A robust method for estimating the velocity of a moving sound source based on line spectral doppler shift

Abstract

The embodiment of the application discloses a kind of robust methods for estimating the speed of moving sound source based on line spectrum Doppler frequency shift, method includes: initialization calculation parameter, the calculation parameter includes window function length, Fourier transform length and overlap length;According to the calculation parameter, short-time Fourier transform is carried out to obtain the STFT time-frequency spectrogram of received signal;According to the STFT time-frequency spectrogram, the Doppler relevant parameter of non-iteration is calculated, the Doppler relevant parameter includes line spectrum frequency, speed, positive lateral time and positive lateral distance;According to STFT time-frequency spectrogram, calculate time-frequency energy concentration;According to Doppler CT time-frequency analysis algorithm, the Doppler relevant parameter is iteratively processed;When reaching the iteration stop requirement, the final Doppler relevant parameter is output.The robustness of Doppler relevant parameter estimation process such as moving sound source speed, positive lateral distance in low signal-to-noise ratio environment is significantly improved.

Description

Technical Field

[0001] This invention relates to the field of signal processing technology, and more specifically to a robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift. Background Technology

[0002] Due to the rotation of mechanical structures, targets such as automobiles, airplanes, and ships inevitably radiate noise during their movement. Generally, target radiated noise signals can be divided into two types: line spectrum noise and broadband noise. Line spectrum noise has a significantly higher source level than broadband noise and can carry characteristic information of the target sound source over longer distances. Therefore, line spectrum noise information is often used for target identification and azimuth tracking. When there is relative motion between the line spectrum sound source and the receiving point, the instantaneous frequency of the received line spectrum signal will experience a Doppler shift. Based on the Doppler shift, Doppler parameters such as the velocity of the moving sound source and its lateral distance can be estimated.

[0003] Currently, the most common method for estimating the velocity of moving sound sources is to extract the instantaneous frequency curve of the line spectrum and then estimate it using the least squares curve fitting technique based on the theoretical formula of Doppler frequency shift. Early researchers estimated the Doppler frequency shift curve in the received signal using phase interpolation. With the rise of time-frequency analysis techniques, in 1994, Ferguson and Quinn proposed using the short-time Fourier transform (STFT, also known as spectrogram) and the Wigner-Ville distribution (WVD) to extract the Doppler frequency shift curve from the received signal. Subsequently, to further improve the estimation accuracy of the Doppler frequency shift curve, researchers explored numerous high-resolution time-frequency analysis algorithms, including polynomial Wigner–Ville distribution, polynomial chirplet transform (PCT), matched Wigner transform (MWT), enhanced spline-kernelled chirplet transform (ESCT), Doppler chirplet transform (DopplerCT), and Wigner–Ville distribution cross-term rejection (WVD-CTR). Compared to the classic STFT method, high-resolution time-frequency analysis algorithms produce line spectrum signals with higher time-frequency energy concentration in the time-frequency plots, thus improving the extraction accuracy of the Doppler frequency shift curve to some extent. Theoretically, this contributes to robust motion source velocity estimation under low signal-to-noise ratio conditions. However, in practical applications, it has been found that simply increasing the time-frequency energy concentration of the line spectrum signal in the time-frequency plot has limited effect on improving the robustness of parameter estimation under low signal-to-noise ratio conditions.

[0004] High-resolution time-frequency analysis algorithms typically improve the time-frequency energy concentration and parameter estimation accuracy of a signal through iteration. This involves repeatedly calculating the time-frequency graph, extracting the instantaneous frequency curve of the line spectrum, and estimating the velocity of the moving sound source based on curve fitting until convergence. The most error-prone part of this process is the initial iteration. During the initial iteration, the time-frequency energy concentration of the line spectrum signal in the time-frequency graph is relatively low. When the signal-to-noise ratio (SNR) is low, noise makes it difficult to accurately extract the instantaneous frequency curve of the line spectrum, leading to an incorrect velocity estimate of the moving sound source after curve fitting. Continuing the next iteration based on this incorrect parameter estimation result only makes convergence of the parameter estimation results difficult. Based on the above analysis, these algorithms are actually quite sensitive to noise and generally struggle to adapt to low SNR environments. However, in practical applications, low SNR environments are unavoidable due to factors such as the distance between the receiver and the sound source and high background noise.

[0005] Therefore, it is necessary to design a method for estimating the velocity of moving sound sources that is suitable for low signal-to-noise ratio environments. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift, aiming to improve the robustness of the estimation process of Doppler-related parameters such as the velocity and lateral distance of a moving sound source in low signal-to-noise ratio environments.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] Firstly, a robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift includes:

[0009] Initialize the calculation parameters, which include the window function length, Fourier transform length, and overlap length;

[0010] Perform a short-time Fourier transform based on the calculated parameters to obtain the STFT time spectrum of the received signal;

[0011] The Doppler parameters at the non-iteration time are calculated based on the STFT time spectrum diagram. The Doppler parameters include line spectrum frequency, velocity, positive transverse time, and positive transverse distance.

[0012] Calculate the time-frequency energy concentration based on the STFT time-frequency spectrum diagram;

[0013] The relevant parameters of Doppler are iteratively processed according to the DopplerCT time-frequency analysis algorithm;

[0014] Once the iteration stops, the final Doppler parameters are output.

[0015] The further technical solution is as follows: the initialization calculation parameters include the window function length, the Fourier transform length, and the overlap length, including:

[0016] Set the calculation parameters required for calculating the STFT time-frequency diagram;

[0017] Set the calculation parameters required to calculate the DopplerCT time-frequency plot.

[0018] The further technical solution is that the window function length and Fourier transform length required for the DopplerCT time-frequency graph are both greater than the window function and Fourier transform length required for the STFT time-frequency graph.

[0019] The further technical solution is as follows: The Doppler-related parameters at the non-iteration stage are calculated based on the STFT time-spectrum diagram. These Doppler-related parameters include line spectrum frequency, velocity, transverse time, and transverse distance. The formula for calculating the Doppler-related parameters at the non-iteration stage is:

[0020]

[0021] In the formula, P(t,f) represents the time-frequency graph, t represents time, f represents frequency, σ represents the thickness of the instantaneous frequency curve, and f u f is the upper bound of the frequency range. d This is the lower bound of the frequency range.

[0022] The further technical solution is as follows: In the calculation of time-frequency energy concentration based on the STFT time-frequency spectrum, the formula for calculating the time-frequency energy concentration is:

[0023]

[0024] In the formula, H α (P) represents the Renyi entropy of P(t,f), which is used to measure the time-frequency energy concentration of P(t,f). α represents the order of the Renyi entropy, and ∫∫P(u,v)dudv represents the total energy of P(t,f) calculated by integration.

[0025] The further technical solution is as follows: the iterative processing of Doppler-related parameters based on the DopplerCT time-frequency analysis algorithm includes:

[0026] The DopplerCT time-frequency diagram was calculated based on relevant Doppler parameters.

[0027] The Doppler parameters after iterative updates were calculated based on the DopplerCT time-frequency plot.

[0028] Calculate the time-frequency energy concentration based on the DopplerCT time-frequency plot;

[0029] The iteration should be terminated based on the time-frequency energy concentration.

[0030] The further technical solution is as follows: the DopplerCT time-frequency diagram is calculated based on relevant Doppler parameters, and the formula for calculating the DopplerCT time-frequency diagram is as follows:

[0031]

[0032]

[0033] In the formula, DopplerCT(t,f,p) represents the DopplerCT time-frequency diagram calculated under parameter p, where τ represents the integral of t, y represents the received line spectrum signal, w represents the window function, c represents the speed of sound, and parameter p = [f0,c,v,τ]. c ,d c [i.e., line spectrum frequency f0, sound speed c, velocity v, and positive transverse time τ] c and the vertical and horizontal distance d c .

[0034] Secondly, a robust device for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift includes an initialization unit, an STFT time-spectrum diagram calculation unit, a preliminary Doppler parameter estimation unit, a time-frequency energy concentration calculation unit, an iterative processing unit, and a final Doppler parameter output unit.

[0035] The initialization unit is used to initialize the calculation parameters, which include the window function length, the Fourier transform length, and the overlap length.

[0036] The STFT time spectrum calculation unit is used to perform a short-time Fourier transform according to the calculation parameters to obtain the STFT time spectrum of the received signal.

[0037] The preliminary Doppler parameter estimation unit is used to calculate the Doppler-related parameters before iteration based on the STFT time spectrum diagram. The Doppler-related parameters include line spectrum frequency, velocity, positive transverse time, and positive transverse distance.

[0038] The time-frequency energy concentration calculation unit is used to calculate the time-frequency energy concentration based on the STFT time-frequency spectrum diagram;

[0039] The iterative processing unit is used to iteratively process Doppler-related parameters according to the DopplerCT time-frequency analysis algorithm;

[0040] The final Doppler parameter output unit is used to output the final Doppler parameters after the iteration stop requirement is met.

[0041] Thirdly, a computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift as described above.

[0042] Fourthly, a computer-readable storage medium storing a computer program including program instructions that, when executed by a processor, cause the processor to perform the robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift as described above.

[0043] The advantages of this invention compared to existing technologies are as follows: By initializing the calculation parameters, a short-time Fourier transform (STFT) is performed based on these parameters to obtain the STFT time-frequency spectrum of the received signal; Doppler-related parameters are calculated from the STFT time-frequency spectrum before iteration; the time-frequency energy concentration is calculated from the STFT time-frequency spectrum; the Doppler-related parameters are iteratively processed using the DopplerCT time-frequency analysis algorithm; and the final Doppler-related parameters are output after the iteration stops. By utilizing the energy accumulation distribution in the time-frequency spectrum to estimate Doppler-related parameters and using the time-frequency energy concentration as the criterion for iterative convergence, the problem of extracting the instantaneous frequency curve of the line spectrum is avoided, thus significantly improving the robustness of the estimation process for Doppler-related parameters such as the velocity and lateral distance of moving sound sources in low signal-to-noise ratio environments.

[0044] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other objectives, features and advantages of the present invention more obvious and understandable, preferred embodiments are described in detail below. Attached Figure Description

[0045] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 A flowchart illustrating a robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift, provided as a specific embodiment of the present invention;

[0047] Figure 2 A schematic block diagram of a robust device for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift, provided for a specific embodiment of the present invention;

[0048] Figure 3A schematic block diagram of a computer device provided for a specific embodiment of the present invention;

[0049] Figure 4 The time-frequency diagram of STFT in a simulation when the signal-to-noise ratio is 0dB in the 116-122Hz frequency band of the truck line spectrum is provided for a specific embodiment of the present invention;

[0050] Figure 5 The iteration result of DopplerCT in a simulation when the signal-to-noise ratio is 0dB in the 116-122Hz frequency band of the truck line spectrum provided in an embodiment of the present invention;

[0051] Figure 6 A comparison chart of Doppler parameter estimation results provided in the embodiments of the present invention. Detailed Implementation

[0052] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0054] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0055] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0056] This invention provides a robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift. This method is particularly suitable for low signal-to-noise ratio (SNR) environments and can significantly improve the robustness of the estimation process for Doppler-related parameters such as the velocity and lateral distance of a moving sound source in low SNR environments.

[0057] This embodiment of the invention uses a segment of line spectrum noise signal from a truck actually received by the sensor as an example. The actual parameters of the truck noise signal are approximately: line spectrum frequency 118.7Hz, speed of movement 6.1m / s, horizontal distance 36.5m, horizontal distance 15.8s, air speed of sound 347m / s, total observation time 30s, and original sampling rate 1.2kHz. In this example, only one line spectrum within the 116-122Hz frequency band is analyzed; therefore, it is downsampled to 300Hz to reduce computational load. To simulate the effect of noise, white noise is added to achieve an in-band signal-to-noise ratio of 0dB. The specific details are explained below.

[0058] like Figure 1 As shown, a robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift includes the following steps: S10-S60.

[0059] S10. Initialize the calculation parameters, including the window function length, Fourier transform length, and overlap length.

[0060] The line spectrum noise signal from the truck can be received using a microphone-based sound acquisition device. Once the sound signal emitted by the sound source is acquired, preprocessing is required, including noise reduction, filtering, and enhancement.

[0061] In one embodiment, step S10 specifically includes the following steps: S101 and S102.

[0062] S101. Set the calculation parameters required for calculating the STFT time-frequency diagram;

[0063] S102. Set the calculation parameters required for calculating the DopplerCT time-frequency plot.

[0064] In this embodiment, for steps S101 and S102, the window function length and Fourier transform length required for the DopplerCT time-frequency plot are both greater than those required for the STFT time-frequency plot. Specifically, the three parameters required for calculating the STFT time-frequency plot are set as follows: window function length N. w =2048, Fourier transform length N F =4096 and overlap length N o =N w -30. Set the three parameters needed to calculate the DopplerCT time-frequency plot: window function length N. w =6144, Fourier transform length N F =8192 and overlap length N o =N w -30.

[0065] It should be noted that the window function and Fourier transform length of the DopplerCT time-frequency plot are both longer than those of the STFT time-frequency plot. This is mainly because high-resolution algorithms can achieve high resolution by utilizing longer window functions and Fourier transform lengths.

[0066] S20. Perform a short-time Fourier transform based on the calculated parameters to obtain the STFT time spectrum of the received signal.

[0067] The STFT time-frequency diagram represents the distribution of sound in time and frequency. It is obtained by performing a Short-time Fourier Transform (STFT) on the sound signal. It decomposes the signal in time and frequency, and can clearly show the changes of sound in different times and frequencies, which can then be used to analyze the frequency components and variation patterns of the sound signal.

[0068] In this embodiment, according to N w N F and N o Using three parameters, a Short-Time Fourier Transform (STFT) is performed to calculate the STFT time-frequency spectrum of the received signal, denoted as STFT(t,f). Based on the STFT time-frequency spectrum, the frequency range of the line spectrum signal and the "thickness" of the instantaneous frequency curve are roughly determined, with the upper bound of the frequency range denoted as f. u Hz, lower bound is f d Hz, the "thickness" of the instantaneous frequency curve is ΔHz.

[0069] like Figure 4 As shown, Figure 4 The figure shows the time-frequency plot of the STFT in the simulation when the signal-to-noise ratio is 0dB within the 116-122Hz frequency band of the truck's line spectrum. The dashed line in the figure represents the instantaneous frequency curve of the line spectrum extracted based on the peak energy. Figure 4 It can be seen that the instantaneous frequency curve undergoes a sudden change due to noise between 25 and 30 seconds. Based on the time-frequency diagram, the upper limit of the frequency range is roughly set as f. u =122Hz, lower bound is f d =116Hz, the "thickness" of the instantaneous frequency curve is Δ=0.5Hz.

[0070] S30. Calculate the relevant Doppler parameters before iteration based on the STFT time spectrum diagram. The relevant Doppler parameters include line spectrum frequency, velocity, positive transverse time, and positive transverse distance.

[0071] The formula for calculating the relevant Doppler parameters before iteration is as follows:

[0072] In the formula, P(t,f) represents the STFT time-frequency diagram, t represents time, f represents frequency, σ represents the thickness of the instantaneous frequency curve, and f u f is the upper bound of the frequency range. d This is the lower bound of the frequency range.

[0073] Using the above formula and least squares curve fitting (a commonly used parameter estimation method, the calculation formula is not elaborated here), the relevant Doppler parameters p = [f0, v, τ] are estimated. c ,d c [i.e., spectral frequency f0, velocity v, and positive transverse time τ] c and the vertical and horizontal distance d c The above formula optimization problem can be solved using mature numerical optimization methods (such as the interior point method). The value of σ in the formula is related to the "thickness" of the instantaneous frequency curve in p(t,f), therefore, the value of σ in the formula is... (That is, the half-peak width of the standard Gaussian function is Δ). In the initial iteration, the time-frequency energy concentration is low, so σ should be a large value. In subsequent iterations, as the time-frequency energy concentration increases, the value of σ can be gradually decreased to improve the accuracy of parameter estimation.

[0074] like Figure 6 As shown, Figure 6 A comparison chart of Doppler parameter estimation results, from Figure 6 It can be seen that the Doppler parameters estimated by the method of the present invention in the 0th iteration STFT (i.e., before iteration) are closer to the true values ​​of each parameter. However, since the instantaneous frequency curve changes abruptly due to noise at 25 to 30 seconds, the parameter values ​​(especially the positive and negative distances) estimated by the curve fitting method in the prior art deviate significantly from the true values.

[0075] S40. Calculate the time-frequency energy concentration based on the STFT time-frequency spectrum diagram.

[0076] Time-frequency energy concentration reflects the distribution of a sound signal in time and frequency. Regions with high time-frequency energy concentration indicate concentrated sound signal energy, while those with low concentration indicate dispersed signal energy. By calculating time-frequency energy concentration, we can better understand the characteristics of sound signals and thus more accurately estimate the velocity of moving sound sources.

[0077] The formula for calculating the time-frequency energy concentration based on the STFT time-frequency spectrum is as follows:

[0078]

[0079] In the formula, H α(P) represents the Renyi entropy of P(t,f), which is used to measure the time-frequency energy concentration of P(t,f). α represents the order of the Renyi entropy, and ∫∫P(u,v)dudv represents the total energy of P(t,f) calculated by integration.

[0080] The time-frequency energy concentration is calculated based on the above formula (Renyi entropy). The smaller the value of the formula, the higher the time-frequency energy concentration. In the formula, α > 0 represents the order of Renyi entropy. When α = 3, the above formula can well measure the time-frequency distribution of most signals and has the best stability.

[0081] like Figure 4 As shown, from Figure 4 It can be seen that, Figure 4 The time-frequency energy concentration at this point is 3.29.

[0082] S50. Iteratively process the relevant parameters of Doppler based on the DopplerCT time-frequency analysis algorithm.

[0083] When estimating Doppler correlation parameters, iterative calculations are required, and each iteration improves the accuracy of the estimation.

[0084] In one embodiment, step S50 specifically includes the following steps: S501-S504.

[0085] S501. The DopplerCT time-frequency diagram is calculated based on the relevant Doppler parameters.

[0086] The formula for calculating the DopplerCT time-frequency diagram based on relevant Doppler parameters is as follows:

[0087]

[0088]

[0089] In the formula, DopplerCT(t,f,p) represents the DopplerCT time-frequency diagram calculated under parameter p, where τ represents the integral of t, y represents the received line spectrum signal, w represents the window function, c represents the speed of sound, and parameter p = [f0,c,v,τ]. c ,d c [i.e., line spectrum frequency f0, sound speed c, velocity v, and positive transverse time τ] c and the vertical and horizontal distance d c .

[0090] Please refer to Figure 5 ,Will Figure 5 and Figure 4 The comparison shows that, Figure 5 The resolving power of the midline spectrum has been significantly improved.

[0091] S502. Calculate the updated Doppler parameters based on the DopplerCT time-frequency diagram.

[0092] This step calculates the relevant Doppler parameters during the iteration process based on the DopplerCT time-frequency plot, which can estimate the velocity of the moving sound source more accurately compared to step S30. Of course, the calculation formula can be the same as that in step S30, but P(t,f) in the formula should be taken as DopplerCT(t,f,p), that is, the DopplerCT time-frequency plot.

[0093] S503. Calculate the time-frequency energy concentration based on the DopplerCT time-frequency diagram.

[0094] Since the time-frequency energy concentration is used as one of the conditions for determining whether to stop the iteration, it needs to be calculated in the steps of the iteration process.

[0095] In this embodiment, the calculation process of this step is the same as that of step S40, but it is necessary to change STFT(t,f) to the DopplerCT time-frequency diagram obtained in step S501, i.e., DopplerCT(t,f,p).

[0096] S504. Determine whether the iteration should terminate based on the time-frequency energy concentration.

[0097] The iteration stops when the difference in time-frequency energy concentration between two consecutive DopplerCT time-frequency maps is within a set range.

[0098] S60. When the iteration stops, output the final Doppler parameters.

[0099] Two conditions must be met for iteration to stop. The first condition is the set number of iterations. After reaching a certain number of iterations, the iteration terminates to avoid looping. Experiments have shown that convergence is generally achieved in 3 to 5 iterations, so a maximum of 5 iterations can be set. The second condition is the time-frequency energy concentration degree mentioned above. The time-frequency energy concentration degree is set with the condition shown in the following formula. If the following formula is true, the iteration stops. The formula is as follows:

[0100] |H k -H k-1 |<δ, where δ represents the difference in time-frequency energy concentration between two adjacent iterations of the DopplerCT time-frequency map, which is generally taken as 0.01 to 0.1. In this embodiment, it is set to 0.01. If the difference in time-frequency energy concentration between two iterations of the DopplerCT time-frequency map is less than 0.01, the iteration is stopped.

[0101] like Figure 5As shown, based on the time-frequency energy concentration, the time-frequency energy concentrations for the four iterations are 1.763, 1.638, 1.622, and 1.620, respectively. In the fourth iteration, the difference in time-frequency energy concentration δ between the DopplerCT time-frequency maps obtained from adjacent iterations is less than 0.01, indicating iteration convergence. After four iterations, the output Doppler parameters are: estimated line spectrum frequency f0, velocity v, and positive transverse time τ. c and the vertical and horizontal distance d c like Figure 6 As shown, after multiple iterations, the estimated Doppler parameters are closer to their true values.

[0102] In summary, even when the instantaneous frequency curve is difficult to extract accurately due to noise (abrupt changes occur), this invention can still ensure robust iteration until convergence, achieving higher time-frequency resolution than the Short Time Fourier Transform (STFT), and obtaining parameter estimation results that are closer to the true values ​​than curve fitting methods.

[0103] Figure 2 This is a schematic block diagram of a robust device for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift, provided in an embodiment of the present invention. Corresponding to the above-described robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift, this embodiment of the present invention also provides a robust device 100 for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift.

[0104] like Figure 2 As shown, a robust device 100 for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift includes an initialization unit 110, an STFT time-spectrum calculation unit 120, a preliminary Doppler parameter estimation unit 130, a time-frequency energy concentration calculation unit 140, an iterative processing unit 150, and a final Doppler parameter output unit 160. The initialization unit 110 initializes the calculation parameters, including the window function length, Fourier transform length, and overlap length. The STFT time-spectrum calculation unit 120 performs a short-time Fourier transform based on the calculation parameters to obtain the STFT time-spectrum of the received signal. The preliminary Doppler parameter estimation unit 130 calculates the Doppler-related parameters before iteration based on the STFT time-spectrum, including line spectrum frequency, velocity, transverse time, and transverse distance. The time-frequency energy concentration calculation unit 140 calculates the time-frequency energy concentration based on the STFT time-spectrum. The iterative processing unit 150 is used to iteratively process the Doppler-related parameters according to the DopplerCT time-frequency analysis algorithm. The final Doppler parameter output unit 160 is used to output the final Doppler-related parameters after the iteration stops.

[0105] In one embodiment, the initialization unit 110 includes a first setting module and a second setting module. The first setting module is used to set the calculation parameters required for calculating the STFT time-frequency plot. The second setting module is used to set the calculation parameters required for calculating the DopplerCT time-frequency plot.

[0106] In one embodiment, the iterative processing unit 150 includes a first calculation module, a second calculation module, a third calculation module, and a judgment module. The first calculation module is used to calculate the DopplerCT time-frequency map based on Doppler-related parameters. The second calculation module is used to calculate the iteratively updated Doppler-related parameters based on the DopplerCT time-frequency map. The third calculation module is used to calculate the time-frequency energy concentration based on the DopplerCT time-frequency map. The judgment module is used to determine whether the iteration should terminate based on the time-frequency energy concentration.

[0107] The robust method described above for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift can be implemented as a computer program, which can be used in applications such as... Figure 3 It runs on the computer device shown.

[0108] Please see Figure 3 , Figure 3 This is a schematic block diagram of a computer device provided in an embodiment of this application. The computer device 700 can be a server, wherein the server can be a standalone server or a server cluster composed of multiple servers.

[0109] like Figure 3 As shown, the computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the robust method steps for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift as described above.

[0110] The computer device 700 can be a terminal or a server. The computer device 700 includes a processor 720, a memory, and a network interface 750 connected via a system bus 710, wherein the memory may include a non-volatile storage medium 730 and internal memory 740.

[0111] The non-volatile storage medium 730 can store an operating system 731 and a computer program 732. When the computer program 732 is executed, it enables the processor 720 to execute any robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift.

[0112] The processor 720 provides computing and control capabilities to support the operation of the entire computer device 700.

[0113] The internal memory 740 provides an environment for the operation of the computer program 732 in the non-volatile storage medium 730. When the computer program 732 is executed by the processor 720, the processor 720 can execute any robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift.

[0114] This network interface 750 is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device 700 to which the present application is applied. The specific computer device 700 may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements. The processor 720 is used to run program code stored in memory to implement the following steps:

[0115] A robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift includes:

[0116] Initialize the calculation parameters, which include the window function length, Fourier transform length, and overlap length;

[0117] Perform a short-time Fourier transform based on the calculated parameters to obtain the STFT time spectrum of the received signal;

[0118] The Doppler parameters at the non-iteration time are calculated based on the STFT time spectrum diagram. The Doppler parameters include line spectrum frequency, velocity, positive transverse time, and positive transverse distance.

[0119] Calculate the time-frequency energy concentration based on the STFT time-frequency spectrum diagram;

[0120] The relevant parameters of Doppler are iteratively processed according to the DopplerCT time-frequency analysis algorithm;

[0121] Once the iteration stops, the final Doppler parameters are output.

[0122] In one embodiment, the initialization calculation parameters, including the window function length, Fourier transform length, and overlap length, include:

[0123] Set the calculation parameters required for calculating the STFT time-frequency diagram;

[0124] Set the calculation parameters required to calculate the DopplerCT time-frequency plot.

[0125] In one embodiment, the window function length and Fourier transform length required for the DopplerCT time-frequency plot are both greater than those required for the STFT time-frequency plot.

[0126] In one embodiment, the Doppler-related parameters for the non-iterative period are calculated based on the STFT time-spectrum diagram. These Doppler-related parameters include line spectrum frequency, velocity, transverse time, and transverse distance. The formula for calculating the Doppler-related parameters for the non-iterative period is as follows:

[0127]

[0128] In the formula, P(t,f) represents the time-frequency graph, t represents time, f represents frequency, σ represents the thickness of the instantaneous frequency curve, and f u f is the upper bound of the frequency range. d This is the lower bound of the frequency range.

[0129] In one embodiment, the formula for calculating the time-frequency energy concentration degree in the step of calculating the time-frequency energy concentration degree based on the STFT time-frequency spectrum is as follows:

[0130]

[0131] In the formula, H α (P) represents the Renyi entropy of P(t,f), which is used to measure the time-frequency energy concentration of P(t,f). α represents the order of the Renyi entropy, and ∫∫P(u,v)dudv represents the total energy of P(t,f) calculated by integration.

[0132] In one embodiment, the iterative processing of Doppler-related parameters based on the DopplerCT time-frequency analysis algorithm includes:

[0133] The DopplerCT time-frequency diagram was calculated based on relevant Doppler parameters.

[0134] The Doppler parameters after iterative updates were calculated based on the DopplerCT time-frequency plot.

[0135] Calculate the time-frequency energy concentration based on the DopplerCT time-frequency plot;

[0136] The iteration should be terminated based on the time-frequency energy concentration.

[0137] In one embodiment, the DopplerCT time-frequency map is calculated based on relevant Doppler parameters. The formula for calculating the DopplerCT time-frequency map is as follows:

[0138]

[0139]

[0140] In the formula, DopplerCT(t,f,p) represents the DopplerCT time-frequency diagram calculated under parameter p, where τ represents the integral of t, y represents the received line spectrum signal, w represents the window function, c represents the speed of sound, and parameter p = [f0,c,v,τ]. c ,d c [i.e., line spectrum frequency f0, sound speed c, velocity v, and positive transverse time τ] c and the vertical and horizontal distance d c .

[0141] It should be understood that in the embodiments of this application, the processor 720 may be a central processing unit (CPU), or it may be other general-purpose processors, 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, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0142] Those skilled in the art will understand that Figure 3 The structure of the computer device 700 shown does not constitute a limitation on the computer device 700, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0143] In another embodiment of the present invention, a computer-readable storage medium is provided. This computer-readable storage medium may be a non-volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein when executed by a processor, the computer program implements the robust method for estimating the velocity of a moving sound source based on line spectrum Doppler frequency shift disclosed in the embodiments of the present invention.

[0144] Those skilled in the art will readily understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention.

[0145] In the embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Units with the same function may be grouped into one unit. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, or it may be an electrical, mechanical, or other form of connection.

[0146] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0147] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0148] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks.

[0149] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A robust method for estimating the velocity of a moving sound source based on line spectral Doppler shift estimation, characterized in that, The method comprises the following steps: initializing calculation parameters, wherein the calculation parameters comprise window function length, Fourier transform length and overlap length; performing short-time Fourier transform according to the calculation parameters to obtain an STFT time-frequency spectrum of the received signal; calculating Doppler-related parameters without iteration according to the STFT time-frequency spectrum, wherein the Doppler-related parameters comprise line spectrum frequency, velocity, time of crossing and distance of crossing; calculating time-frequency energy concentration according to the STFT time-frequency spectrum; iteratively processing the Doppler-related parameters according to a Doppler CT time-frequency analysis algorithm; outputting final Doppler-related parameters when the iteration stopping requirement is reached.

2. The method of claim 1, wherein, The method of initializing calculation parameters, wherein the calculation parameters comprise window function length, Fourier transform length and overlap length, comprises the following steps: setting calculation parameters required for calculating the STFT time-frequency spectrum; setting calculation parameters required for calculating the Doppler CT time-frequency spectrum.

3. The method of claim 2, wherein, The window function length and the Fourier transform length required for the Doppler CT time-frequency spectrum are both greater than the window function length and the Fourier transform length required for the STFT time-frequency spectrum.

4. The method of claim 1, wherein, The formula for calculating the Doppler-related parameters without iteration according to the STFT time-frequency spectrum, wherein the Doppler-related parameters comprise line spectrum frequency, velocity, time of crossing and distance of crossing, is as follows: ； where P denotes a set of variables, k(t,p) denotes a Doppler profile, denotes a time-frequency plot, denotes time, denotes frequency, denotes the thickness of the instantaneous frequency profile, is the upper bound of the frequency interval, is the lower bound of the frequency interval.

5. The method of claim 1, wherein, The formula for calculating the time-frequency energy concentration according to the STFT time-frequency spectrum is as follows: ； In the formula, denotes the Renyi entropy of The Renyi entropy is used to measure the time-frequency energy aggregation degree of denotes the Renyi entropy order, denotes the total energy of by integral calculation .​ 6. The method of claim 1, wherein, The method of iteratively processing the Doppler-related parameters according to the Doppler CT time-frequency analysis algorithm comprises the following steps: calculating a Doppler CT time-frequency spectrum according to the Doppler-related parameters; calculating iteratively updated Doppler-related parameters according to the Doppler CT time-frequency spectrum; calculating time-frequency energy concentration according to the Doppler CT time-frequency spectrum; judging whether the iteration is terminated according to the time-frequency energy concentration.

7. The method of claim 6, wherein, The formula for calculating the Doppler CT time-frequency spectrum according to the Doppler-related parameters is as follows: ； where f denotes the short-time Fourier transform frequency, denotes the parameter The Doppler CT time-frequency plot calculated under the condition denotes the quantity being integrated, denotes the received line spectrum signal, denotes the window function, denotes the sound velocity, the parameter is the line spectrum frequency , the sound velocity , the velocity , the time of the cross range and the cross range distance .

8. A device for robustly estimating the velocity of a moving sound source based on line spectral Doppler shift estimation, characterized in that The method comprises an initialization unit, an STFT time-frequency spectrum calculation unit, a preliminary Doppler parameter estimation unit, a time-frequency energy concentration calculation unit, an iteration processing unit and a final Doppler parameter output unit. The initialization unit is configured to initialize calculation parameters, wherein the calculation parameters comprise window function length, Fourier transform length and overlap length. The STFT time-frequency spectrum calculation unit is configured to perform short-time Fourier transform according to the calculation parameters to obtain an STFT time-frequency spectrum of the received signal. The preliminary Doppler parameter estimation unit is configured to calculate Doppler-related parameters without iteration according to the STFT time-frequency spectrum, wherein the Doppler-related parameters comprise line spectrum frequency, velocity, time of crossing and distance of crossing. The time-frequency energy concentration calculation unit is configured to calculate time-frequency energy concentration according to the STFT time-frequency spectrum. The iteration processing unit is configured to iteratively process the Doppler-related parameters according to a Doppler CT time-frequency analysis algorithm. The final Doppler parameter output unit is configured to output final Doppler-related parameters when the iteration stopping requirement is reached.

9. A computer device, comprising: The computer program product comprises a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and the processor implements the method for estimating the speed of a moving sound source based on a line spectrum Doppler frequency shift according to any one of claims 1-7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, and the computer program comprises program instructions, and the program instructions are executed by a processor to enable the processor to execute the method for estimating the speed of a moving sound source based on a line spectrum Doppler frequency shift according to any one of claims 1-7.

Images