A method and system for separating and locating signals from swarms of bats

Abstract

This invention discloses a method and system for separating and locating swarming bat signals, primarily addressing the problem of effective separation and three-dimensional localization when swarming bat signals are aliased. The implementation includes: receiving and preprocessing swarming bat signals using a microphone array; estimating the number of bats through energy distribution in the time-frequency domain; estimating the angle of arrival (AHA) of the swarming bat signals using an angle measurement algorithm based on iterative projection to eliminate interference signals; separating the signals using a frequency-domain beamforming algorithm based on the estimated AHA; matching the separated signals using a signal correlation method that cyclically removes the maximum value, and estimating the time difference of arrival (TDOA) of the same bat signal reaching each microphone array based on the cross-correlation characteristics between arrays; and performing three-dimensional localization of the swarming bats by combining the AHA and TDOA. This invention can effectively separate aliased signals from swarming bats and achieve three-dimensional localization of them, and can be used in the design of biomimetic radar.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of radar signal processing and bat acoustics, specifically relating to a method and system for separating and locating signals from swarms of flying bats, which can be used in bat behavior and biomimetic radar. Background Technology

[0002] Bats are animals that navigate and forage in complete darkness using echolocation. They accurately perceive their surroundings and the location of their prey by emitting ultrasonic signals and receiving reflected echoes from objects in their environment. In natural settings where bats fly in swarms, the sound waves emitted by multiple individuals simultaneously superimpose in space, creating a complex and highly aliased sound field. The fact that individual bats can fly undisturbed in this complex and disturbing environment demonstrates their remarkable acoustic perception capabilities, which are worthy of in-depth study and emulation. Researching the flight patterns and signal characteristics of bat swarms is not only of significant value to ecology and animal behavior but also provides inspiration and theoretical basis for the development of acoustic localization, multi-target tracking, and radar anti-jamming technologies. Due to the aliasing of signals from swarming bats, traditional signal analysis methods face significant challenges in signal separation and localization.

[0003] Current research methods for separating the acoustic signals of swarming bats mainly rely on the natural separation characteristics of the signals in the time domain. Specifically, researchers typically use fixed microphone arrays or backpack microphone devices to record the echolocation signals of bats. However, regardless of whether a fixed microphone array or a backpack microphone device is used, it is difficult to avoid the aliasing caused by receiving signals from multiple individuals simultaneously. In their 2008 paper "Flying in silence: Echolocating bats cease vocalizing to avoid sonar jamming," Chiu et al. used 16 fixed microphones arranged in a U-shape on the wall and 3 microphones placed on the ground to record the acoustic signals of bats flying in pairs. In their 2021 paper "Silence and reduced echolocation during flight areas associated with social behaviors in male hoary bats (Lasiurus cinereus)," and in their 2022 paper "Echo reception in group flight by Japanese horseshoe bats, Rhinolophus ferrumequinum nippon," Corcoran et al. used miniature backpack microphones mounted on individual bats to directly collect the acoustic signals of flying individuals. The above studies rely solely on the natural temporal offset of bat signals for separation. However, this approach fails to achieve effective separation when there is severe time-frequency aliasing in the acoustic signals of swarming bats, making it difficult to efficiently extract and accurately analyze swarming bat signals.

[0004] In terms of localization, current research typically uses a small number of microphones and relies on the Time Difference of Arrival (TDOA) between microphones for 3D localization. Simone, in 2016... In their paper "No evidence for spectral jammingavoidance in echolocation behavior of foraging pipistrelle bats," researchers used a horizontal T-shaped planar microphone array and employed the TDOA method for target localization. In their 2023 paper "Calibrated microphone array recordings reveal that a gleaning bat emits low-intensity echolocation calls even in open-space habitat," Léna de Framond et al. used a planar symmetrical star-shaped microphone array and similarly utilized the TDOA method for target localization. These studies, by combining microphone arrays with the TDOA method, can achieve 3D localization of a single bat. However, in scenarios involving swarms of bats, severe signal aliasing increases the estimation error of TDOA, making it difficult to achieve multi-bat localization under swarm conditions. Summary of the Invention

[0005] To address the shortcomings of the existing technologies, this invention proposes a method and system for separating and locating signals from swarms of bats, aiming to effectively separate overlapping signals from multiple bats and achieve three-dimensional localization of bats, thus providing a new technical means for bat behavioral research.

[0006] To achieve the above objectives, the technical solution adopted by the present invention includes the following:

[0007] 1. The method for separating and locating signals of swarming bats provided by the present invention is characterized by comprising:

[0008] Receive signals from swarms of bats using at least two microphone arrays. m,q (t), after amplitude and phase consistency correction, bandpass filtering and frame segmentation, the l-th frame aliasing signal is obtained. Where l = 1, 2, ..., L are the frame numbers, L is the total number of frames, m = 1, 2, ..., M are the array numbers, M is the number of arrays, q = 1, 2, ..., Q are the microphone numbers, and Q is the total number of microphones for each array;

[0009] According to the aliasing signal of the lth frame The energy distribution in the time-frequency domain is used to estimate the number of bats D in that frame. (l) ;

[0010] Based on the number of bats D (l) Iterative projection eliminates interference signals, for the l-th frame signal of each array. Angle measurement was performed to obtain the bat's arrival angle AOA, where

[0011] Based on the bat's angle of arrival (AOA), frequency domain beamforming is used to analyze the signal. Separate the signals to obtain the signals of each bat in the l-th frame. Where d = 1, 2, ..., D (l) Number the bat signals;

[0012] A signal correlation method that cyclically removes the maximum value is used to progressively correlate the bat signals of each array. Matching is performed, and based on the cross-correlation characteristics of the same bat signal between arrays, the time difference of arrival (TDOA) between arrays for the same bat signal is estimated.

[0013] The three-dimensional position of each bat was calculated by combining the bat's angle of arrival (AOA) and the time difference of arrival (TDOA) between arrays.

[0014] Furthermore, the statement based on the number of bats D ( l Iterative projection eliminates interference signals, for the l-th frame signal of each array. Angle measurement was performed to obtain the bat's angle of arrival (AOA), including:

[0015] a) For the signal The MUSIC algorithm was used to measure the angle, and the number of spectral peaks P and the corresponding angle of arrival AOA were obtained.

[0016] b) Determine whether the number of spectral peaks P is less than the target number D. (l) If yes, proceed to step c); otherwise, select the peak with the strongest energy D from the P peaks. l The angle of arrival corresponding to each spectral peak is used as the bat's angle of arrival (AOA).

[0017] c) Continue estimating the signal by eliminating interference signals through iterative projection. The concealed D (l) -P bat signal AOAs are obtained and merged with the P AOAs obtained in a) to get D. (l) A bat reached the angle AOA.

[0018] Furthermore, the method of eliminating interference signals through iterative projection is used to continue estimating the signal. The concealed D (l) -P bat signal AOAs, including:

[0019] i) Initialize the number of known spectral peaks to 0;

[0020] ii) Based on the signal An orthogonal projection matrix is ​​constructed based on the angle of arrival corresponding to the strongest spectral peak in the medium energy range, and this matrix is ​​then used to analyze the signal. Projecting the covariance matrix onto the given covariance matrix yields the updated covariance matrix.

[0021] iii) Apply the MUSIC algorithm to the updated covariance matrix to measure the angle, obtain the current number of spectral peaks and the corresponding angle of arrival, and add the newly obtained number of spectral peaks to the known number of spectral peaks;

[0022] iv) Determine whether the number of known spectral peaks is less than the number of targets being masked, D. (l) -P: If yes, execute v); otherwise, select the D with the strongest energy. (l) -P spectral peaks corresponding to the angles of arrival, which are used as the angles of arrival of the masked bats;

[0023] v) Extract the angle of arrival corresponding to the strongest spectral peak in iii) to construct an orthogonal projection matrix, update the covariance matrix, and then return to iii).

[0024] Furthermore, the statement based on the signal An orthogonal projection matrix is ​​constructed based on the angle of arrival corresponding to the strongest spectral peak in the medium energy range, and this matrix is ​​then used to analyze the signal. Projecting the covariance matrix onto the given covariance matrix yields the updated covariance matrix, which includes:

[0025] Extract signal The angle of arrival (AOA) corresponding to the strongest spectral peak is used to construct the steering vector a of the signal.

[0026] Construct the projection matrix P using the guiding vector a:

[0027]

[0028] Where I is the identity matrix, (·) H Indicates conjugate transpose;

[0029] Using the projection matrix P to represent the signal Projecting the covariance matrix R onto the given covariance matrix yields the updated covariance matrix R0. n =PRP H .

[0030] 2. Based on the same inventive concept, this invention also proposes a system for separating and locating signals of swarming bats, comprising:

[0031] The signal receiving and preprocessing module is used to receive the signals of swarming bats using at least two microphone arrays, and to perform amplitude and phase consistency correction, bandpass filtering and framing processing on them to generate normalized aliased signal frames.

[0032] The bat count estimation module is used to perform time-frequency analysis on the normalized aliased signal frame and estimate the number of bats contained in the signal frame.

[0033] The Angle of Arrival (AOA) estimation module is used to iteratively project and eliminate interference signals based on the estimated number of bats to obtain the AOA of each bat signal.

[0034] The signal separation module is used to separate the signal of each bat based on the angle of arrival (AOA) of each bat signal using a frequency domain beamforming method.

[0035] The Time Difference of Arrival (TDOA) estimation module is used to match the bat signals separated from each array and estimate the TDOA of the matched signals to each array based on the cross-correlation characteristics of the same bat signal between arrays.

[0036] A 3D localization module is used to estimate the 3D spatial position of each bat by combining the angle of arrival (AOA) and time difference of arrival (TDOA).

[0037] Compared with the prior art, the present invention has the following advantages:

[0038] Firstly, this invention employs an iterative projection method to eliminate interference signals in angle measurement. Firstly, the angle of strong interference signals is estimated using the MUSIC algorithm for multiple signal classification. Based on this angle, an orthogonal projection matrix is ​​constructed. This matrix is ​​then used to project the signal into a subspace orthogonal to the strong signal, thereby eliminating the interference of the strong signal and realizing angle measurement of weak signals under the masking of strong signals. Furthermore, all strong signals are gradually eliminated through iteration, improving the angle measurement performance of weak targets in environments with interleaved strong and weak signals.

[0039] Secondly, this invention employs a frequency domain beamforming method to separate time-frequency aliased bat signals. Based on the angle of arrival (AOA), an optimal beamforming weight vector is constructed to form a spatial filter pointing to each bat signal. The signal is then weighted and filtered in the frequency domain, thereby separating the time-frequency aliased broadband bat signals and improving the ability to separate complex aliased signals in multi-bat environments.

[0040] Third, this invention employs a signal correlation method that cyclically eliminates the maximum value. Based on the normalized correlation coefficient of each array's bat signals, a correlation coefficient matrix is ​​constructed. The maximum value is found and the corresponding row and column are set to zero. While achieving strong signal matching, the influence of cross-correlation between strong and weak signals on the autocorrelation of weak signals is reduced. Furthermore, through iteration, the correlation matching of all bat signals is gradually achieved. Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating the implementation of the method for separating and locating signals of swarming bats provided in Embodiment 1 of the present invention.

[0042] Figure 2 This is a block diagram of the swarm bat signal separation and positioning system provided in Embodiment 2 of the present invention.

[0043] Figure 3This is a simulation diagram of the time frequency and waveform of the signal received by two bats during flight using a microphone array, according to the present invention.

[0044] Figure 4 This is a simulation diagram of the AOA estimation results of the present invention for time-frequency aliasing and strong-weak interleaving bat signals.

[0045] Figure 5 This is a simulation diagram showing the results of the present invention in separating time-frequency mixed and strong-weak interleaved bat signals.

[0046] Figure 6 This is a simulation diagram of the reconstructed flight trajectories of two bats and compared with their actual flight trajectories. Detailed Implementation

[0047] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0048] Example 1: Method for Separating and Locating Signals of Swarming Bats

[0049] Reference Figure 1 The implementation steps of this example include the following:

[0050] Step 1: Receiving and preprocessing signals from swarms of bats.

[0051] With the rapid development of radar technology, traditional radar systems face numerous challenges in multi-target detection, noise suppression, and adaptability to complex environments. Against this backdrop, bionics offers new research directions for radar technology innovation, particularly the echolocation system derived from nature, particularly bats. Bats are animals that navigate and forage using echolocation in complete darkness. They accurately perceive their surroundings and the location of prey by emitting ultrasonic signals and receiving reflected echoes from objects in their environment. In natural scenarios where bats fly in swarms, the sound waves emitted simultaneously by multiple individuals superimpose in space, forming a complex and highly aliased sound field. The fact that individual bats can fly undisturbed in this complex and disruptive environment demonstrates their remarkable acoustic perception capabilities, which warrant in-depth study and emulation. This interdisciplinary research is not only of significant value to ecology and animal behavior but also provides inspiration and theoretical basis for the development of acoustic localization, multi-target tracking, and radar anti-jamming technologies. Existing bat data acquisition and analysis methods still cannot completely overcome the key technical bottlenecks in echolocation research on swarm bats, mainly manifested in the severe aliasing of echolocation signals from groups of bats, making effective separation and accurate localization difficult. The present invention proposes a method and system for separating and locating signals of swarming bats, which aims to effectively separate overlapping signals of multiple bats and achieve three-dimensional localization of bats.

[0052] To analyze and accurately locate the signals of swarms of bats, it is first necessary to collect and preprocess the high-frequency ultrasonic signals they emit. The specific implementation includes the following:

[0053] (1.1) Receive signals from swarms of bats using at least two microphone arrays:

[0054] When using microphone arrays to collect ultrasonic waves from swarms of bats in space, planar or stereo arrays are typically used to obtain the target's azimuth and elevation angles. To achieve target localization, at least two microphone arrays are required; increasing the number of arrays allows for the fusion of more spatial information, further improving positioning accuracy.

[0055] Because bat ultrasonic signals have a high frequency, typically above 20kHz to 120kHz, a high-sensitivity ultrasonic microphone is required to ensure the array can accurately capture the signal characteristics of bats. Its frequency response range should cover the bat ultrasonic frequency band, and a sampling frequency much higher than the Nyquist frequency should be used to ensure signal integrity and reduce sampling errors. The resulting multi-channel received signal is mathematically expressed as follows:

[0056] x m (t)=[x m，1 (t),...,x m，q (t),…,x m，Q (t)] T

[0057] In the formula, m = 1, 2, ..., M are the array numbers, M is the number of microphone arrays, q = 1, 2, ..., Q are the microphone numbers, and Q is the number of microphones per array.

[0058] In this embodiment, the bat's main frequency is approximately 72kHz, and the array's sampling rate is set to 250kHz to meet the requirements for high-frequency signal acquisition. Two microphone arrays are used to locate the target.

[0059] (1.2) Receive signal x from each channel of the array m,q (t) Perform amplitude-phase consistency correction:

[0060] In practical systems, the amplitude and phase responses of the microphones, front-end amplifier circuits, and analog-to-digital conversion channels in each channel of the array differ. These differences significantly affect the performance of subsequent array processing algorithms such as spatial spectrum estimation and beamforming. Therefore, before performing array signal processing, it is essential to analyze the received array signal x. m,q The amplitude and phase of (t) are consistent to ensure that the amplitude and phase characteristics of the received signals from all channels are consistent. The specific implementation includes the following:

[0061] (1.2.1) Estimate the correction coefficients for each channel

[0062] To correct for amplitude and phase differences between different channels in the microphone array, a standard narrowband signal source with a frequency of f0 was used to illuminate the microphone array in a noise-free experimental environment. In this embodiment, f0 = 72kHz was selected as the reference frequency for the calibration signal. After each channel of the microphone array synchronously acquired this narrowband signal, the received signal for each channel was obtained. A Fast Fourier Transform (FFT) was performed on the received signal of each channel, and the complex spectrum value Z at frequency point f0 was extracted. q (f0), taking the first channel as the reference, its complex spectrum value is denoted as Z1(f0), and the correction coefficient G for each channel is calculated. q :

[0063]

[0064] (1.2.2) Correct the received signal x of each channel m,q (t)

[0065] Received signal x for each channel of the array m,q (t) Perform a Fast Fourier Transform (FFT) to obtain the frequency domain signal X. m,q (f):

[0066] X m,q (f)=FFT{x m,q (t)}

[0067] For frequency domain signal X m,q (f) Perform amplitude and phase consistency correction to obtain the corrected frequency domain signal X′. m,q (f):

[0068] X′ m,q (f)=G q ·X m,q (f)

[0069] Where m = 1, 2 are array numbers, G q Here are the correction coefficients for each channel, q = 1, 2, ..., Q, where Q is the microphone number and Q is the number of microphones in each array.

[0070] X′ is obtained by inverse fast Fourier transform (IFFT). m,q (f) Convert back to the time domain to obtain the corrected time-domain signal x′ m,q (t):

[0071] x′ m,q (t)=IFFT{X′ m,q (f)}

[0072] (1.3) For the corrected signal x′ m,q (t) performs bandpass filtering

[0073] To preserve the signal within the bat echolocation frequency band and filter out interference and noise outside the echolocation signal band, the corrected signal x′ needs to be... m,q (t) is used for bandpass filtering.

[0074] This embodiment selects the greater horseshoe bat, whose echolocation signal frequency range is 55kHz-72kHz. Based on this frequency range, a finite impulse response (FIR) bandpass filter is designed with the following parameters: passband cutoff frequency 50kHz-75kHz, stopband cutoff frequencies 40kHz and 80kHz, sideband attenuation 60dB, and passband ripple 0.1dB. This filter is then used to filter the received signals from each channel, resulting in the filtered signal x′. m ′ ,q (t).

[0075] (1.4) For the filtered signal x′ m ′ ,q (t) performs frame segmentation processing

[0076] Because the received signal lasts for a long time, contains multiple echolocation pulses from each bat, and the bats are always in motion, causing the pulse emission position to change continuously, it is necessary to perform frame-segmentation processing on the signal.

[0077] Bat echolocation signals are pulsed, and their duration and pulse interval can be used to guide framing. For the greater hooves bat in this embodiment, the duration of its echolocation signal is approximately 10 ms, and the pulse interval is approximately 25 ms. Based on these two parameters, a frame is formed every 35 ms, and the filtered signal x′ is... m ′ ,q (t) Perform frame segmentation processing to generate frame signals. Where l = 1, 2, ..., L are the frame numbers, and L is the total number of frames.

[0078] The l-th frame signal after amplitude and phase consistency correction and bandpass filtering is written in matrix form.

[0079]

[0080] Where m = 1, 2 are array numbers, q = 1, 2, ..., Q are microphone numbers, and Q is the number of microphones in each array.

[0081] Step 2: Estimate the number of bats.

[0082] Before performing target signal separation and localization on bats, it is necessary to first determine the number of individual bats in each frame of the signal. In each frame of the signal to be processed... In the middle, the number of individual bats D (l) It can be estimated using time-frequency graphs. The specific implementation includes the following:

[0083] (2.1) The signal to be processed is obtained through time-frequency analysis methods, such as Short Time Fourier Transform (STFT), Wigner-Ville distribution transform, and wavelet transform. The time-frequency distribution is obtained, and a time-frequency graph is generated.

[0084] This embodiment employs, but is not limited to, STFT transformation to calculate the time-frequency distribution.

[0085]

[0086] Where w(τ-t) is a window function, such as a Gaussian window, a Hanning window, and a Hamming window. This embodiment uses a Hamming window.

[0087] (2.2) In The number of independent energy concentration regions in the generated time-frequency graph is the bat population D. (l) .

[0088] Due to the differences in frequency, time domain, and energy distribution of different bat signals, they appear as multiple independent energy concentration regions in the time-frequency diagram. Therefore, by counting the number of these energy concentration regions, the number D of bats simultaneously vocalizing can be determined. (l) The estimate.

[0089] Step 3: Estimate the angle of arrival (AOA) of the bat signal.

[0090] This invention employs a multi-signal classification (MUSIC) method based on iterative projection to eliminate interference signals to estimate the l-th frame signal of each array. The angle of arrival (AHA) of each bat signal, including azimuth α and pitch θ, is projected into a signal subspace orthogonal to strong interference signals through multiple iterations. This eliminates the masking effect of strong interference on weak signals, allowing the AHA of weak bat targets to be accurately estimated. The specific implementation includes the following:

[0091] (3.1) Spatial spectrum obtained by the MUSIC method:

[0092] (3.1.1) Calculate the l-th frame signal of the m-th array. The covariance matrix R:

[0093]

[0094] Where Ε[·] represents the mathematical expectation operation, (·) H Indicates conjugate transpose;

[0095] (3.1.2) Perform eigenvalue decomposition on R to extract the eigenvector matrix E of the noise subspace. n :

[0096]

[0097] Among them, E s and E n Let Λ represent the eigenvector matrices of the signal subspace and the noise subspace, respectively. s and Λ n This is the corresponding eigenvalue matrix;

[0098] (3.1.3) Based on the noise subspace eigenvector matrix E n Orthogonal to the signal steering vector, the spatial spectrum search function P(α,θ) is constructed using this orthogonality:

[0099]

[0100] in, The guide vectors of the array, each guide vector corresponding to an angle of arrival, (x q ,y q ,z q ) represents the coordinates of the q-th microphone in the array with the center of the microphone array as the origin, and λ is the wavelength of the bat signal;

[0101] (3.1.4) Determine the spatial spectrum based on the spectral search function P(α,θ). That is, when the steering vector a(α,θ) points to the actual arrival direction of the bat signal, the spatial spectrum function corresponding to this angle will show an obvious peak. By traversing all possible angles in the entire search space, the complete spatial spectrum is obtained.

[0102] (3.2) Count the number P of spatial spectral peaks, and then calculate the number of peaks P and the number of peaks D. (l) Choose an appropriate angle measurement strategy based on the relationship between the two:

[0103] When the number of spectral peaks P is greater than or equal to the number of bats D (l) When selecting the peak with the strongest energy D from the P spectral peaks... (l) The angle of arrival corresponding to each spectral peak is used as the bat's angle of arrival (AOA).

[0104] When the number of spectral peaks P is less than the number of bats D ( l When this occurs, a masking phenomenon appears. Subsequent steps are then used to estimate the masked D. (l) -P bat signal arrival angles;

[0105] (3.3) Iterative projection eliminates interference signals and estimates the masked D. (l) -P bat signals AOA:

[0106] (3.3.1) Initialize the number of known spectral peaks to 0;

[0107] (3.3.2) The angle of arrival (α) of the strongest energy peak in the spatial spectrum obtained from (3.1.4) max ,θ max Construct the guiding vector a(α) max ,θ max ):

[0108]

[0109] Among them, (x q ,y q ,z q ) represents the coordinates of the q-th microphone in the array with the center of the microphone array as the origin, and λ is the wavelength of the bat signal;

[0110] (3.3.3) Based on the guiding vector a(α) max ,θ max We obtain the orthogonal projection matrix P that is orthogonal to it:

[0111]

[0112] Where I is the identity matrix of P×P;

[0113] (3.3.4) Project the signal covariance matrix R onto the orthogonal projection matrix P to obtain the updated covariance matrix R. n :

[0114] R n =PRP H ;

[0115] (3.3.5) Apply the MUSIC algorithm to the updated covariance matrix to measure the angle, obtain the current number of spectral peaks and the corresponding angle of arrival, and add the newly obtained number of spectral peaks to the known number of spectral peaks;

[0116] (3.3.6) Determine whether the number of known spectral peaks is less than the number of targets being masked, D. (l) -P:

[0117] If so, then execute (3.3.7); otherwise, select the D with the strongest energy. (l) -P spectral peaks corresponding to the angles of arrival, which are used as the angles of arrival of the masked bats;

[0118] (3.3.7) Extract the angle of arrival corresponding to the strongest spectral peak in (3.3.5) to construct an orthogonal projection matrix, update the covariance matrix, and then return to (3.3.5).

[0119] Step 4: Use a frequency domain beamforming algorithm to separate each bat signal from the aliased signal.

[0120] Once the angle of each bat signal is known, a spatial filter can be constructed pointing in the direction of each bat signal while suppressing signals from other directions, thus achieving the separation of each signal. This embodiment selects the frequency-domain linearly constrained minimum variance (LCMV) beamforming algorithm to separate aliased signals. The specific implementation includes the following:

[0121] (4.1) Calculate the frequency domain covariance matrix:

[0122] (4.1.1) The l-th frame signal of the m-th array The signal is transformed to the frequency domain using a Fast Fourier Transform (FFT) to obtain the frequency domain signal.

[0123]

[0124] frequency domain signal The data across all frequency points is represented as follows:

[0125]

[0126] Among them, f j Let J be the frequency at the j-th frequency point, where j = 1, 2, ..., J is the frequency point number, and J is the total number of frequency points.

[0127] (4.1.2) Calculate the frequency domain signal covariance matrix

[0128]

[0129] (4.2) Calculate the optimal beamforming weight vector:

[0130] (4.2.1) For the d-th bat signal of the m-th array, calculate the frequency f. j The corresponding frequency domain steering vector a(f) j ,α m,d ,θ m,d ):

[0131]

[0132] Where, α m,d θ is the azimuth angle of the d-th bat signal estimated by the m-th array. m,d Let m be the elevation angle of the d-th bat signal estimated by the m-th array, where m = 1, 2 are the array numbers, and d = 1, 2, ..., D. (l) The bat signal is numbered, D (l) For the number of bat signals, (x q ,y q ,z qLet λ be the coordinates of the q-th microphone in the array with the array center as the origin, where q = 1, 2, ..., Q is the microphone number, Q is the number of microphones in each array, and λ is the number of microphones in each array. j For frequency f j The corresponding wavelength;

[0133] (4.2.2) Calculate the frequency f of the interference signal j The corresponding frequency domain steering vector a(f) j ,α m,i ,θ m,i ):

[0134]

[0135] Where, α m,i Let θ be the azimuth angle of the i-th interference signal estimated by the m-th array. m,i Estimate the elevation angle of the i-th interference signal for the m-th array, where i = 1, 2, ..., I is the interference signal number and I is the total number of interference signals;

[0136] (4.2.3) The frequency domain steering vector a(f) in the direction of the desired suppression of interference signal is... j ,α m,i ,θ m,i The interference signal steering vector matrix A is formed. I :

[0137] A I =[a(f j ,α m,1 ,θ m,1 ),...,a(f j ,α m,i ,θ m,i ),...,a(f j ,α m,I ,θ m,I )];

[0138] (4.2.4) Based on the frequency f of the d-th bat signal in the m-th array j Frequency domain steering vector a(f) at the location j ,α m,d ,θ m,d ) and interference signal steering vector matrix A I , forming a frequency domain steering vector matrix

[0139]

[0140] (4.2.5) For the d-th bat signal of the m-th array, according to the frequency domain covariance matrix... and frequency domain steering vector matrix Calculate frequency fj The corresponding optimal weights for frequency domain LCMV beamforming

[0141]

[0142] in,(·) -1 To perform the inverse operation;

[0143] (4.3) Separate individual bat signals:

[0144] (4.3.1) Using weight vectors For the received signal Each frequency point is weighted separately to obtain the separated signal at each frequency point.

[0145]

[0146] (4.3.2) Based on the separated signal at each frequency point Obtain the frequency domain representation of the separated signal

[0147]

[0148] (4.3.3) Frequency domain representation of the d-th bat signal of the m-th array to be separated Perform an inverse fast Fourier transform to obtain its time-domain representation.

[0149]

[0150] Where m = 1, 2 are array numbers, and d = 1, 2, ..., D (l) Number the bat signals;

[0151] (4.4) Based on the arrival angle of all bat signals, calculate the optimal weight vector for each bat signal and perform beamforming to separate all bat signals in each frame of each array.

[0152] Step 5: Estimate the Time Difference of Arrival (TDOA) of each bat signal to each array.

[0153] To achieve three-dimensional localization of bat signals, in addition to the angle of arrival (AOA), the time difference of arrival (TDOA) of the signals arriving at each array also needs to be estimated. Since the signals received by the arrays originate from multiple individual bats, it is first necessary to identify and match signals from the same bat in different arrays, and then calculate their TDOA using the cross-correlation properties between the matched signals. The specific implementation includes the following:

[0154] (5.1) Matching signals from the same bat:

[0155] In multi-target scenarios, received signals often originate from multiple different bats. It is necessary to correlate the signals separated from each array to match signals originating from the same bat. Based on the correlation of signals from the same bat across arrays, a signal correlation method that iteratively eliminates the maximum value is adopted to progressively correlate the bat signals from each array. Perform matching. The specific implementation includes the following:

[0156] (5.1.1) Calculate the normalized cross-correlation coefficient c between the bat signals in the array. r,n :

[0157]

[0158] in, The r-th signal separated from array 1, where r = 1, 2, ..., D (l) The numbering of the signals separated from array 1. This refers to the nth signal separated from array 2, where n = 1, 2, ..., D. (l) D is the number of the signal separated from array 2. (l) This represents the number of bats in that frame of signal. These represent the time-domain mean values ​​of the signals separated from the first and second arrays, respectively.

[0159] (5.1.2) Based on the normalized cross-correlation coefficient c between the bat signals in the array r,n , constitute D (l) ×D (l) The normalized cross-correlation matrix C:

[0160]

[0161] (5.1.3) Find the maximum value in the normalized cross-correlation matrix C. The index of the maximum value is the signal number of the same bat separated from the two arrays. Denote the signals corresponding to these two numbers as the matching signal y. 1,k (t) and y 2,k (t), k = 1, 2, ..., D (l) The bat is numbered D (l) This represents the number of bats in that frame of signal;

[0162] (5.1.4) A method of gradually eliminating matched signals is used to reduce the interference of bats with loud calls on the matching process. Specifically, after finding the maximum value in the current matrix each time, its index is recorded as a matching result, and the row and column where the index is located are set to zero. Then, the search for a new maximum value in the matrix is ​​continued, and this process is repeated until the association matching of all signals is completed.

[0163] (5.2) Estimating the Time Difference of Arrival (TDOA) of Bat Signals to Different Arrays using the cross-correlation method:

[0164] (5.2.1) Based on the matching signal y of the kth bat 1,k (t) and y 2,k (t) Calculate the cross-correlation function R k (τ):

[0165]

[0166] The cross-correlation function R k The value of τ corresponding to the peak value of (τ) is the time difference of arrival (TDOA) of the k-th bat signal arriving at the two arrays, denoted as t. k k = 1, 2, ..., D (l) The bat is numbered D (l) This represents the number of bats in that frame of signal;

[0167] (5.2.2) Perform the process of (5.2.1) on all matched signals to obtain the time difference of arrival (TDOA) of each bat's signal to each array.

[0168] Step 6: Estimate the bat's location based on the angle of arrival (AOA) and time difference of arrival (TDOA).

[0169] The principle of TDOA and AOA joint localization is to determine the target position by simultaneously solving the TDOA and AOA measurement equations. Geometrically, the target position is determined by the intersection of the time-difference positioning hyperboloid and the angle-difference positioning ray; solving for this intersection point enables 3D localization of the bat. The specific implementation includes the following:

[0170] (6.1) The azimuth angle α of the k-th bat signal estimated from the m-th array. m,k and pitch angle θ m,k Construct the angle information matrix γ m,k :

[0171]

[0172] Where, γ m,k This is an angle information matrix, where m = 1, 2 are array numbers, and k = 1, 2, ..., D. (l) Numbering bats, γ 1,k and γ 2,k Let represent the angle information matrices of the first array and the second array, respectively. T Indicates transpose;

[0173] (6.2) Based on the angle information matrix γ of the first array 1,k The second array's angle information matrix γ 2,kand the center position coordinates s of each microphone array m =[x m ,y m ,z m ] T Construct the first angle coefficient matrix Second angle coefficient matrix

[0174]

[0175] (6.3) Based on the arrival time difference t of the k-th bat signal to each array k Calculate the difference in arrival distance r between each array. k :

[0176] r k =t k *c

[0177] Where c is the speed at which bat signals propagate;

[0178] (6.4) The azimuth angle α of the k-th bat signal estimated by array 1 1,k and pitch angle θ 1,k Construct a unit vector b:

[0179] b = [cosθ] 1,k cosα 1,k cosθ 1,k sinα 1,k sinθ 1,k ] T ;

[0180] (6.5) Based on the arrival distance difference r of the k-th bat signal to each array k The center position coordinates s of each microphone array m =[x m ,y m ,z m ] T Construct the first distance coefficient matrix h with the unit vector b. r,k Second distance coefficient matrix G r,k :

[0181]

[0182] G r,k =[2(s1-s2-r k b) T ]

[0183] Where ||·|| denotes the norm, and s1=[x1,y1,z1] TLet s2 be the coordinates of the center position of the first array, where s2 = [x2, y2, z2]. T The coordinates of the center position of the second array;

[0184] (6.6) Joint first angle coefficient matrix Second angle coefficient matrix First distance coefficient matrix h r,k Second distance coefficient matrix G r,k Construct the first coefficient matrix h k Second coefficient matrix G k :

[0185]

[0186]

[0187] (6.7) Based on the first coefficient matrix h k Second coefficient matrix G k Calculate the position coordinates u of the k-th bat. k =[x k ,y k ,z k ] T :

[0188]

[0189] (6.8) Repeat steps (6.1) to (6.7) for the angle of arrival (AOA) and time difference of arrival (TDOA) of all bat signals to estimate the position of each bat and reconstruct the movement trajectory of each bat based on the position of each bat at different times.

[0190] Example 2: Signal Separation and Localization System for Swarming Bats

[0191] Based on the same inventive concept, this invention also proposes a system for separating and locating signals of swarming bats, such as... Figure 2 As shown, it includes:

[0192] The signal receiving and preprocessing module is used to receive the signals of swarming bats using at least two microphone arrays, and to perform amplitude and phase consistency correction, bandpass filtering and framing processing on them to generate normalized aliased signal frames.

[0193] The bat count estimation module is used to perform time-frequency analysis on the normalized aliased signal frame and estimate the number of bats contained in the aliased signal frame.

[0194] The Angle of Arrival (AOA) estimation module is used to iteratively project and eliminate interference signals based on the estimated number of bats to obtain the AOA of each bat signal.

[0195] The signal separation module is used to separate the signal of each bat based on the angle of arrival (AOA) of each bat signal using a frequency domain beamforming method.

[0196] The Time Difference of Arrival (TDOA) estimation module is used to match the bat signals separated from each array and estimate the TDOA of the matched signals to each array based on the cross-correlation characteristics of the same bat signal between arrays.

[0197] The three-dimensional positioning module is used to estimate the three-dimensional spatial position of each bat based on the angle of arrival (AOA) and time difference of arrival (TDOA).

[0198] The technical effects of the present invention will be further explained below with reference to simulation experiments.

[0199] I. Simulation conditions:

[0200] Two 8×8 uniform planar microphone arrays were used to receive echolocation signals from bats. The echolocation signals emitted by bats consist of a constant frequency band and a downward frequency modulated band. The speed of sound is c = 340 m / s, the center frequency is about 68000 Hz, and the element spacing is half a wavelength.

[0201] Hardware environment: CPU is Intel(R) Core(TM) i9-14900HX with a clock speed of 2.20GHz, memory is 32.0GB, and operating system is 64-bit.

[0202] Software environment: Microsoft Windows 11 Home Chinese Edition, MATLAB 2024a simulation software.

[0203] Simulation scenario: Array 1 is positioned at (0,0,0), and Array 2 is positioned at (0,0.5,0), in meters (m). Two bats, 1 and 2, fly towards the microphone array. The echolocation calls emitted by each bat are received by the microphone array, where:

[0204] The call of Bat 1 has a constant frequency of 68196Hz, a frequency modulation bandwidth of 10kHz, and an amplitude of 0.2.

[0205] The call of Bat 2 has a constant frequency of 68141Hz, a frequency modulation bandwidth of 10kHz, and an amplitude of 9.6.

[0206] Bat 1 took 1.16 seconds to turn at a constant speed from position (3, -0.58, -0.91) to position (1.61, -1.11, -0.51), and made a total of 29 sounds during the flight.

[0207] Bat 2 took 1.17 seconds to turn at a constant speed from position (3, -0.7, -0.21) to (0.44, -0.86, -0.67) and made 41 sounds.

[0208] Both bat signals had a constant frequency band of 10ms and a frequency modulation band of 2ms.

[0209] II. Simulation Content

[0210] Simulation 1: Simulating the time-frequency diagram and waveform diagram of the signal received by the microphone channel at the origin of microphone array 1 when two bats fly simultaneously. The results are as follows: Figure 3 As shown, Figure 3 (a) is the time-frequency diagram of the received signal. Figure 3 (b) shows the corresponding time-domain waveform. Figure 3 It can be seen that when two bats fly at the same time, the signals received by the microphone array exhibit obvious time-frequency aliasing.

[0211] Simulation 2: For the signal in Simulation 1 that exhibits time-frequency aliasing and has a significant energy difference between the two signals in one frame, the angle measurement is performed using the AOA algorithm based on iterative projection to eliminate strong interference signals from this invention. The results are as follows: Figure 4 As shown. Wherein:

[0212] Figure 4 (a) is the time-frequency diagram of this frame of signal, including weak signal 1 and strong signal 2. From this, it can be intuitively estimated that the number of signals is 2;

[0213] Figure 4 (b) is the spatial spectrum of this frame of signal obtained using the MUSIC algorithm. Only a single spectral peak can be observed in this spatial spectrum, indicating that the strong signal masks the weak signal, making it impossible to effectively identify the weak signal.

[0214] Figure 4 (c) is the spatial spectrum estimated by applying the AOA estimation algorithm for eliminating strong interference signals based on iterative projection in this invention. The spatial spectrum realizes the successful estimation of the AOA of weak signals.

[0215] Depend on Figure 4 The results show that the present invention can effectively estimate the angle of weak signals under the condition of strong signal masking.

[0216] Simulation 3: For the signal in Simulation 2 that exhibits time-frequency aliasing and has a significant energy difference between the two signals in one frame, the frequency-domain LCMV beamforming algorithm used in this invention is employed to effectively separate the signals, and the separated time-frequency diagram is plotted. The results are as follows: Figure 5 As shown. Among them, Figure 5 (a) shows the separation result of signal 1. Figure 5 (b) shows the separation result of signal 2. From Figure 5 It can be seen that the present invention can effectively separate the signals of each bat.

[0217] Simulation 4: Based on two microphone arrays, using the algorithm employed in this invention to estimate the 3D position of bats based on the angle of arrival (AOA) and time difference of arrival (TDOA), the 3D position of the flying bats is estimated and the flight trajectories of the two bats are reconstructed. The results are as follows: Figure 6 As shown in the image, this figure compares the actual flight paths of two bats with the paths reconstructed by the algorithm. Figure 6 As can be seen, this invention can accurately reconstruct the movement trajectory of flying bats.

[0218] In summary, the proposed method for signal separation and localization of swarming bats can effectively estimate the AOA of bat signals with significant differences in signal strength and successfully separate the signals of different individual bats. Based on the cooperation of two microphone arrays, the movement trajectory of flying bats can be accurately reconstructed, providing important technical support for bat behavior and biomimetic radar research.

[0219] The above description is merely two specific examples of the present invention and does not constitute any limitation on the present invention. Obviously, those skilled in the art, after understanding the content and principles of the present invention, may make various modifications and changes in form and details without departing from the principles and structure of the present invention. However, these modifications and changes based on the ideas of the present invention are still within the scope of protection of the claims of the present invention.

[0220] It should be noted that the step numbers in the specification and claims of this invention are only for the purpose of clearly describing the embodiments of this invention and facilitating understanding, and their order is not limited.

Claims

1. A method for separating and locating signals from swarms of bats, characterized in that, include: Receive signals from swarms of bats using at least two microphone arrays. After amplitude and phase consistency correction, bandpass filtering, and framing processing, the first... Frame aliasing signal ,in For frame numbering, Total number of frames Number the array. For the array number, Number the microphones. The total number of microphones for each array; According to the Frame aliasing signal The energy distribution in the time-frequency domain is used to estimate the number of bats in that frame. ; Based on the number of bats Iterative projection eliminates interference signals, for each array's first... Frame signal Angle measurement was performed to obtain the bat's arrival angle AOA, where ;include: 4a) For signals The MUSIC algorithm was used to measure angles and obtain the number of spectral peaks. and the corresponding angle of arrival (AOA); 4b) Determine the number of spectral peaks Is it less than the target number? : If so, proceed to step 4c). Otherwise, from Select the peak with the strongest energy from the spectral peaks. The angle of arrival corresponding to each spectral peak is used as the bat's angle of arrival (AOA). 4c) Continue estimating the signal by eliminating interference signals through iterative projection. Hidden in A bat signal AOA, and obtained from 4a) Merge the AOAs to get One bat reached the angle AOA; Based on the bat's angle of arrival (AOA), frequency domain beamforming is used to analyze the signal. Separate to obtain the first Signals of each bat in the frame ,in Numbering bat signals; including: 5a) Initialize the number of known spectral peaks to 0; 5b) According to the signal An orthogonal projection matrix is ​​constructed based on the angle of arrival corresponding to the strongest spectral peak in the medium energy range, and this matrix is ​​then used to analyze the signal. Projecting the covariance matrix onto the given covariance matrix yields the updated covariance matrix. 5c) Apply the MUSIC algorithm to the updated covariance matrix to measure the angle, obtain the current number of spectral peaks and the corresponding angle of arrival, and add the newly obtained number of spectral peaks to the known number of spectral peaks; 5d) Determine if the number of known spectral peaks is less than the number of targets to be masked. : If so, then execute 5e). Otherwise, select the one with the strongest energy. The angle of arrival corresponding to each spectral peak is used as the angle of arrival of the masked bat. 5e) Extract the angle of arrival corresponding to the strongest spectral peak in 5c) to construct an orthogonal projection matrix, update the covariance matrix, and then return to 5c); A signal correlation method that cyclically removes the maximum value is used to progressively correlate the bat signals of each array. Matching is performed, and based on the cross-correlation characteristics of the same bat signal between arrays, the time difference of arrival (TDOA) between arrays for the same bat signal is estimated. The three-dimensional position of each bat was calculated by combining the bat's angle of arrival (AOA) and the time difference of arrival (TDOA) between arrays.

2. The method for separating and locating signals of swarming bats according to claim 1, characterized in that, The amplitude-phase consistency correction includes: 2a) Using a known frequency A standard narrowband signal source illuminates each microphone array and simultaneously acquires signals from each channel. ; 2b) Extracting the acquired signal Complex spectrum value at the center frequency The complex spectrum value of the first channel is selected. For reference, calculate the amplitude and phase correction coefficients. ; 2c) Correction coefficient Data received by the array Multiply the frequency domain signal to obtain the frequency domain corrected signal, and then convert it to the time domain to obtain the amplitude-phase consistency corrected signal.

3. The method for separating and locating signals of swarming bats according to claim 1, characterized in that, According to the first Frame aliasing signal The energy distribution in the time-frequency domain is used to estimate the number of bats in that frame. ,include: 3a) For the first Frame aliasing signal Perform time-frequency analysis to generate a time-frequency graph of the signal in that frame; 3b) Identify and count the independent regions of concentrated bat signal energy in the time-frequency graph to obtain the number of bats. .

4. The method for separating and locating signals of swarming bats according to claim 1, characterized in that, According to the signal An orthogonal projection matrix is ​​constructed based on the angle of arrival corresponding to the strongest spectral peak in the medium energy range, and this matrix is ​​then used to analyze the signal. Projecting the covariance matrix onto the given covariance matrix yields the updated covariance matrix, which includes: 6a) Signal Extraction The angle of arrival (AOA) corresponding to the strongest spectral peak, including the azimuth angle. and pitch angle Construct the steering vector of the signal. : ； in, For the first in the array The coordinates of each microphone are relative to the center of the array. The wavelength of the bat signal; 6b) Using guide vectors Constructing the projection matrix : ； in, As a unit array, Indicates conjugate transpose; 6c) Using projection matrix For signals covariance matrix By projection, the updated covariance matrix is ​​obtained. .

5. The method for separating and locating signals of swarming bats according to claim 1, characterized in that, The frequency domain beamforming method is used to analyze the signal based on the bat's angle of arrival (AOA). Separate to obtain the first Signals of each bat in the frame ,include: 7a) For signals Perform a Fast Fourier Transform (FFT) to obtain the frequency domain signal. And represent it using data from all frequency points: ； in, For the first Frequency at each frequency point Frequency point numbering, This represents the total number of frequency points. 7b) Calculate the frequency domain signal covariance matrix : ； in, Represents the mathematical expectation. Indicates conjugate transpose; 7c) Regarding the first The array of the first Calculate the frequency of a bat signal. Corresponding frequency domain steering vector : ； in, For the first The first array estimate of the first The azimuth angle of a bat signal. For the first The array estimate of the first The pitch angle of a bat signal, Number the array. Number of microphone arrays Numbering bat signals For bat signal count, For the first in the array The coordinates of each microphone are relative to the center of the array. Number the microphones. Number of microphones per array For frequency The corresponding wavelength; 7d) Calculate the frequency of the interference signal. Corresponding frequency domain steering vector : ； in, For the first The array estimate of the first The azimuth angle of the interference signal. For the first The array estimate of the first... The elevation angle of the interfering signal, Number the interference signal. This represents the total number of interfering signals. Frequency domain steering vector in the direction where interference signals are to be suppressed Composition of interference signal steering vector matrix : ； 7e) According to the The array of the first bat signal frequency Frequency domain steering vector at the location and interference signal steering vector matrix , forming a frequency domain steering vector matrix : ； 7f) Regarding the first The array of the first Calculate the frequency of a bat signal. The corresponding optimal weights for frequency domain LCMV beamforming : ； in, To perform the inverse operation; 7g) Using weight vectors For the received signal Each frequency point is weighted separately to obtain the separated signal at each frequency point. By further combining the separated signals at each frequency point, the frequency domain representation of the separated signals is obtained. : ； 7h) Frequency domain components of each bat signal Performing an inverse fast Fourier transform (IFFT) yields the time-domain signals of each bat. .

6. The method for separating and locating signals of swarming bats according to claim 1, characterized in that, The signal correlation method, which employs cyclic elimination of maximum values, progressively processes the bat signals of each array. Matching is performed, and based on the cross-correlation characteristics of the same bat signal between arrays, the inter-array time difference of arrival (TDOA) for the same bat signal is estimated, including: 8a) Calculate the bat signals between arrays Normalized cross-correlation coefficients between them, construct Correlation coefficient matrix ; 8b) In Find the maximum value in the middle, take the bat signal corresponding to the row and column where the maximum value is located as the matched bat signal, and set the corresponding row and column to zero. Repeat this step until all bat signals are matched. 8c) Calculate the cross-correlation function for the matched bat signals and estimate the time difference of arrival (TDOA) between arrays based on the position of the cross-correlation peak.

7. The method for separating and locating signals of swarming bats according to claim 1, characterized in that, The combined bat arrival angle (AOA) and inter-array time difference of arrival (TDOA) estimate the three-dimensional position of each bat, including: 9a) Construct an angle information matrix based on the bat arrival angle (AOA) of each array. : ； in, and They represent the first The array estimate of the first The azimuth and elevation angles of a bat signal. Indicates transpose; 9b) Based on the angle information matrix of the first array The angle information matrix of the second array and the center coordinates of each microphone array Construct the first angle coefficient matrix Second angle coefficient matrix : ； ； 9c) According to the first Time difference of arrival of each bat signal to each array Calculate the difference in arrival distance to each array. : ； in, The speed at which bat signals travel; 9d) Based on the estimated first array, the... Azimuth angle of a bat signal and pitch angle Construct unit vectors : ； 9e) According to the The difference in arrival distance of each bat signal to each array Center coordinates of each microphone array and unit vector Construct the first distance coefficient matrix Second distance coefficient matrix : ； ； in, Represents the norm; 9f) Joint first angle coefficient matrix Second angle coefficient matrix First distance coefficient matrix Second distance coefficient matrix Construct the first coefficient matrix Second coefficient matrix : ； ； 9g) Based on the first coefficient matrix Second coefficient matrix Calculate the first The location coordinates of the bat : 。 8. A system for separating and locating signals of swarming bats, characterized in that, include: The signal receiving and preprocessing module is used to receive the signals of swarming bats using at least two microphone arrays, and to perform amplitude and phase consistency correction, bandpass filtering and framing processing on them to generate normalized aliased signal frames. The bat count estimation module is used to perform time-frequency analysis on the normalized aliased signal frame and estimate the number of bats contained in the signal frame. Angle of Arrival (AOA) estimation module, used to iteratively project and eliminate interference signals based on the estimated number of bats, to obtain the AOA for each bat signal; including: 4a) For signals The MUSIC algorithm was used to measure angles and obtain the number of spectral peaks. and the corresponding angle of arrival (AOA); 4b) Determine the number of spectral peaks Is it less than the target number? : If so, proceed to step 4c). Otherwise, from Select the peak with the strongest energy from the spectral peaks. The angle of arrival corresponding to each spectral peak is used as the bat's angle of arrival (AOA). 4c) Continue estimating the signal by eliminating interference signals through iterative projection. Hidden in A bat signal AOA, and obtained from 4a) Merge the AOAs to get One bat reached the angle AOA; The signal separation module is used to separate the signal of each bat based on its angle of arrival (AOA) using a frequency domain beamforming method; it includes: 5a) Initialize the number of known spectral peaks to 0; 5b) According to the signal An orthogonal projection matrix is ​​constructed based on the angle of arrival corresponding to the strongest spectral peak in the medium energy range, and this matrix is ​​then used to analyze the signal. Projecting the covariance matrix onto the given covariance matrix yields the updated covariance matrix. 5c) Apply the MUSIC algorithm to the updated covariance matrix to measure the angle, obtain the current number of spectral peaks and the corresponding angle of arrival, and add the newly obtained number of spectral peaks to the known number of spectral peaks; 5d) Determine if the number of known spectral peaks is less than the number of targets to be masked. : If so, then execute 5e). Otherwise, select the one with the strongest energy. The angle of arrival corresponding to each spectral peak is used as the angle of arrival of the masked bat. 5e) Extract the angle of arrival corresponding to the strongest spectral peak in 5c) to construct an orthogonal projection matrix, update the covariance matrix, and then return to 5c); The Time Difference of Arrival (TDOA) estimation module is used to match the bat signals separated from each array and estimate the TDOA of the matched signals to each array based on the cross-correlation characteristics of the same bat signal between arrays. A 3D localization module is used to estimate the 3D spatial position of each bat by combining the angle of arrival (AOA) and time difference of arrival (TDOA).

Images