CN112540346A - Sound source positioning method based on signal-to-noise ratio weight optimization updating - Google Patents
Sound source positioning method based on signal-to-noise ratio weight optimization updating Download PDFInfo
- Publication number
- CN112540346A CN112540346A CN202011438854.XA CN202011438854A CN112540346A CN 112540346 A CN112540346 A CN 112540346A CN 202011438854 A CN202011438854 A CN 202011438854A CN 112540346 A CN112540346 A CN 112540346A
- Authority
- CN
- China
- Prior art keywords
- sound
- signal
- noise ratio
- sound source
- calculating
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000005457 optimization Methods 0.000 title claims abstract description 12
- 230000005236 sound signal Effects 0.000 claims abstract description 116
- 230000002159 abnormal effect Effects 0.000 claims abstract description 22
- 230000005540 biological transmission Effects 0.000 claims abstract description 19
- 230000004807 localization Effects 0.000 claims description 16
- 230000003595 spectral effect Effects 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 13
- 238000005314 correlation function Methods 0.000 claims description 10
- 238000009432 framing Methods 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 9
- 238000001914 filtration Methods 0.000 claims description 7
- 238000001514 detection method Methods 0.000 claims description 4
- 230000007613 environmental effect Effects 0.000 claims description 3
- 238000009499 grossing Methods 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 28
- 230000008569 process Effects 0.000 description 6
- 239000000654 additive Substances 0.000 description 5
- 230000000996 additive effect Effects 0.000 description 5
- 230000003321 amplification Effects 0.000 description 4
- 238000003199 nucleic acid amplification method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 238000005311 autocorrelation function Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 210000005069 ears Anatomy 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013341 scale-up Methods 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
- G01S5/18—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using ultrasonic, sonic, or infrasonic waves
- G01S5/22—Position of source determined by co-ordinating a plurality of position lines defined by path-difference measurements
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
The application discloses sound source positioning method based on signal-to-noise ratio weight optimization updating, which is used for carrying out sound source positioning on an element which makes sound in equipment to be detected, and comprises the following steps: step 1, amplifying a first sound signal which is collected and emitted by a sound source according to a preset proportion, and performing gain adjustment to generate a second sound signal; step 2, when the second sound signal is judged not to be an abnormal noise signal, sequentially calculating the current environment signal-to-noise ratio corresponding to the second sound signal and the signal-to-noise ratio related weight mean value of the second sound signal, and calculating the transmission delay of the second sound signal corresponding to the reference sound sensor according to the signal-to-noise ratio related weight mean value; and 3, calculating the positioning information of the sound source according to the transmission time delay and the position coordinates of the sound sensor, wherein the positioning information is used for positioning the sound source. Through the technical scheme in this application, can accurately capture the specific positional information at device sound production position fast, realize the intelligent real-time location of sound source.
Description
Technical Field
The application relates to the technical field of sound source positioning, in particular to a sound source positioning method based on signal-to-noise ratio weight optimization updating.
Background
Various sounds such as noise, abnormal sound and the like are common in daily life and industrial production, the original sound positioning function of human ears is only used for solving the basic problems of life and survival, and the positioning precision is very limited. Therefore, the application of the sound source positioning technology is very important, and the product can quickly position a noise source or an abnormal sound position, so that an engineer is helped to quickly solve the problem, and the working efficiency is greatly improved.
Nowadays, microphone arrays are widely applied to the field of sound signal processing such as speech enhancement and speech recognition, sound source localization is a technology for determining a spatial source position of sound, and is one of key technologies of array signal processing based on microphone arrays, and is a hot research topic of society.
In the prior art, a traditional time delay estimation algorithm is usually adopted for sound source positioning, since reverberation is severe in an indoor environment and noise is strong in an outdoor environment, positioning performance is greatly affected, and a traditional sound source positioning method is poor in positioning performance in an environment with high reverberation and strong noise and low signal-to-noise ratio (SNR), and cannot meet requirements of various actual scenes.
Disclosure of Invention
The purpose of this application lies in: the specific position information of the sounding part of the device is captured quickly and accurately, and the intelligent real-time positioning of the sound source is realized.
The technical scheme of the application is as follows: the utility model provides a sound source localization method based on that SNR weight optimizes the renewal, this method is used for treating the sound source localization of the component that makes sound in the equipment that will detect, and the method includes: step 1, amplifying a first sound signal which is collected and emitted by a sound source according to a preset proportion, and performing gain adjustment to generate a second sound signal, wherein the first sound signal is collected by a plurality of sound sensors; step 2, when the second sound signal is judged not to be an abnormal noise signal, sequentially calculating a current environment signal-to-noise ratio corresponding to the second sound signal and a signal-to-noise ratio related weight mean value of the second sound signal, and calculating transmission delay of the second sound signal corresponding to a reference sound sensor according to the signal-to-noise ratio related weight mean value, wherein the reference sound sensor comprises a first sound sensor and a second sound sensor; and 3, calculating the positioning information of the sound source according to the transmission time delay and the position coordinates of the sound sensor, wherein the positioning information is used for positioning the sound source.
In any one of the above technical solutions, further, in step 2, the method for calculating the current environment signal-to-noise ratio corresponding to the second sound signal specifically includes:
step 201, selecting any two sound sensors as reference sound sensors, and calculating two corresponding short-time stationary signals through filtering and windowing and framing processing according to a second sound signal corresponding to the reference sound sensors;
step 202, when the short-time energy and the short-time zero crossing rate of the two short-time stationary signals are judged to be larger than the preset threshold value, a second sound signal corresponding to the two short-time stationary signals is judged to be an abnormal noise signal, the abnormal noise signal is discarded, the step 201 is executed again, and otherwise, the step 203 is executed;
step 203, when the second sound signal corresponding to the two short-time stationary signals is judged not to be the abnormal noise signal, calculating the current environment signal-to-noise ratio, wherein the calculation formula of the current environment signal-to-noise ratio is as follows:
SNR(λ)＝aSNR(λ-1)+(1-a)SNR_0
wherein λ is the frame number of the second sound signal after framing, a is a smoothing factor, SNR (λ -1) is the current environment SNR of the second sound signal of the previous frame, and SNR _0 is the energy ratio of the second sound signal of the λ -th frame to the second sound signal of the λ -1-th frame.
In any one of the above technical solutions, further, in step 2, the snr dependent weight at least includes: the method for calculating the SNR (signal-to-noise ratio) of the current environment, the cross-power spectral density function of a stationary signal, the weighting function of the SNR, the cross-correlation function of a sound signal and the related weight of the SNR specifically comprises the following steps: step 211, performing fast fourier transform on two short-time stationary signals of the second sound signal corresponding to any two sound sensorsTransforming and calculating a cross-power spectral density function R12(λ, w); step 212, determining an adjustment factor according to the signal-to-noise ratio of the current environment, and according to a cross-power spectral density function R12(lambda, w), calculating a signal-to-noise ratio weighting function
Step 213, calculating a cross-power spectral density function R12(lambda, w) and signal-to-noise ratio weighting function
And inverse fourier transforming the product to generate the acoustic signal cross-correlation function r12(λ,w)。
In any one of the above technical solutions, further, calculating a transmission delay of the second sound signal corresponding to the reference sound sensor specifically includes: step 221, orderly grouping the sound sensors in pairs, and recording the sound sensors as reference sound sensors; step 222, for each group of reference sound sensors, the sound signal cross-correlation function r12Carrying out peak value detection on the value sequence of (lambda, w) to obtain a sampling point corresponding to the maximum discrete value point; step 223, multiplying the sampling point by the sampling point interval time to generate the predicted time delay of the second sound signal corresponding to the reference sound sensor; and 224, repeatedly executing the steps 221 to 224 for each group of reference sound sensors by adopting a loop traversal algorithm, calculating the average value of the predicted time delay, and recording the average value as the transmission time delay.
In any of the above technical solutions, further, the positioning information of the sound source at least includes: the azimuth angle of the sound source is theta, the pitch angle of the sound source is mu, and the distance from the sound source to each sound sensor is.
The beneficial effect of this application is:
according to the technical scheme, the sound source positioning method and device are combined with corresponding hardware equipment and deployed in an actual application scene, specific position information of a sound production part of the device can be captured quickly and accurately, intelligent real-time positioning of the sound source is achieved, and the sound source positioning method and device have good performance and popularization and application prospects.
The five-element microphone array is arranged to collect sound, the sound source signal microprocessor is used for carrying out secondary amplification operation on sound signals, the signal-to-noise ratio weight function combining the sound signal power value provided by the application is reused, then reverberation and noise of the current environment can be self-adapted, and finally the time delay and position characteristics are combined to carry out specific calculation of the azimuth coordinate on a target sound source, so that sound source positioning is realized. The disclosed method can meet the practical requirements on positioning precision and reasoning speed, and realizes the application on the front-end artificial intelligent embedded equipment.
Compared with the traditional sound source positioning algorithm at present, the method combines hardware modularization equipment to process signals, combines machine learning, mode recognition and other related algorithms to perform related analysis and processing on collected sound signal streams, innovatively provides a signal-to-noise ratio weight function combined with a sound signal power value and a sound source positioning method combining the signal-to-noise ratio weight and a time delay characteristic, breaks through a conventional thought, has obvious performance characteristics of noise resistance and reverberation resistance, obviously improves experimental performance, and mainly embodies the following two aspects:
(1) the microphone has strong sound pickup capability. According to the method, the sound source signal microprocessor is used for amplifying the sound signal for multiple times, so that the output voltage of the microphone is multiplied, and the input sound signal stream can meet the requirement of post signal processing.
(2) The anti-noise and anti-reverberation effect is obvious. According to the method, the traditional time delay estimation algorithm is not used for sound source positioning, the method is correspondingly adjusted through the change of the current signal to noise ratio, the signal to noise ratio weight which is continuously optimized and updated at different moments is obtained, functions such as signal to noise ratio weighting are related to the power value of the sound signal, and the accuracy and stability of the functions and the weight value are improved.
Drawings
The advantages of the above and/or additional aspects of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a schematic flow diagram of a sound source localization method based on SNR weight optimization update according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a five-element microphone array structure according to one embodiment of the present application;
FIG. 3 is a schematic diagram of a weight function calculation process according to one embodiment of the present application.
Detailed Description
In order that the above objects, features and advantages of the present application can be more clearly understood, the present application will be described in further detail with reference to the accompanying drawings and detailed description. It should be noted that the embodiments and features of the embodiments of the present application may be combined with each other without conflict.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, however, the present application may be practiced in other ways than those described herein, and therefore the scope of the present application is not limited by the specific embodiments disclosed below.
In the embodiment, a five-element microphone array is used for sound collection, and is firstly transmitted into a sound source signal microprocessor during collection, wherein the sound source signal microprocessor comprises an integrated operational amplifier, a gain adjustment potentiometer and other modules, the sound signals are subjected to secondary amplification through microprocessing, and then the five-channel sound signals are subjected to digital conversion through an AD collection module, so that subsequent input sound signals are obtained. Then, the embodiment provides a signal-to-noise ratio weighting function combined with a sound signal power value, which can be adjusted correspondingly with the change of the current signal-to-noise ratio, and performs final calculation of a sound source position coordinate by combining a signal-to-noise ratio weighting average value, a microphone array time delay and basic position information.
The experimental results show that: the method in the embodiment is combined with corresponding hardware equipment, is deployed in an actual application scene, can quickly and accurately capture specific position information of a sounding part of the device, realizes intelligent real-time positioning of a sound source, and has good performance and popularization and application prospects for positioning of the sound source in indoor and outdoor environments.
As shown in fig. 1, the present embodiment provides a sound source localization method based on snr weight optimization update, which is used for performing sound source localization on an acoustic element in a device to be detected, and includes:
It should be noted that the output voltage range of the conventional omnidirectional type condenser microphone is about 0mV-60mV, and thus weak electrical signals cannot meet the work of signal processing in the later period at all, so that the experimental effect is not obvious, and the positioning accuracy is low. Therefore, it is necessary to perform a process of scaling up the signal collected by the microphone.
In this embodiment, microphones are used as sound sensors and form a five-element microphone array, as shown in fig. 2, a three-dimensional rectangular coordinate system is established with the geometric center of the device to be detected as an origin, the microphones in the five-element microphone array are respectively marked as numbers 1, 2,3,4, and 5, and the coordinates of the five microphones 1, 2,3,4, and 5 are S in sequence1(0,0,l)、S2(l,0,0)、S3(0,l,0)、S4(-l,0,0) and S5(0, -l,0), wherein the value of the length l can be obtained by artificial measurement as the distance from the microphone to the geometric center of the device to be detected.
The equipment to be detected can make a normal sound in the operation process, but when some element or some elements in the equipment to be detected are abnormal, the equipment to be detected can make an abnormal sound. At this time, the element can be used as a sound source, and the anomaly detection of the equipment to be detected is realized by positioning the sound source.
Therefore, the five-element microphone array is used for collecting the sound (including normal sound and abnormal sound) of the equipment to be detected and then entering a sound source signal microprocessor, wherein the sound source signal microprocessor comprises modules such as an integrated operational signal amplifier, a gain adjustment potentiometer and the like, the omnidirectional type capacitive microphone and all signal processing components are known commercial devices, and the connection method between all the modules and the components can be mastered by those skilled in the art.
This embodiment adopts the comparatively stable integrated operational signal amplifier of performance, carries out the secondary amplification according to predetermineeing the proportion to the first sound signal of gathering, introduces gain adjustment potentiometre module simultaneously for when debugging in the later stage, reduce the precision error that produces because the hardware characteristic is different. The processor may scale up the sound signal and use it as the second sound signal for subsequent operations.
And 2, when the second sound signal is judged not to be an abnormal noise signal (namely the current second sound signal is a normal sound signal), calculating a current environment signal-to-noise ratio corresponding to the second sound signal, calculating a signal-to-noise ratio related weight mean value of the second sound signal according to the current environment signal-to-noise ratio, and calculating transmission delay of the second sound signal corresponding to a reference sound sensor according to the signal-to-noise ratio related weight mean value, wherein the reference sound sensor comprises a first sound sensor and a second sound sensor.
In this embodiment, the microphone 1 is set as the first sound sensor, and the microphone 5 is set as the second sound sensor.
Specifically, as shown in fig. 3, in this embodiment, on the basis of the conventional adaptive snr weight optimization method, the power characteristics of the sound signal are blended, so that the weight function deeply extracts the characteristic information of the sound signal. Obtaining an enhanced five-channel second sound signal through a five-element microphone array and a sound source signal microprocessor, preprocessing the sound signal and calculating a series of related formulas to determine the signal-to-noise ratio related weights of the current environment signal-to-noise ratio, the cross power spectral density function of steady signals, the signal-to-noise ratio weighting function, the sound signal cross-correlation function and the like. And finally calculating the time delay characteristics of the five microphones according to the obtained signal-to-noise ratio related weight, thereby calculating the subsequent sound source azimuth coordinate.
The embodiment shows a method for calculating a current environmental signal-to-noise ratio corresponding to a second sound signal when it is determined that the second sound signal is not an abnormal noise signal, specifically including:
step 201, selecting any two sound sensors as the reference sound sensors, and calculating two corresponding short-time stationary signals through filtering and windowing and framing processing according to a second sound signal corresponding to the reference sound sensors.
Any two microphone channels in the five-element microphone array are used as reference sound sensors, and band-pass filtering processing is carried out on second sound signals which are collected by the reference sound sensors and subjected to amplification and gain adjustment, so that two paths of sound signals subjected to band-pass filtering are obtained. And then performing windowing and framing processing on the two paths of band-pass filtered sound signals to obtain two paths of short-time stable signals.
The sound signals collected by the two microphones are assumed to be:
x1(t)＝a1s1(t)+n1(t)
x2(t)＝a2s1(t+D)+n2(t)
wherein a is1、a2Is a sound attenuation factor, since the sound source is a near-field signal, it can be considered that a1、a2The value 1, D is the time delay of the sound signal to the selected two microphones, n1(t)、n2(t) is the noise signal received by the selected two microphones at time t.
And then, the selected second sound signal is subjected to band-pass filtering processing, so that the noises of a low frequency band and a high frequency band are eliminated, and two paths of sound signals subjected to band-pass filtering are provided for subsequent processing.
After the two paths of band-pass filtered sound signals are obtained, a Hamming window function is used for framing the two paths of band-pass filtered sound signals to obtain two paths of short-time stable signals, and a frame-to-frame overlapping method is generally adopted for windowing and framing. The two short-time steady signals obtained are:
s1(λ,n)＝x1×(n+d×(λ-1)×N)×o(n)
s2(λ,n)＝x2×(n+d×(λ-1)×N)×o(n)
where o (N) is a Hamming window function, N is the length of the window function o (N), d is a shift parameter between adjacent frames, λ is the number of frames, and N is the sound sequence length.
Step 202, when the short-time energy and the short-time zero crossing rate of the two short-time stationary signals are judged to be larger than the preset threshold, the second sound signal corresponding to the two short-time stationary signals is judged to be an abnormal noise signal, the abnormal noise signal is discarded, the step 201 is executed again, and otherwise, the step 203 is executed.
Specifically, the first sound signal collected by the five-element microphone array includes a normal sound signal and a background noise signal, and if the sound source does not produce sound, the collected sound signal is only the background noise signal (i.e., the normal noise signal). Specifically, when both the short-time energy (energy of the sound signal for a short period of time) and the short-time zero-crossing rate (the number of times the signal waveform crosses the horizontal axis (zero level) per unit time) of the detected two-way short-time stationary signal are larger than a preset threshold, it is determined that the current sound signal is an abnormal noise signal.
Step 203, when it is determined that the second sound signal corresponding to the two short-time stationary signals is not an abnormal noise signal, calculating the current environment signal-to-noise ratio, where a calculation formula of the current environment signal-to-noise ratio is:
SNR(λ)＝aSNR(λ-1)+(1-a)SNR_0
wherein λ is the frame number of the second sound signal after framing, a is a smoothing factor, SNR (λ -1) is the current environment SNR of the second sound signal of the previous frame, and SNR _0 is the energy ratio of the second sound signal of the λ -th frame to the second sound signal of the λ -1-th frame.
Further, when the second sound signal corresponding to the two-path short-time stationary signal is judged to be an abnormal noise signal, the current environment signal-to-noise ratio of the current frame second sound signal is updated according to the current environment signal-to-noise ratio of the previous frame second sound signal.
In this embodiment, the snr dependent weights at least include: SNR (lambda) of current environment signal to noise ratio and cross-power spectral density function R of stationary signal12(lambda, w), signal-to-noise ratio weighting function
Sound signal cross correlation function r12(λ,w)。
The embodiment also shows a method for calculating the signal-to-noise ratio related weight, which specifically includes:
step 211, performing fast fourier transform on two paths of short-time stationary signals of the second sound signals corresponding to any two sound sensors, and calculating the cross-power spectral density function R12(λ, w), the corresponding calculation formula is:
in the formula, S1(lambda, w) is a finite-length sequence with the time window function length of N of the second sound signal of the lambda frame corresponding to the first path of sound sensor, and S1(lambda, w) is a finite long sequence with the time window function length N of the second sound signal of the lambda frame corresponding to the second path of sound sensor,
is S2And (lambda, w), wherein the first and second sound sensors are any two sound sensors in the five-element microphone array.
Step 212, determining an adjustment factor according to the signal-to-noise ratio of the current environment, and determining a cross-power spectral density function R according to the cross-power spectral density function12(λ, w) calculating the signal-to-noise ratio weighting function
Specifically, in order to resist larger noise and reverberation, and in consideration of the fact that the energy of the second sound signal after being amplified in time is still small, in this embodiment, an adjustment factor ρ in a direct relationship with the current environmental signal-to-noise ratio SNR (λ) is introduced, a value of the adjustment factor ρ is obtained through multiple experimental tests in the sound source environment, the value depends on the current signal-to-noise ratio SNR (λ), different signal-to-noise ratios SNR (λ), the adjustment factor ρ takes different values, and the higher the SNR (λ), the larger the value of the adjustment factor ρ is, which is a specific value taking manner:
when SNR (lambda) is less than or equal to 10dB, the value range of the adjusting factor rho is more than or equal to 0.25 and less than or equal to 0.55;
when SNR (lambda) is less than 10dB and less than or equal to 30dB, the value range of the adjusting factor rho is more than 0.55 and less than or equal to 0.85;
when 30dB is less than SNR (lambda), the value range of the adjusting factor rho is 0.85< rho is less than or equal to 0.95.
The additive noise is an independent noise that interferes with a desired signal such as thermal noise, and the additive noise is added to the signal, and exists regardless of the presence or absence of the signal. Therefore, the SNR weighting function takes into account additive noise
The calculation formula of (2) is as follows:
wherein rho is an adjusting factor in direct proportion to the SNR (lambda) of the current environment signal-to-noise ratio1(w) and phi2And (w) is the autocorrelation function of the two paths of second sound signals.
SNR weighting function without considering additive noise
The calculation formula of (2) is as follows:
wherein rho is an adjusting factor in direct proportion to the SNR (lambda) of the current environment signal-to-noise ratio1(w) and phi2And (w) is the autocorrelation function of the two paths of second sound signals.
In this embodiment, in order to ensure the accuracy of sound source localization, additive noise in the actual environment is considered, and therefore, the snr weighting function
The calculation formula of (2) is as follows:
step 213, calculating the cross-power spectral density function R12(λ, w) and said signal-to-noise ratio weighting function
And inverse fourier transforming said product to generate said acoustic signal cross-correlation function r12(λ, w), the corresponding calculation formula is:
in the formula, α means the number of times of accumulation calculation for each frame length in the time domain of the selected window function.
In this embodiment, calculating the transmission delay of the second sound signal corresponding to the reference sound sensor specifically includes:
step 221, orderly grouping the sound sensors in pairs, and recording the sound sensors as the reference sound sensors;
step 222, for each group of reference sound sensors, the sound signal cross-correlation function r12Carrying out peak value detection on the value sequence of (lambda, w) to obtain a sampling point corresponding to the maximum discrete value point;
step 223, multiplying the sampling point by the sampling point interval time to generate the predicted time delay of the second sound signal corresponding to the reference sound sensor;
and 224, repeatedly executing the steps 221 to 224 for each group of reference sound sensors by adopting a loop traversal algorithm, calculating the average value of the predicted time delay, and recording the average value as the transmission time delay.
Specifically, the microphones of the five-element microphone array are divided into numbers 1, 2,3,4 and 5, and the coordinates of the five microphones 1, 2,3,4 and 5 are taken as S1(0,0,l)、S2(l,0,0)、S3(0,l,0)、S4(-l,0,0) and S5(0,-l,0)。
And (3) grouping each 2 of the five microphones into ten groups, and circularly traversing each group and finally obtaining the transmission delay through the steps 221 to 224.
Setting the microphone 1 corresponding to the first sound sensor and the microphone 5 corresponding to the second sound sensor, and calculating the transmission time delay tau of the sound source reaching the microphones 2,3,4 and 5 relative to the microphone 1 by adopting a traversing mode through the processδ1(δ 2,3,4,5), corresponding to a difference in acoustic path length Δ rδ1And satisfies the following conditions: Δ rδ1＝c·τδ1(δ — 2,3,4,5), where c is the speed of sound, fixed constant, and c 340. Correspondingly, the time of sound source arriving at the microphones 1, 2,3,4 is relative to the transmission time delay tau of No. 5δ′(δ′＝1,2,3,4)。
And 3, positioning the sound source according to the transmission time delay and the position coordinates of the sound sensor.
The azimuth angle of the sound source is set to theta, the pitch angle of the sound source is set to mu, and the distances from the sound source to each microphone (sound sensor) are respectively set to z1、z2、z3、z4And z5. The process of locating the sound source is the above-mentioned location information (azimuth angle theta, pitch angle mu, and distance z)1、z2、z3、z4And z5) And solving the calculation process, wherein the corresponding calculation formula is as follows:
in the formula, ziIs representative of the distance a sound source (where the sound emitting element is located) reaches each microphone (sound sensor).
The technical solution of the present application is described in detail above with reference to the accompanying drawings, and the present application provides a sound source localization method based on snr weight optimization updating, which is used for performing sound source localization on an element emitting sound in a device to be detected, and includes: step 1, amplifying a first sound signal which is collected and emitted by a sound source according to a preset proportion, and performing gain adjustment to generate a second sound signal; step 2, when the second sound signal is judged not to be an abnormal noise signal, sequentially calculating the current environment signal-to-noise ratio corresponding to the second sound signal and the signal-to-noise ratio related weight mean value of the second sound signal, and calculating the transmission delay of the second sound signal corresponding to the reference sound sensor according to the signal-to-noise ratio related weight mean value; and 3, calculating the positioning information of the sound source according to the transmission time delay and the position coordinates of the sound sensor, wherein the positioning information is used for positioning the sound source. Through the technical scheme in this application, can accurately capture the specific positional information at device sound production position fast, realize the intelligent real-time location of sound source.
The steps in the present application may be sequentially adjusted, combined, and subtracted according to actual requirements.
The units in the device can be merged, divided and deleted according to actual requirements.
Although the present application has been disclosed in detail with reference to the accompanying drawings, it is to be understood that such description is merely illustrative and not restrictive of the application of the present application. The scope of the present application is defined by the appended claims and may include various modifications, adaptations, and equivalents of the invention without departing from the scope and spirit of the application.
Claims (5)
1. A sound source localization method based on snr weight optimization update, wherein the method is used for sound source localization of an acoustic element in a device to be detected, the method comprises:
step 1, amplifying a first sound signal which is collected and emitted by a sound source according to a preset proportion, and performing gain adjustment to generate a second sound signal, wherein the first sound signal is collected by a plurality of sound sensors;
step 2, when the second sound signal is judged not to be an abnormal noise signal, sequentially calculating a current environment signal-to-noise ratio corresponding to the second sound signal and a signal-to-noise ratio related weight mean value of the second sound signal, and calculating transmission delay of the second sound signal corresponding to a reference sound sensor according to the signal-to-noise ratio related weight mean value, wherein the reference sound sensor comprises a first sound sensor and a second sound sensor;
and 3, calculating the positioning information of the sound source according to the transmission time delay and the position coordinates of the sound sensor, wherein the positioning information is used for positioning the sound source.
2. The sound source localization method based on snr weight optimization updating of claim 1, wherein in the step 2, the method for calculating the current environmental snr corresponding to the second sound signal specifically includes:
step 201, selecting any two sound sensors as the reference sound sensors, and calculating two corresponding short-time stationary signals through filtering and windowing and framing processing according to a second sound signal corresponding to the reference sound sensors;
step 202, when the short-time energy and the short-time zero crossing rate of the two short-time stationary signals are judged to be larger than a preset threshold value, a second sound signal corresponding to the two short-time stationary signals is judged to be an abnormal noise signal, the abnormal noise signal is discarded, the step 201 is executed again, and otherwise, the step 203 is executed;
step 203, when it is determined that the second sound signal corresponding to the two short-time stationary signals is not an abnormal noise signal, calculating the current environment signal-to-noise ratio, where a calculation formula of the current environment signal-to-noise ratio is:
SNR(λ)＝aSNR(λ-1)+(1-a)SNR_0
wherein λ is the frame number of the second sound signal after framing, a is a smoothing factor, SNR (λ -1) is the current environment SNR of the second sound signal of the previous frame, and SNR _0 is the energy ratio of the second sound signal of the λ -th frame to the second sound signal of the λ -1-th frame.
3. The sound source localization method based on snr weight optimization updating of claim 1, wherein in the step 2, the snr related weight at least comprises: the method for calculating the signal-to-noise ratio related weight comprises the following steps of SNR (signal-to-noise ratio) of the current environment, a cross-power spectral density function of a stationary signal, a signal-to-noise ratio weighting function and a sound signal cross-correlation function, and specifically comprises the following steps:
step 211, two short circuits of the second sound signals corresponding to any two sound sensorsPerforming fast Fourier transform on the time stationary signal, and calculating the cross-power spectral density function R12(λ,w)；
Step 212, determining an adjustment factor according to the signal-to-noise ratio of the current environment, and determining a cross-power spectral density function R according to the cross-power spectral density function12(λ, w) calculating the signal-to-noise ratio weighting function
Step 213, calculating the cross-power spectral density function R12(λ, w) and said signal-to-noise ratio weighting function
And inverse fourier transforming said product to generate said acoustic signal cross-correlation function r12(λ,w)。
4. The sound source localization method according to any one of claims 1 to 3, wherein the calculating of the transmission delay of the second sound signal corresponding to the reference sound sensor specifically includes:
step 221, orderly grouping the sound sensors in pairs, and recording the sound sensors as the reference sound sensors;
step 222, for each group of reference sound sensors, the sound signal cross-correlation function r12Carrying out peak value detection on the value sequence of (lambda, w) to obtain a sampling point corresponding to the maximum discrete value point;
step 223, multiplying the sampling point by the sampling point interval time to generate the predicted time delay of the second sound signal corresponding to the reference sound sensor;
and 224, repeatedly executing the steps 221 to 224 for each group of reference sound sensors by adopting a loop traversal algorithm, calculating the average value of the predicted time delay, and recording the average value as the transmission time delay.
5. The sound source localization method based on snr weight optimization updating of any of claims 1 to 3, wherein the localization information of the sound source at least comprises: the azimuth angle of the sound source is theta, the pitch angle of the sound source is mu, and the distance from the sound source to each sound sensor is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011438854.XA CN112540346A (en) | 2020-12-07 | 2020-12-07 | Sound source positioning method based on signal-to-noise ratio weight optimization updating |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011438854.XA CN112540346A (en) | 2020-12-07 | 2020-12-07 | Sound source positioning method based on signal-to-noise ratio weight optimization updating |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112540346A true CN112540346A (en) | 2021-03-23 |
Family
ID=75019987
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011438854.XA Pending CN112540346A (en) | 2020-12-07 | 2020-12-07 | Sound source positioning method based on signal-to-noise ratio weight optimization updating |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112540346A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102854494A (en) * | 2012-08-08 | 2013-01-02 | Tcl集团股份有限公司 | Sound source locating method and device |
| JP2017058286A (en) * | 2015-09-17 | 2017-03-23 | 一般財団法人電力中央研究所 | Method for specifying fault part of semiconductor power converter, device and program for specifying fault part, and semiconductor power converter including fault part specification function |
| CN110082725A (en) * | 2019-03-12 | 2019-08-02 | 西安电子科技大学 | Auditory localization delay time estimation method, sonic location system based on microphone array |
| CN111025233A (en) * | 2019-11-13 | 2020-04-17 | 阿里巴巴集团控股有限公司 | Sound source direction positioning method and device, voice equipment and system |
-
2020
- 2020-12-07 CN CN202011438854.XA patent/CN112540346A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102854494A (en) * | 2012-08-08 | 2013-01-02 | Tcl集团股份有限公司 | Sound source locating method and device |
| JP2017058286A (en) * | 2015-09-17 | 2017-03-23 | 一般財団法人電力中央研究所 | Method for specifying fault part of semiconductor power converter, device and program for specifying fault part, and semiconductor power converter including fault part specification function |
| CN110082725A (en) * | 2019-03-12 | 2019-08-02 | 西安电子科技大学 | Auditory localization delay time estimation method, sonic location system based on microphone array |
| CN111025233A (en) * | 2019-11-13 | 2020-04-17 | 阿里巴巴集团控股有限公司 | Sound source direction positioning method and device, voice equipment and system |
Non-Patent Citations (1)
| Title |
|---|
| 张坤: "基于嵌入式的麦克风阵列声源定位系统算法研究", 《中国优秀硕士学位论文全文数据库（电子期刊） 信息科技辑》, no. 01, pages 27 - 28 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107644650B (en) | Improved sound source positioning method based on progressive serial orthogonalization blind source separation algorithm and implementation system thereof | |
| CN111429939B (en) | Sound signal separation method of double sound sources and pickup | |
| CN111445920B (en) | Multi-sound source voice signal real-time separation method, device and pickup | |
| CN111044973B (en) | MVDR target sound source directional pickup method for microphone matrix | |
| CN102103200A (en) | Acoustic source spatial positioning method for distributed asynchronous acoustic sensor | |
| CN102565759B (en) | Binaural sound source localization method based on sub-band signal to noise ratio estimation | |
| CN105388459B (en) | The robust sound source space-location method of distributed microphone array network | |
| CN106373589B (en) | A kind of ears mixing voice separation method based on iteration structure | |
| CN109188362B (en) | Microphone array sound source positioning signal processing method | |
| CN107167770B (en) | A kind of microphone array sound source locating device under the conditions of reverberation | |
| Liu et al. | Continuous sound source localization based on microphone array for mobile robots | |
| WO2020024816A1 (en) | Audio signal processing method and apparatus, device, and storage medium | |
| CN110534126B (en) | Sound source positioning and voice enhancement method and system based on fixed beam forming | |
| CN110726972B (en) | Voice sound source positioning method using microphone array under interference and high reverberation environment | |
| CN105976827B (en) | A kind of indoor sound localization method based on integrated study | |
| CN103278801A (en) | Noise imaging detection device and detection calculation method for transformer substation | |
| CN105679329A (en) | Microphone array voice enhancing device adaptable to strong background noise | |
| CN111798869B (en) | Sound source positioning method based on double microphone arrays | |
| CN112363112B (en) | Sound source positioning method and device based on linear microphone array | |
| CN112394324A (en) | Microphone array-based remote sound source positioning method and system | |
| CN112270915A (en) | Active noise reduction method for indoor space | |
| CN112540346A (en) | Sound source positioning method based on signal-to-noise ratio weight optimization updating | |
| Nakano et al. | Automatic estimation of position and orientation of an acoustic source by a microphone array network | |
| CN110838303A (en) | Voice sound source positioning method using microphone array | |
| Firoozabadi et al. | Combination of nested microphone array and subband processing for multiple simultaneous speaker localization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |