correlated, and the calculation formula of the cross-correlation can include formula (1):

[0071]

[0072] Among them, R1(m) represents the first cross-correlation result; m represents the sequence x2(n) shifted by m units.

[0073] The obtained cross-correlation sequence R1(m) is subjected to cross-correlation operation with the third segment signal sequence. The calculation formula of the cross-correlation may include formula (2):

[0074]

[0075] Wherein, R1(n) is R1(m) in the above formula (1).

[0076] When implementing signal denoising by utilizing the strong correlation property of the signal itself and obtaining the denoised signal, its spectrum can be obtained by performing a discrete Fourier transform on it. The calculation formula can include formula (3):

[0077]

[0078] where 0 ≤ k < N - 1; R2(n) is R2(m) in the above formula (2); N is the number of frequency domain sampling points; X R (k) is the spectrum after segmented cross-correlation.

[0079] After obtaining the spectrum data, it can also be transformed to obtain the corresponding magnitude spectrum |X R (k)|.

[0080] In step S12, wavelet low-frequency reconstruction processing is performed on the magnitude spectrum data to extract the corresponding envelope signal.

[0081] In the embodiment of this example, the above step S12 may include:

[0082] Step S121, based on the wavelet basis and the number of wavelet decomposition layers determined in advance, using a low-pass filter to recursively decompose the magnitude spectrum data to obtain target low-frequency data;

[0083] Step S122, based on the target low-frequency data, perform recursive reconstruction according to the determined number of wavelet decomposition layers to obtain the envelope signal.

[0084] Specifically, the wavelet basis and the number of wavelet decomposition layers can be configured according to prior experience. For example, the db4 wavelet basis can be configured; the number of wavelet decomposition layers is 10 layers. Based on the determined db4 wavelet basis, the filter coefficients of the corresponding wavelet filter can be determined. Among them, the filter coefficients can include wavelet translation coefficients and scaling coefficients; they can be used as the coefficients of low-pass filters and high-pass filters.

[0085] In the embodiment of this example, in the above step S121, it may specifically include: determining the filter coefficients according to the wavelet basis determined in advance; in the (n + 1)-th layer, perform a convolution operation on the decomposition result of the n-th layer and the filter coefficients, perform downsampling on the convolution operation result, and use the downsampling result as the decomposition result of the (n + 1)-th layer; and repeat this process until the preset number of wavelet decomposition layers to obtain the target low-frequency coefficients. Among them, n is a positive integer, and the magnitude spectrum data is configured as the input parameter of the first layer.

[0086] Specifically, for the magnitude spectrum |X R(k)|, when performing wavelet decomposition, the Mallat algorithm can be used to recursively decompose through the low-pass filter g(-n) to obtain the corresponding low-frequency information. Simultaneously, the high-pass filter h(-n) can also be used to recursively decompose to obtain the corresponding high-frequency information. Specifically, when performing wavelet decomposition, at the first level, the amplitude spectrum |X R (k)| and the filter coefficient are used as the input parameters of the low-frequency filter, and a convolution operation is performed on them. Then, the convolution result sequence is downsampled, such as sampling at alternate points, to obtain the low-frequency information of the first layer. Then, the low-frequency information and the filter coefficient of the first layer are used as the input parameters of the second layer, and a convolution operation is performed on them. Then, the convolution result sequence is downsampled to obtain the low-frequency information of the second layer. And so on, the low-frequency information of each layer of decomposition can be obtained; until the set number of decomposition layers is met, the wavelet decomposition ends. The calculation formula of the wavelet decomposition can specifically include formula (4):

[0087]

[0088] Among them, A j+1 (n) is the low-frequency information; j is the number of layers, k is the translation unit; G is the low-pass filter coefficient.

[0089] For high-frequency information, the same method can also be used for recursive decomposition. The specific calculation formula can include formula (5):

[0090]

[0091] Among them, AD j+1 (n) is the high-frequency information; j is the number of layers, k is the translation unit; H is the high-pass filter coefficient.

[0092] For example, refer to Figure 2 As shown, taking the decomposition of the amplitude spectrum data into three layers as an example, the amplitude spectrum |X R (k)|Perform the first decomposition to obtain the low-frequency information cA1 and high-frequency information cD1 of the first layer; perform the second decomposition on the low-frequency information cA1 of the first layer to obtain the low-frequency information cA2 and high-frequency information cD2 of the second layer; perform the third decomposition on the low-frequency information cA2 of the second layer to obtain the low-frequency information cA3 and high-frequency information cD3 of the third layer.

[0093] In this example implementation, in the above-mentioned step S122, it can specifically include: setting the high-frequency information of the nth layer to zero, upsampling the low-frequency data of the nth layer, performing a convolution operation based on the upsampling result and the filter coefficient, and configuring the convolution operation result as the low-frequency information of the n-1th layer; repeating this step to the top layer to complete the recursive reconstruction and obtain the envelope signal; wherein n is a positive integer; and the target low-frequency data is configured as the low-frequency data of the last layer.

[0094] Specifically, after recursively decomposing the amplitude spectrum signal to obtain the low-frequency information of the last layer, since the envelope information of the signal is contained in the low-frequency information, low-frequency reconstruction is adopted. Specifically, the high-frequency information obtained by the last layer of decomposition can be set to zero, and then the low-frequency information can be upsampled, such as sampling at alternate points, and then convolved with the wavelet filter coefficients mentioned in the wavelet decomposition to achieve a reconstruction. The reconstruction result of this time is then used as the low-frequency information of the previous layer; and so on, until the first layer of coefficients is reconstructed, the envelope Y(k) of the signal can be reconstructed. Specifically, the calculation formula for reconstruction can include formula (6):

[0095]

[0096] For example, taking the decomposition of the amplitude spectrum data into three layers as an example, the reconstruction process can include formula (7):

[0097] cA3+cD3=cA2

[0098] cA2+cD2=cA1

[0099] cA1+cD1=Y(k)

[0100] Among them, cA n is low-frequency information; cD n is the high frequency information; Y(k) is the envelope signal.

[0101] Since the envelope information of the signal is contained in the low-frequency information, we use low-frequency reconstruction. The high-frequency information cD obtained by decomposing each layer n After setting to zero, reconstructing the signal can complete the low-frequency reconstruction.

[0102] In step S13, the boundary of the envelope signal is determined to estimate the bandwidth of the signal to be processed according to the boundary of the envelope signal.

[0103] In this example implementation, the above step S13 may include:

[0104] Step S131, normalizing the envelope signal, and performing a first forward difference operation on the normalized result to obtain a first difference result;

[0105] Step S132, performing segment processing on the first differential result, and performing a second forward differential operation on the segment processing result to obtain a second differential result;

[0106] Step S133, using a preset threshold to filter the maximum value of the second difference result to obtain a maximum value set;

[0107] Step S134, determining a left boundary and a right boundary according to the maximum value set;

[0108] Step S135, determining a first minimum value index based on the left boundary and configuring it as an upper frequency band of the spectrum, and determining a second minimum value index based on the right boundary and configuring it as a lower frequency band of the spectrum;

[0109] Step S136: estimating the bandwidth of the signal to be processed according to the upper frequency band of the frequency spectrum and the lower frequency band of the frequency spectrum.

[0110] Specifically, for the amplitude spectrum |X R The reconstructed signal Y(k)| is normalized, and its formula may include formula (8):

[0111]

[0112] Among them, G(k) is the normalized result.

[0113] Then, we can perform a forward difference operation on G(k), which can be expressed as Then target Segment processing:

[0114]

[0115] Then, you can Perform the forward difference operation again, recorded as in:

[0116] like This point is the minimum value;

[0117] like This point is the maximum value.

[0118] Based on the result of the second forward difference operation, all extreme points of G(k) can be determined, and a threshold K is set to filter out all maximum points that satisfy a value greater than K, thereby obtaining a maximum value set; specifically, the formula (9) can be included:

[0119]

[0120] in, is the normalized spectral envelope mean, = is the signal length.

[0121] In the maximum value set selected based on the above method, the leftmost and rightmost values ​​are respectively used as the left boundary left1 and the right boundary right1. Then, the minimum value index left2 closest to the left boundary left1 is found as the upper frequency band of the spectrum; and the minimum value index right2 closest to the right boundary right1 is found as the lower frequency band of the spectrum. The bandwidth is determined based on the upper frequency band and the lower frequency band of the spectrum. Then the calculation formula of the zero-crossing full bandwidth of the signal can include formula (10):

[0122] band=right2-left2

[0123] In some exemplary embodiments, a BPSK (Binary Phase Shift Keying) signal is used for simulation. The simulation parameters may include: carrier frequency: 60 MHz; code rate: 20 MB; number of code elements: 10,000; sampling rate: 400 MHz; signal-to-noise ratio: 20 dB; wavelet basis: db4; number of wavelet decomposition layers: 10. When the effective signal s(n) is transmitted in the channel, it is affected by the noise signal q(n) generated by the channel, and the noisy signal x(n) = s(n) + q(n) is obtained. Figure 3 As shown in the figure, it is the amplitude spectrum of the pure BPSK signal after adding noise, which is used to simulate the scene where the signal is interfered by noise during channel transmission. It can be seen that the zero-crossing full bandwidth point of the amplitude spectrum after adding noise is completely buried by noise. Direct bandwidth estimation will produce a large accuracy error. After the noisy signal is denoised by the segmented cross-correlation processing method, its amplitude spectrum is analyzed. The amplitude spectrum of the data after segmented correlation is shown in Figure 4 As shown. In wavelet decomposition and reconstruction, different wavelet bases give different reconstruction results. Figure 5 As shown in FIG. 1 , the amplitude diagram is shown, which shows the amplitude spectrum of the segmented correlation data reconstructed using Haar wavelet and dbN (Daubechies) wavelet. Figure 5As shown in the figure, due to the poor regularity of Haar wavelet, the waveform reconstructed by it has poor differentiability and is not smooth enough, which increases the error and complexity of the subsequent bandwidth estimation algorithm. The waveform reconstructed by dbN wavelet is relatively smooth, which can effectively reduce the impact of signal randomness on bandwidth estimation. dbN wavelet is also called compactly supported orthogonal wavelet, and N is the order of the wavelet. The dbN wavelet has good regularity, so the smoothness error introduced by the wavelet as a sparse basis is not easy to detect, making the reconstructed signal relatively smooth. The characteristic of dbN wavelet is that as the order (sequence N) increases, the order of the vanishing moment increases, and the higher the vanishing moment, the better the smoothness, the stronger the localization ability in the frequency domain, and the better the frequency band division effect, but it will weaken the compact support in the time domain, and at the same time greatly increase the amount of calculation, and the real-time performance will deteriorate. Therefore, the dbN wavelet needs to weigh its advantages and disadvantages when determining the size of N. According to the characteristics of the dbN wavelet described in the previous article, the analysis Figure 6 It can be seen that when N is equal to 4, the relative error of bandwidth estimation is small. When N is further increased, the relative error of bandwidth estimation is not greatly optimized. However, since the increase of N will increase the amount of calculation and the real-time performance of the signal will deteriorate, the db4 wavelet basis is selected. The decomposition of the signal is a recursive decomposition process. When the set number of decomposition layers is reached, the recursion ends. According to Figure 7 The simulation results show that when the number of decomposition layers reaches eight, increasing the number of decomposition layers again will cause the relative error of bandwidth estimation to drop rapidly. When the number of wavelet decomposition layers reaches 10, the relative error of bandwidth estimation is the smallest. Figure 8 , Fig. 9 As shown in the figure, when the number of decomposition layers is 8, the low-frequency information of this layer also contains some detailed information of the signal. The reconstructed envelope smoothing effect is poor due to the influence of this detailed information, causing the upper frequency band point left2 and the lower frequency band point right2 of the bandwidth estimation to move up. Therefore, the estimation error is large. Until the 10th layer, the low-frequency information of this layer can better reflect the envelope information of the signal. If the number of decomposition layers is increased again, some envelope detail information will be lost.

[0124] Traditional bandwidth estimation algorithms include the root mean square method, autocorrelation method, energy concentration method, etc. These algorithms are all based on -3dB bandwidth research. Since the autocorrelation bandwidth estimation algorithm is simple in principle, it is still widely used in practical engineering. Fig.10 As shown, the traditional autocorrelation bandwidth estimation algorithm and the bandwidth estimation algorithm based on wavelet reconstruction proposed in the present disclosure are selected for performance comparison analysis. Under various signal-to-noise ratio conditions, the bandwidth estimation effect based on wavelet reconstruction is better than the traditional algorithm. This algorithm can well overcome the negative impact of signal randomness on bandwidth estimation. It has good robustness. As the signal-to-noise ratio increases, the error continues to decrease. When the signal-to-noise ratio is greater than 10db, the accuracy can reach more than 96%.

[0125] The simulation experiments verify that the bandwidth estimation algorithm proposed in the present invention is superior to the traditional bandwidth estimation algorithm, which provides great convenience for the design of filter parameters in single-channel multi-signal selection and has certain research significance.

[0126] The signal bandwidth estimation method based on wavelet reconstruction provided by the embodiment of the present disclosure is as follows: Fig.11 As shown in the figure, the signal to be processed is first segmented, and the cross-correlation operation is performed to achieve signal noise reduction; the cross-correlation data is discrete Fourier transformed to obtain its spectrum and transformed into amplitude spectrum data; the amplitude spectrum data is reconstructed by wavelet low-frequency reconstruction to obtain the reconstructed signal Y(k); in the process of extreme point search, the reconstructed signal is normalized to obtain G(k), and a difference operation, segmentation processing, and secondary difference operation are performed to determine the extreme point; in the boundary search process, the left boundary and the boundary are searched to determine, and then the zero-crossing full bandwidth boundary is estimated based on the determined boundary. This method reduces the influence of noise by means of data segmentation cross-correlation, and extracts the envelope signal by means of low-frequency reconstruction by selecting the appropriate number of wavelet decomposition layers through wavelet decomposition and low-frequency reconstruction, so as to suppress the negative influence of signal randomness as much as possible. Under various signal-to-noise ratio conditions, the bandwidth estimation algorithm based on wavelet reconstruction is better than the traditional bandwidth estimation algorithm. And it has good robustness.

[0127] It should be noted that the above figures are only schematic illustrations of the processes included in the method according to an exemplary embodiment of the present invention, and are not intended to be limiting. It is easy to understand that the processes shown in the above figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0128] For further reference, Fig.12 As shown, in the implementation of this example, a signal bandwidth estimation device 120 based on wavelet reconstruction is also provided, and the device includes: a signal preprocessing module 1201, a signal reconstruction module 1202, and a bandwidth estimation module 1203. Among them,

[0129] The signal preprocessing module 1201 can be used to perform segment processing on the signal to be processed to obtain multiple segment signals, and preprocess the multiple segment signals to obtain noise-reduced amplitude spectrum data.

[0130] The signal reconstruction module 1202 may be used to perform wavelet low-frequency reconstruction processing on the amplitude spectrum data to extract a corresponding envelope signal.

[0131] The bandwidth estimation module 1203 may be used to determine the boundary of the envelope signal, so as to estimate the bandwidth of the signal to be processed according to the boundary of the envelope signal.

[0132] In some exemplary embodiments, the signal preprocessing module 1201 can be used to perform cross-correlation operations on multiple segmented signals according to the signal segmentation order to obtain a noise reduction signal; perform discrete Fourier transform on the noise reduction signal to obtain corresponding spectrum data; and convert the spectrum data into corresponding amplitude spectrum data.

[0133] In some exemplary embodiments, the signal reconstruction module 1202 can be used to recursively decompose the amplitude spectrum data using a low-frequency filter based on a wavelet basis of a predetermined order and a wavelet decomposition layer number to obtain target low-frequency data; and recursively reconstruct the target low-frequency data according to the determined number of wavelet decomposition layers to obtain the envelope signal.

[0134] In some exemplary embodiments, the signal reconstruction module 1202 may include: determining the filter coefficients based on a wavelet basis of a predetermined order; at the n+1 layer, performing a convolution operation on the decomposition result of the n layer and the filter coefficients, downsampling the convolution operation result, and using the downsampling result as the decomposition result of the n+1 layer; and repeating the process to a preset number of wavelet decomposition layers to obtain the target low-frequency coefficients, wherein n is a positive integer and the amplitude spectrum data is configured as an input parameter of the first layer.

[0135] In some exemplary embodiments, the signal reconstruction module 1202 may include: setting the high-frequency information of the nth layer to zero, upsampling the low-frequency data of the nth layer, performing a convolution operation based on the upsampling result and the filter coefficient, and configuring the convolution operation result as the low-frequency information of the n-1th layer; repeating this step to the top layer to complete the recursive reconstruction and obtain the envelope signal; wherein n is a positive integer; and the target low-frequency data is configured as the low-frequency data of the last layer.

[0136] In some exemplary embodiments, the bandwidth estimation module 1203 can be used to normalize the envelope signal, and perform a first forward difference operation on the normalized result to obtain a first difference result; segment the first difference result, and perform a second forward difference operation on the segmented processing result to obtain a second difference result; use a preset threshold to filter the second difference result for maximum values ​​to obtain a maximum value set; determine a left boundary and a right boundary based on the maximum value set; determine a first minimum value index based on the left boundary and configure it as an upper frequency band of the spectrum, and determine a second minimum value index based on the right boundary and configure it as a lower frequency band of the spectrum; and estimate the bandwidth of the signal to be processed based on the upper frequency band of the spectrum and the lower frequency band of the spectrum.

[0137] In some exemplary embodiments, the bandwidth of the signal to be processed is the zero-crossing full bandwidth of the signal to be processed.

[0138] The specific details of each module in the above-mentioned signal bandwidth estimation device 120 based on wavelet reconstruction have been described in detail in the corresponding signal bandwidth estimation method based on wavelet reconstruction, so they will not be repeated here.

[0139] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0140] Fig.13 A schematic diagram of a terminal device suitable for implementing an embodiment of the present invention is shown.

[0141] It should be noted that Fig.13 The terminal device 1000 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0142] like Fig.13 As shown, the terminal device 1000 includes a central processing unit (CPU) 1001, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 1002 or the program loaded from the storage part 1008 to the random access memory (RAM) 1003. In the RAM 1003, various programs and data required for system operation are also stored. The CPU 1001, the ROM 1002, and the RAM 1003 are connected to each other through a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0143] The following components are connected to the I / O interface 1005: an input section 1006 including a keyboard, a mouse, etc.; an output section 1007 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1010 as needed, so that a computer program read therefrom is installed into the storage section 1008 as needed.

[0144] In particular, according to an embodiment of the present invention, the process described below with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part 1009, and / or installed from a removable medium 1011. When the computer program is executed by a central processing unit (CPU) 1001, various functions defined in the system of the present application are executed.

[0145] Specifically, the terminal device may be a smart mobile terminal device such as a mobile phone, a tablet computer or a laptop computer, or may be a smart terminal device such as a desktop computer.

[0146] It should be noted that the computer-readable medium shown in the embodiment of the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate, or transmit programs for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0147] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0148] The units involved in the embodiments of the present invention may be implemented by software or hardware, and the units described may also be arranged in a processor. The names of these units do not, in some cases, limit the units themselves.

[0149] It should be noted that, as another aspect, the present application also provides a computer-readable medium, which may be included in an electronic device; or may exist independently without being installed in the electronic device. The above-mentioned computer-readable medium carries one or more programs, and when the above-mentioned one or more programs are executed by an electronic device, the electronic device implements the method described in the following embodiments. For example, the electronic device may implement the following Figure 1 The steps shown.

[0150] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to an exemplary embodiment of the present invention, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0151] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary technical means in the art that are not disclosed in the present disclosure. The specification and examples are to be considered as exemplary only, and the true scope and spirit of the present disclosure are indicated by the claims.

[0152] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A signal bandwidth estimation method based on wavelet reconstruction, characterized in that: The method comprises: Performing segment processing on the signal to be processed to obtain a plurality of segment signals, and preprocessing the plurality of segment signals to obtain amplitude spectrum data after noise reduction; Performing wavelet low-frequency reconstruction processing on the amplitude spectrum data to extract the corresponding envelope signal; Determining a boundary of the envelope signal to estimate a bandwidth of the signal to be processed according to the boundary of the envelope signal comprises: Normalizing the envelope signal, and performing a first forward difference operation on the normalized result to obtain a first difference result; Performing segment processing on the first differential result, and performing a second forward differential operation on the segment processing result to obtain a second differential result; Using a preset threshold to filter the second difference result for a maximum value to obtain a maximum value set; Determine a left boundary and a right boundary according to the maximum value set; Determine a first minimum value index based on the left boundary and configure it as an upper frequency band of the spectrum, and determine a second minimum value index based on the right boundary and configure it as a lower frequency band of the spectrum; The bandwidth of the signal to be processed is estimated according to the upper frequency band of the frequency spectrum and the lower frequency band of the frequency spectrum.

2. The signal bandwidth estimation method based on wavelet reconstruction according to claim 1 is characterized in that: The preprocessing of the plurality of segmented signals to obtain noise-reduced amplitude spectrum data comprises: Performing cross-correlation operation on multiple segmented signals according to the signal segmentation order to obtain a noise reduction signal; Performing discrete Fourier transform on the noise reduction signal to obtain corresponding spectrum data; The frequency spectrum data is converted into corresponding amplitude spectrum data.

3. The signal bandwidth estimation method based on wavelet reconstruction according to claim 1 is characterized in that: The performing wavelet low-frequency reconstruction processing on the amplitude spectrum data to extract the corresponding envelope signal includes: Based on a wavelet basis of a predetermined order and a wavelet decomposition layer number, the amplitude spectrum data is recursively decomposed using a low-frequency filter to obtain target low-frequency data; Recursive reconstruction is performed based on the target low-frequency data according to the determined number of wavelet decomposition layers to obtain the envelope signal.

4. The signal bandwidth estimation method based on wavelet reconstruction according to claim 3 is characterized in that: The step of recursively decomposing the amplitude spectrum data by using a low-frequency filter to obtain target low-frequency data includes: Determine the filter coefficients according to a wavelet basis of a predetermined order; At the n+1th layer, a convolution operation is performed on the decomposition result and the filter coefficient of the nth layer, the convolution operation result is downsampled, and the downsampled result is used as the decomposition result of the n+1th layer; and, the process is repeated to a preset number of wavelet decomposition layers to obtain the target low-frequency coefficient; wherein n is a positive integer, and the amplitude spectrum data is configured as an input parameter of the first layer.

5. The signal bandwidth estimation method based on wavelet reconstruction according to claim 3 or 4, characterized in that: The recursive reconstruction based on the target low-frequency data according to the determined wavelet decomposition layer number to obtain the envelope signal includes: The high-frequency information of the nth layer is set to zero, the low-frequency data of the nth layer is up-sampled, a convolution operation is performed based on the up-sampling result and the filter coefficient, and the convolution operation result is configured as the low-frequency information of the n-1th layer; this step is repeated to the top layer to complete the recursive reconstruction and obtain the envelope signal; wherein n is a positive integer; the target low-frequency data is configured as the low-frequency data of the last layer.

6. The signal bandwidth estimation method based on wavelet reconstruction according to claim 1 is characterized in that: The bandwidth of the signal to be processed is the zero-crossing full bandwidth of the signal to be processed.

7. A signal bandwidth estimation device based on wavelet reconstruction, characterized in that: The device comprises: A signal preprocessing module is used to perform segment processing on the signal to be processed to obtain multiple segment signals, and preprocess the multiple segment signals to obtain amplitude spectrum data after noise reduction; A signal reconstruction module, used for performing wavelet low-frequency reconstruction processing on the amplitude spectrum data to extract the corresponding envelope signal; A bandwidth estimation module, used to determine the boundary of the envelope signal to estimate the bandwidth of the signal to be processed according to the boundary of the envelope signal, including: normalizing the envelope signal, and performing a first forward difference operation on the normalized result to obtain a first difference result; segmenting the first difference result, and performing a second forward difference operation on the segmented processing result to obtain a second difference result; using a preset threshold to filter the second difference result for a maximum value to obtain a maximum value set; determining a left boundary and a right boundary according to the maximum value set; determining a first minimum value index based on the left boundary and configuring it as an upper frequency band of the spectrum, and determining a second minimum value index based on the right boundary and configuring it as a lower frequency band of the spectrum; estimating the bandwidth of the signal to be processed according to the upper frequency band of the spectrum and the lower frequency band of the spectrum.

8. A computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the signal bandwidth estimation method based on wavelet reconstruction according to any one of claims 1 to 6 is implemented.

9. A terminal device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to execute the signal bandwidth estimation method based on wavelet reconstruction according to any one of claims 1 to 6 by executing the executable instructions.

Images