KR101700424B1 - Device and method for detecting abnormal signals to a speaker - Google Patents

Abstract

The present invention relates to a device for detecting a speaker abnormal signal of a broadcasting system capable of clearly transmitting sound information by preventing skewed broadcasting due to malfunction of speaker, the device including an audio system (10), a power amplifier (20), a speaker (30), and a microphone (40) receiving an output signal. The device for detecting a speaker abnormal signal comprises: a level detector (120) detecting an indoor noise signal, a signal received to the microphone, and an audio output signal to detect whether broadcasting is performed or not; a first FFT envelope detector (130) detecting information on a frequency of the audio output signal and a change in the frequency information in a broadcasting state; a second FFT envelope detector (140) detecting information on a frequency of a signal received to the microphone and a change in the frequency information; a noise envelope detector (160) copying the signal of the second FFT envelope detector (140) as a noise signal in a non-broadcasting state in the level detector (110) and storing the signal; a first synchronization buffer (150) for synchronizing a signal of the second FFT envelope detector (140); a second synchronization buffer (170) for synchronizing a signal of the noise envelope detector (160); and a correlation calculator (180) calculating a correlation between an audio output signal (x(t)) and a speaker output signal (y(t)) with reference to information of the first and second FFT envelope detectors (120, 140) and a signal of the noise envelope detector (160).

Description

TECHNICAL FIELD [0001] The present invention relates to an apparatus and a method for detecting a speaker distortion signal,

The present invention relates to detection of a speaker distortion signal, and more particularly to an apparatus and method for detecting an output error of a speaker in real time.

Particularly, according to the present invention, a failure of a speaker is determined by comparing a correlation between a signal of an audio equipment and a speaker output signal by frequency.

In addition, noise information generated in a listening space is referred to together so that it can be used even in a noisy environment.

In general, a broadcasting system is a one-to-many sound communication method, and is used when a plurality of speakers installed in a near or remote place are controlled by a single management system.

That is, the broadcasting system is installed in a building having a plurality of divided spaces such as a hospital, a school, a performance hall, and an apartment, and transmits sound signals related to announcements such as announcements and emergency situations through a plurality of speakers.

As an example, a broadcasting system according to the prior art is disclosed in the background art of Korean Patent Registration No. 10-0923488 (hereinafter referred to as Document 1).

That is, the broadcasting system 10 disclosed in the above document 1 includes a broadcasting control device 12, an output selecting device 14, an audio amplifier device 16, and a speaker device 18 as shown in Fig. The broadcast control unit 12 controls overall broadcasting, receives or generates an audio signal from the outside, and outputs the audio signals to the output selection device 14. [ Further, the output selection device 14 selects and controls the audio signals input from the broadcast control device 12, and outputs the selected audio signals to the audio amplifier device 16. The audio amplifier unit 16 amplifies the input audio signal and outputs the amplified audio signal to the speaker unit 18 provided in each of the broadcasting areas A1, A2, and A3, And emits the input audio signal to each broadcasting area ('A1', 'A2', 'A3').

However, in order for the broadcasting through the speaker to exhibit its function, it is very important to always keep normal operation by detecting in real time whether the internal circuit for operating the amplifier (AMP) and the speaker for sound output during broadcasting is in failure.

That is, the prior art disclosed in the above-mentioned document has a problem that it is difficult to accurately analyze whether the broadcasting line is broken when the speaker line is separated from the broadcasting room or a remote place, or when a plurality of speakers are operated at the same time.

In other words, an output error of a speaker in a broadcasting system can be accurately determined by analyzing both the acoustical listening due to the failure of the power amplifier and the speaker line, the cause of the disconnection and short circuit of the broadcasting line, and the signal level of the speaker line.

Therefore, in order to solve the problem according to the first prior art as described above, the present applicant (Garak Electronics Co., Ltd.) has filed and filed a Korean Patent Registration No. 10-1469064 (hereinafter referred to as "Document 2") .

That is, in the broadcasting system that conducts announcement broadcasting through a speaker installed in a nearby or remote place, the output state and the line state of the broadcasting line are recorded in real time The present invention relates to a real-time analysis system and method for a broadcast circuit capable of real-time analysis and protection of a failure of a broadcast line.

2 is a block diagram of a real-time analysis system for a broadcast line according to the second prior art. FIG. 4 is a detailed view of an output signal control module applied to a real-time analysis system of a broadcasting line according to the second prior art and its peripheral configuration, and FIG. to be.

2 to 4, the real-time analysis system for a broadcasting line according to the second prior art includes an input signal control module 100, a power amplifier 175, an output signal transmission module 200, And a read module 300.

As shown in FIGS. 2 and 3, the input signal control module 100 is a means for transferring a plurality of audio input signals and audio electric signals to an external power amplifier 175, A signal generator 120, an input signal gain controller 130, an input selector MUX 140, and an input signal detector 150. The signal input unit 110, the signal generator 120, the input signal gain controller 130,

Specifically, the audio signal input unit 110 is a sound source transmitting means such as a wired / wireless microphone, a CD player, and a tuner for transmitting audio signals related to announcements, such as performances, announcements and emergency situations.

The signal generator 120 generates an electrical signal for sound transmission to an external speaker.

The input signal gain controller 130 converts an electric signal output from the signal generator 120 into a digital signal and adjusts the gain and outputs the digital signal.

The MUX 140 selects and outputs an audio signal output from the audio signal input unit 110 and a signal output from the signal generator 120 and the input signal gain controller 130.

3, the input signal detecting unit 150 detects the magnitude of a signal output to the power amplifier 175. The input signal detecting unit 150 includes a first buffer 151 And a second buffer 153. The first buffer 151 is connected to a power amplifier 175 for outputting an audio signal and the second buffer 153 is connected to an AC The signal passed through the smoothing circuit 155 is sent to an analog-to-digital converter (ADC) 157, which converts the audio signal into a digital signal To be transmitted to the detection and reading module 300.

The input signal control module 100 includes a communication interface 160 and the detection and reading module 300, the input signal gain control unit 130, the input selection unit (MUX) 140, and the ADC 157, To the communication.

A plurality of power amplifiers 175 are connected to an output terminal of the input signal control module 100 to amplify and output a signal output from the input signal control module 10.

2 and 4, the output signal control module 200 controls the output signal control module 200 to output the audio signal and the electric signal output through the power amplifier 175 to a speaker 205 line near or at a remote place And a speaker line connection unit 210 and an output signal selection unit 220. The speaker line connection unit 210 and the output signal selection unit 220 are connected to each other.

Specifically, the speaker line connection unit 210 is a means for connecting or disconnecting a plurality of speakers 205, which are located near or remote from the plurality of power amplifiers 175, so as to be able to detect the state of audio broadcasting or lines .

The output signal selector 220 selects one of the output signals from the speaker line connection unit 210 to detect an output signal of the speaker line 205 according to the connection of the speaker line connection unit 210 4, the sensor 230 connected to the output terminal of the output signal selector 220 is used to detect the current flowing through the speaker line, The detected current is converted into a digital signal through an analog-to-digital converter (ADC) 240 and then transmitted to a broadcast line condition analyzer 330 in the detection and reading module 300.

At this time, the selected speaker 205 line according to the connection of the output signal selection unit 220 receives the sound signal through the output signal listening unit 324 and the status check speaker 325 in the detection and reading module 300 And the impedance matching unit 322 in the detecting and reading module 300 is composed of a matching control switch and a matching transformer so that the impedance of the speaker 205 can be adjusted in consideration of the impedance of the signal output from the output signal control module 200. [ So that the signal can be detected without affecting the output signal of the line.

Meanwhile, the detection and reading module 300 analyzes the output signal transmitted through the output signal transmission module 200 to display the presence or absence of an abnormality in the power amplifier 175 and the speaker 205, The input signal selection control unit 310, the detection signal collection unit 312, the input level detection and conversion unit 314, the input comparison unit 316, the output signal selection control unit 318, the output signal reception control unit The output level detecting unit 326, the broadcast line condition analyzing unit 330, the recording unit 332, the speaker output preventing unit 322, the output signal receiving unit 324, the status checking speaker 325, the output level detecting unit 326, A broadcast line status display unit 336, a network data transfer unit 338, an algorithm storage unit 339, and a controller 340. [

Specifically, the input signal selection control unit 310 controls the input signal gain control unit 130 and the input selection unit (MUX) 140 of the input signal control module 100.

The detection signal collection unit 312 receives a signal output through the input signal detection unit 150 of the input signal control module 100.

The input level detection and conversion unit 314 converts the input signal detected through the detection signal collection unit 312 into a digital level and outputs the digital level.

The input comparator 316 compares the signal output from the input level detecting and converting unit 314 with a predetermined steady state signal to control the input signal gain controller 130 in the input signal control module 100 do.

The output signal selection control unit 318 controls the speaker line connection unit 210 and the output signal selection unit 220 of the output signal control module 200 so that one of the output signals from the output signal selection unit 220 .

The output signal reception control unit 320 receives the signal output from the output signal selection unit 220 of the output signal control module 200 and outputs an acoustic listening signal indicating whether the speaker 205 is abnormal or not, .

The impedance matching unit 322 detects the signal without affecting the output signal in consideration of the impedance of the signal output from the output signal reception control unit 320.

The output signal receiving unit 324 allows the acoustic signal output through the impedance matching unit 322 to be output through the status checking speaker 325.

The output level detector 326 converts the output signal from the impedance matcher 322 into a digital signal level.

The broadcast line condition analyzing unit 330 receives the input signal level from the input level detecting and converting unit 314 and the output signal level from the output level detecting unit 326 and outputs the state of the power amplifier 175 and the speaker (205) Analyze the condition of the track.

The recording unit 332 stores the state of the speaker 205, the impedance state thereof, and the output signal state from the output signal control module 200.

The speaker output preventing unit 334 cancels the connection of the speaker line connection unit 210 when there is an error in the line of the speaker 205 referring to the state of the recording unit 332. [

The broadcast line state display unit 336 displays the states of the power amplifier 175 and the speaker 205 on the outside by referring to the state of the recording unit 332.

The network data transmission unit 338 transmits detection contents according to the state of the recording unit 332 to the web and the smart phone through the network and supports remote control.

The algorithm storage unit 339 stores a program capable of analyzing the state of the broadcasting line in real time according to the control of the input signal control module 100, the output signal control module 200, and the detection and reading module 300 do.

The controller 340 controls the functions of the units 310 to 339 in the detection and reading module 300 to analyze the state of the broadcasting line in real time.

Therefore, according to the second prior art, the output state and the line state of the broadcasting line including the power amplifier and the speaker line are heard and sensed in real time, and the failure state of the broadcasting line is analyzed in real time. It is possible to easily confirm the output state of a plurality of speaker lines and to operate a more stable broadcasting system.

On the other hand, nonetheless, there is a case where an audio signal which is not clear due to the output distortion of the speaker is transmitted from the audio enhancing device. The distortion of the output of the speaker is considered to be caused by the sound equipment, the error of the transmission line of the speaker, and the failure of the speaker equipment itself. Errors in the sound equipment or the speaker transmission line are easy for the user to examine. However, in the case of a failure of the speaker equipment itself, when the sound equipment and the speaker are separated from each other, the operator can not check the error of the speaker in real time and transmits the distorted signal to the distorted listener.

Such a problem is a cause of hindering the purpose of use in equipment requiring clear sound signal output such as conference equipment, performance equipment, public hall broadcasting equipment and the like.

Document 1: Korean Patent Registration No. 10-0923488 (Registered on October 19, 2009) Document 2: Korean Patent Registration No. 10-1469064 (Registered on Nov. 28, 2014)

An object of an apparatus and a method for detecting a speaker distortion signal according to the present invention is to prevent a distorted broadcast due to a failure of a speaker itself, thereby securing clear sound information transmission.

Therefore, it should be possible to use it for conference broadcasting, school broadcasting, public announcement broadcasting, and public broadcasting equipment which need clear broadcasting, but especially for the outer guard which needs to examine whether the distributed speaker is broken or not, I think that utilization is high.

These and other objects of the present invention will be apparent to those skilled in the art without departing from the scope of the present invention by the appended claims.

In order to achieve the above object, an apparatus for detecting a speaker distortion signal according to the first aspect of the present invention includes an audio system (10) for pre-amplifying an audio signal or adding a sound effect to output it as an audio output signal x A power amplifier 20 for amplifying the audio output signal x (t) to a power amplifier output signal; a speaker 30 for outputting a loudspeaker signal; y (t)) (as a detection device of the speaker distortion signal of a broadcast system comprises a 40), of the room noise signal (n (t) microphone for sound receiver), the signal (n (t) is received sound to the microphone + y (t) ) and an audio output signal ( x (t) ), respectively, to detect whether the signal is in a broadcast state or in a non-broadcast state; A first FFT envelope detector 130 for detecting frequency information of the audio output signal x (t) in a broadcast state and a change of the frequency information; A second FFT envelope detector 140 for detecting frequency information of the signal n (t) + y (t) received by the microphone and a change of the frequency information; A noise envelope detector 160 for copying and storing a signal of the second FFT envelope detector 140 as a noise signal when the level detector 110 is in a non-broadcasting state; A first synchronization buffer (150) for signal synchronization of the second FFT envelope detector (140); A second synchronization buffer (170) for signal synchronization of the noise envelope detector (160); (T ) and the speaker output signal y (t) by referring to the information of the first and second FFT envelope detectors 120 and 140 and the noise envelope detector 160 signal, A correlation calculator 180 for calculating the degree of correlation of the input signal; And a control unit.

Preferably, the level detector 120 receives the noise signal n (t ), the signal n (t) + y (t) received by the microphone, the audio output signal x (t ) in further comprising an ADC (110), each digital converting a signal, is smaller than the moving average value (ℓ (t)) the extracted moving average value (ℓ (t)) is obtained for a particular threshold value (ℓ _th) noise signal state And if it is not, it is determined as a broadcast signal.

) 및 표준 편차(S_x(f))를 구하고, 상기 제2 FFT 엔벨롭 검출기(140)는 마이크로폰 입력 신호( n(t) + y(t) )에서 주파수 별 평균값(

) 및 표준 편차(S_yn(f))를 구하고, 상기 잡음 엔벨롭 검출기(160)는 잡음 신호( n(t) )에서 평균값(

)를 계산하며, 상기 상관도 계산기(180)가, 상기 평균값들로부터 스피커의 출력 신호(y(t))의 추정되는 평균값(

)을 하기 수학식 (1)과 같이 구하고, 주파수에 의한 상관도(r_xy)를 하기 수학식 (2)와 같이 구하여, 상기 상관도(r_xy)가 기준값(r₀) 미만이면 왜곡이 있는 것으로 판단하는 것을 특징으로 한다.
Also, preferably, the first FFT envelope detector 130 subtracts the frequency-averaged value ( x (t) ) from the audio output signal x

) And a standard deviation S _x (f), and the second FFT envelope detector 140 obtains a frequency-averaged value ( ) from the microphone input signal n (t) + y (t)

) And standard deviation (S _yn (f) obtaining a), the noise envelope detector 160, the noise signal (n (t) from the average value) (

, And the correlation calculator 180 calculates an estimated average value of the output signal y (t) of the speaker from the average values

(R _xy ) is found by the following equation (1), and the correlation degree r _xy according to the frequency is obtained by the following equation (2). If the correlation degree r _xy is less than the reference value r ₀ , .

On the other hand, the detection method of the speaker distortion signal according to the second aspect of the present invention for achieving the above object is, (a) an audio output signal (x (t)) detected at the output of the audio system 10 and the speaker ( 30) microphone 40. the speaker output signal (y (t detected in)) and a noise signal (n (t)), the sum a microphone input signal (n (t) + y (t) of a sound receiver an output signal) Converting each of the digital signals into a digital signal (S21); (b) detecting a level value of the digital signal by the level detector 120 (S22); (c) extracting a moving average value ? (t) from a level value of the digital signal detected in the step (b) as a moving average; (d) comparing the moving average value ? (t) extracted in the step (c) with a specific threshold value ? _th (S24); (e) the judgment made at step (d) is smaller than the extracted value (ℓ (t)) a specific threshold value (ℓ _th) noise signal state and (S26), or determining the broadcast signal (S25) ; (f) a step (S27) of extracting the moving average value ( t (t) ) and FFT-transforming and collecting frequency information of the signal (S27); (h) extracting a change of each frequency by an envelope (S29); And (k) obtaining a degree of correlation from the envelope of each frequency-dependent change (S34); And a control unit.

(G) facilitating (S28) the analysis of the frequency information of the FFT-transformed signal in step (f) with ease; And further comprising:

In addition, preferably, the (h) step and the (k) between the steps, (i) extracting the yen along the envelope information on the state of the logic signal value (b) related to the moving average value (ℓ (t)), signal (S30) to the synchronization buffer 150 or the noise synchronization buffer 170; And (j) obtaining a frequency-based average value and a standard deviation using the past values and the current values of the signal synchronization buffer 150 and the noise synchronization buffer 170, S32 and S33; And the degree of correlation is calculated based on the frequency-based average value and the standard deviation.

) 및 표준 편차(S_x(f))를 구하고, 마이크로폰 입력 신호( n(t) + y(t) )에서 주파수 별 평균값(

) 및 표준 편차(S_yn(f))를 구하며, 잡음 엔벨롭 검출기(160)는 잡음 신호( n(t) )에서 평균값(

)를 계산하는 단계; (k2) 상기 (k1) 단계에서의 평균값 및 표준편차로부터 스피커의 출력 신호(y(t))의 추정되는 평균(

) 을 하기 수학식 (1)로부터 구하는 단계; 및 (k3) 상기 (k1) 및 (k2) 단계에서의 값들로부터 하기 수학식 (2)에 의해 주파수에 의한 상관도(r_xy)를 구하는 단계; 로 이루어지며, 상기 상관도(r_xy)가 기 설정된 기준값(r₀) 미만이면 스피커의 출력에 왜곡이 있는 것으로 판단하는 것을 특징으로 한다.More preferably, the (k) comprises: a microphone 4 in the speaker output signal (y (t)) and a noise signal (n (t)), the sum a microphone input signal (n (t) of the input + y ( t)) and the audio output signal (x (t)) being dangyeyi calculating a correlation by the frequency of, (k1) at the average value per frequency audio output signal (x (t)) (

) And standard deviation (S _x (f)), a microphone input signal (n (t), obtains the + y (t)) from the frequency-dependent mean value (

) And a standard deviation S _yn (f), and the noise envelope detector 160 obtains an average value ( n (t) ) from the noise signal n

); (k2) Estimated average of the output signal y (t) of the speaker from the average value and the standard deviation in the step (k1)

) From the following equation (1); And (k3) obtaining a correlation degree (r _xy ) by frequency from the values at steps (k1) and (k2) by the following equation (2); And determines that the output of the speaker is distorted if the correlation degree (r _xy ) is less than a preset reference value (r ₀ ).

…………………(1)

... ... ... ... ... ... ... (One)

Where a is the gain by the acoustic system and the power amplifier.

…………………(2)

... ... ... ... ... ... ... (2)

Where n is the number of frequencies sampled in the FFT.

According to the apparatus and method for detecting a speaker distortion signal according to the present invention, it is possible to prevent a distorted broadcast due to a failure of the speaker itself, thereby ensuring clear sound information transmission.

Furthermore, the technology of the present invention can be applied to conference broadcasting, school broadcasting, public announcement broadcasting, and public broadcasting equipment that require clear broadcasting, and in particular, It is highly utilized in evacuation equipment for evacuation, and it can be ensured that the evacuation equipment has clear clarity.

Additional features and advantages of the present invention will become more apparent from the following description.

1 is a diagram showing an example of a broadcasting system according to a first prior art;
2 is a diagram showing the overall configuration of a real-time analysis system for a broadcasting line according to a second prior art.
3 is a diagram illustrating a detailed configuration of an input signal control module applied to a real-time analysis system of a broadcasting line according to the second conventional technique.
4 is a detailed view showing an output signal control module applied to a real-time analysis system of a broadcasting line according to the second prior art and its peripheral configuration.
5 is a block diagram of a broadcasting system including an apparatus for detecting a speaker distortion signal according to a preferred embodiment of the present invention.
6 is a systematic flow diagram of an FFT envelope detector of a method for detecting a speaker distortion signal according to an embodiment of the present invention;
7 is a flowchart of a method for determining a speaker failure according to the correlation diagram of the present invention.

Hereinafter, an apparatus and method for detecting a speaker distortion signal according to a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.

Prior to the detailed description of the present invention, the same or corresponding components in the drawings are denoted by the same reference numerals, and if a detailed description of related known configurations or functions is deemed to blur the gist of the present invention, Is omitted.

Also, in the present specification, the following embodiments and embodiments are to be considered as being illustrative and not restrictive, and the invention is not to be limited to the details given herein but is to be accorded with the scope of the appended claims and their equivalents, It can be changed to another embodiment.

FIG. 5 is a block diagram of a broadcasting system including an apparatus for detecting a speaker distortion signal according to an exemplary embodiment of the present invention, FIG. 6 is a block diagram of a system of an FFT envelope detector in a method for detecting a speaker distortion signal according to an embodiment of the present invention. And FIG. 7 is a flowchart of a method for determining a speaker failure according to the correlation diagram of the present invention.

(Apparatus for Detecting Speaker Distortion Signal of Optimal Embodiment of Present Invention)

First, an apparatus for detecting a speaker distortion signal according to a preferred embodiment of the present invention will be described with reference to FIG.

The broadcasting system relating to the apparatus for detecting a speaker distortion signal of the present invention is an audio system (a speaker system) for pre-amplifying and / or adding a sound effect to an audio output signal x (t) A power amplifier 20 for amplifying the audio output signal x (t) to a power amplifier output signal, a speaker 30 for outputting an amplification signal, A microphone 40 for receiving an output signal y (t ) of the speaker 30, and an apparatus 100 for detecting a speaker distortion signal according to the present invention.

The apparatus 100 for detecting a speaker distortion signal according to the present invention is a device for detecting a speaker distortion signal in accordance with the present invention which detects a noise signal n (t) in a room, a signal n (t) + y (t) by detecting a signal x (t)) shows the state that the non-broadcast status that the detection level of the detector changes in the frequency information and the frequency information) (level detector) (120) , the audio output signal (x (t on the air condition), which A second FFT envelope detector 130 for detecting the signal n (t) + y (t) received by the microphone, a first FFT envelope detector 130 for detecting a level detector 110 A noise envelope detector 160 for storing a signal of the second FFT envelope detector 140 as a noise signal and storing the copied signal as a noise signal and a second FFT envelope detector 140 for signal synchronization, A second synchronization buffer 1 for signal synchronization of the noise envelope detector 160, (T ) and the speaker output signal y (t (t) ) by referring to the information of the first and second FFT envelope detectors 120 and 140 and the noise envelope detector 160, , A correlation calculator 180 for calculating a correlation of the input image data, and a display unit 190 for displaying a calculation result.

(Method for discriminating the speaker distortion signal in the optimum embodiment of the present invention)

Now, with reference to Figs. 6 and 7, a method of detecting a speaker distortion signal will be described in detail.

In the present invention, first and second FFT envelope detectors 130 and 140 are provided for determining the noise signal or the broadcast output signal according to the level of the signal and calculating the mean and standard deviation of each frequency while extracting the frequency information for each signal. .

Specifically, the speaker output signal (y (t)) and a noise signal (n (t)) detected in the audio output signal (x (t)) detected at the output terminal and the microphone 40 of the audio system 10 each sum of the microphone input signal (n (t) + y (t)) is, while by the ADC (110) detecting the level value of the signal by (S21), the level detector 120 after being converted into a digital signal ( S22), and a value is extracted by a moving average (S23). If the extracted moving average value l (t) is compared with the specific threshold value l _{th in step} S24 and the extracted value l (t ) is smaller than the specific threshold value l _th , S26) or a broadcast signal (S25). The judged result is converted into a logic signal value (b) for control. That is, if the extracted value l (t) is smaller than the specific threshold value ? _Th , '0' is given to the logic signal value b (S26), and the extracted value ? (T) If it is larger than the specific threshold value ? _Th , '1' is given to the logic signal value b (S25).

Meanwhile, in the present invention, the moving average value ( t (t) ) is extracted and the frequency information of the signal is collected by the FFT 125 (S27). Then, the analysis is easily smoothed (S28), and the change of each frequency is extracted as an envelope (S29).

)과 표준 편차(standard deviation, S(f))를 구한다(S32,S33). 마지막으로, 구해진 값으로 상관도를 구한다(도 7의 S34).
The extracted envelope information is transmitted to the synchronization buffer 150 or the noise synchronization buffer 170 according to the state of the logic signal value b related to the moving average value l (t) (S30 ). Using the past and present values of each buffer,

) And a standard deviation S (f) are obtained (S32, S33). Lastly, correlation is obtained from the obtained values (S34 in Fig. 7).

) 및 표준 편차(S(f))로서 상관도(r_xy)를 구하고, 상관도에 의한 스피커 고장 판단 방법에 대하여, 도 7을 참조하여 상술한다. Now, the average value (

) And the standard deviation obtained for the correlation (r _xy) as the (S (f)), with respect to the speaker failure determination method according to the correlation, will be described with reference to FIG.

A method for detecting a failure of the speaker in the present invention, a microphone 4 in the speaker output signal (y (t)) and a noise signal (n (t)), the sum a microphone input signal (n (t) of the input + y ( If t)) and to calculate the correlation by the frequency of the audio output signal (x (t)) (S34 ), correlation is present the different frequencies is characterized by also determined arbitrarily speaker is failed.

) 및 표준 편차(S_x(f))를 구하고, 제2 FFT 엔벨롭 검출기(140)는 마이크로폰 입력 신호( n(t) + y(t) )에서 주파수 별 평균값(

) 및 표준 편차(S_yn(f))를 구하며, 잡음 엔벨롭 검출기(160)는 잡음 신호( n(t) )에서 평균값(

)를 계산한다. That is, the first FFT envelope detector 130 of FIG. 5 extracts the frequency-averaged value ( x (t) ) from the audio output signal x

) And a standard deviation S _x (f), and the second FFT envelope detector 140 obtains a frequency-averaged value ( ) from the microphone input signal n (t) + y (t)

) And a standard deviation S _yn (f), and the noise envelope detector 160 obtains an average value ( n (t) ) from the noise signal n

).

The estimated average of the output signal y (t) of the speaker is expressed by Equation (1) below,

Where a is the gain by the acoustic system and the power amplifier.

At this time, the coefficient of correlation (r _xy ) according to the frequency is expressed by Equation (2)

Where n is the number of frequencies sampled in the FFT.

The correlation (r _xy ) has a value of 1 or less due to the influence of noise. If it is 1, it is a perfect match. If it is -1, it is a conduction. If it is 0, it is uncorrelated.

) 와 평균값(

)가 상관도가 낮은 상태로 판단하며, 스피커의 출력에 왜곡이 있는 것으로 판단하고(S36), 이를 표시부(190)를 통해 디스플레이한다.
In the present invention, if the correlation (r _xy) the group is less than a reference value set (r ₀₎ and determines whether or less (S35), the correlation (r _xy) is a predetermined reference value (r ₀₎ the average value (

) And the average value

And determines that the output of the speaker is distorted (S36), and displays the distorted output through the display unit 190. [0052]

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, It should be understood that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

(Prior art)
100: input signal control module 110: audio signal input part
120: Signal generator 130: Input signal gain controller
140: input selector (MUX) 150: input signal detector
175: Power Amplifier 200: Output Signal Control Module
205: Speaker 210: Speaker line connection
220: output signal selector 300: detecting and reading module
310: input signal selection control unit 312: detection signal collection unit
314: input level detection and conversion unit 316: input comparison unit
318: Output signal selection control unit 320: Output signal reception control unit
322: Impedance matching unit 324: Output signal receiving unit
325: Status check speaker 326: Output level detector
330: Broadcast line condition analysis unit 332:
334: Speaker output preventing section 336: Broadcast line status display section
338: Network data transfer unit 339: Algorithm storage unit
340: controller
(Invention)
10: Audio System 20: Power Amplifier
30: Speaker 40: Microphone
100: a speaker distortion signal detecting device according to the present invention
110: ADC 120: Level detector
130: first FFT envelope detector 140: second FFT envelope detector
150: first synchronization buffer 160: noise envelope detector
170: Second synchronization buffer
180: Correlation calculator 190: Display

Claims (7)

And an audio system 10 amplifies preconditioning the audio signal, or add a sound effect and outputting the audio output signal (x (t)), a public address for amplifying the audio output signal (x (t)) to the power amplifier output signal A speaker 30 for outputting a loudspeaker signal and a microphone 40 for receiving an output signal y (t) of a loudspeaker 30, As a detection device,
The respective detection signals from the room noise signal (n (t)), the signal (n (t) + y ( t)) being received sound to the microphone, an audio output signal (x (t)) that the broadcasting state or a non-broadcast status A level detector 120 for detecting the level;
A first FFT envelope detector 130 for detecting frequency information of the audio output signal x (t) in a broadcast state and a change of the frequency information;
A second FFT envelope detector 140 for detecting frequency information of the signal n (t) + y (t) received by the microphone and a change of the frequency information;
A noise envelope detector 160 for copying and storing a signal of the second FFT envelope detector 140 as a noise signal when the level detector 120 is in a non-broadcasting state;
A first synchronization buffer (150) for signal synchronization of the second FFT envelope detector (140);
A second synchronization buffer (170) for signal synchronization of the noise envelope detector (160); And
(T ) and the speaker output signal y (t) by referring to the information of the first and second FFT envelope detectors 130 and 140 and the noise envelope detector 160, A correlation calculator 180 for calculating a correlation of the input signal;
And a speaker distortion signal detecting unit for detecting the speaker distortion signal. The method according to claim 1,
The level detector 120 outputs a signal of the indoor noise signal n (t) , a signal of the microphone n (t) + y (t) , and an audio output signal x (t) Further comprising an ADC (110) for digital conversion,
Moving average value (ℓ (t)) moving average value (ℓ (t)) extracted obtain a speaker distortion signal, characterized in that for determining a noise signal state is less than a certain threshold (ℓ _th), otherwise, to the broadcast signal . The method according to claim 1,
The first FFT envelope detector 130 multiplies the frequency-averaged value ( x (t) ) from the audio output signal x

) And the standard deviation S _x (f)
The second FFT envelope detector 140 receives the frequency-averaged mean value ( n (t) ) from the microphone input signal n (t) + y

) And standard deviation S _yn (f)
The noise envelope detector 160 calculates an average value ( n (t) ) from the noise signal n

),
The correlation calculator 180 calculates an average value of the output signal y (t) of the speaker from the average values

(R _xy ) is found by the following equation (1), and the correlation degree r _xy according to the frequency is obtained by the following equation (2). If the correlation degree r _xy is less than the reference value r ₀ , And the speaker distortion signal is detected.

... ... ... ... ... ... ... (One)
Where a is the gain by the acoustic system and the power amplifier.

... ... ... ... ... ... ... (2)
Where n is the number of frequencies sampled in the FFT. A method for detecting a speaker distortion signal,
(a) and the speaker output signal (y (t)) detected by the microphone 40 for sound receiver an output signal of the audio output signal (x (t)) and a speaker (30) is detected at the output of the audio system 10 Converting the microphone input signal n (t) + y (t) , which is the sum of the noise signals n (t) , into a digital signal (S21);
(b) detecting a level value of the digital signal by the level detector 120 (S22);
(c) extracting a moving average value ? (t) from a level value of the digital signal detected in the step (b) as a moving average;
(d) comparing the moving average value ? (t) extracted in the step (c) with a specific threshold value ? _th (S24);
(e) the judgment made at step (d) is smaller than the extracted value (ℓ (t)) a specific threshold value (ℓ _th) noise signal state and (S26), or determining the broadcast signal (S25) ;
(f) a step (S27) of extracting the moving average value ( t (t) ) and FFT-transforming and collecting frequency information of the signal (S27);
(h) extracting a change of each frequency by an envelope (S29); And
(k) obtaining a degree of correlation from the envelope of each frequency-dependent change (S34);
And detecting a distortion of the speaker distortion signal. 5. The method of claim 4,
(g) facilitating the frequency information of the signal FFT-transformed in step (f) to facilitate analysis (S28);
And detecting a distortion of the speaker distortion signal. 5. The method of claim 4,
Between the step (h) and the step (k)
(i) in accordance with the extracted envelope information to the state of the logic signal value (b) related to the moving average value (ℓ (t)), comprising: transmitting a signal synchronization buffer 150 or noise synchronization buffer 170 for (S30 ); And
(j) obtaining average values and standard deviations (S32 and S33) of the frequency using the past values and the current values of the signal synchronization buffer 150 and the noise synchronization buffer 170;
Further comprising:
And the correlation is found based on the frequency-based average value and the standard deviation. The method according to claim 6,
The step (k)
(T) + y (t) ) which is the sum of the speaker output signal y (t) and the noise signal n (t) input to the microphone 4 and the audio output signal x ) Of the frequency bands,
(k1) The frequency-based average value ( x (t) ) from the audio output signal

) And standard deviation (S _x (f)), a microphone input signal (n (t), obtains the + y (t)) from the frequency-dependent mean value (

) And a standard deviation S _yn (f), and the noise envelope detector 160 obtains an average value ( n (t) ) from the noise signal n

);
(k2) Estimated average of the output signal y (t) of the speaker from the average value and the standard deviation in the step (k1)

) From the following equation (1); And
(k3) obtaining a correlation degree (r _xy ) by frequency from the values at the steps (k1) and (k2) by the following equation (2);
Lt; / RTI >
And if the correlation degree (r _xy ) is less than a preset reference value (r ₀ ), it is determined that the output of the speaker is distorted.

... ... ... ... ... ... ... (One)
Where a is the gain by the acoustic system and the power amplifier.

... ... ... ... ... ... ... (2)
Where n is the number of frequencies sampled in the FFT.

Images