CN203552692U - Self-adapting denoiser for low and medium frequency noise in indoor environments - Google Patents

Abstract

The utility model discloses a self-adapting denoiser for low and medium frequency noise in indoor environments. The self-adapting denoiser comprises a primary microphone, a reverse sound source, a frequency adjusting device and a secondary microphone, wherein the secondary microphone is a high-sensitivity omnidirectional electret secondary microphone with frequency response of 20 HZ to 20 KHz; the primary microphone and the secondary microphone are connected with a 16 bit analog-to-digital converter respectively through audio cables, the 16 bit analog-to-digital converter is connected with an I/O reading-writing parallel port of the frequency adjusting device through a 16 pin parallel port, and the reverse sound source is connected with the MCU (Microprogrammed Control Unit) processor of the frequency adjusting device through a serial digital-to-analog converter. The product is particularly suitable for processing low frequency noise in household environments and is capable of significantly reducing the low frequency noise generated by electronic and electric equipment indoors, in order to perfect the living environment and improve the living quality. The equipment is small in volume and low in cost, the system can be used once being electrified, and no complicated installation process is needed.

Description

A kind of adaptive noise reduction device for indoor environment medium and low frequency noise

Technical field

The utility model belongs to electronics active noise reduction technical field, specifically relates to a kind of adaptive noise reduction device that has remarkable noise reduction for the low frequency noise in indoor environment.

Background technology

Low frequency noise pollutes large public hazards that become urban environment.Scientific research shows that low frequency noise can impair one's health, and people works long hours, lives in low frequency noise environment, large to the stimulation of central nervous system, and severe patient can cause central nervous system function disorderly.In addition, low frequency noise can cause that sleep is bad, absent minded, and the symptoms such as decrease of memory, then cause mood confusing, emotional lability, and patience reduces, short-tempered, and even produces a series of diseases such as hypertension, ulcer, diabetes.What cause due to low frequency noise is phonosensitive nerve damage and damage of hair cell, is difficult to it to repair, and does not also have at present specific medicament can treat this damage.

At present, in family room, the common processing means of noise reduction are to adopt the passive materials such as sound-absorbing, sound insulation to carry out coating processing to indoor wall, door and window, but the acoustic attenuation meeting of sound-absorbing, sound-proof material reduces and variation with frequency, passive noise control mechanism centering, high band noise are comparatively effective, and little to the effect of low frequency noise.

For " active " of low frequency noise, controlling (Active Noise Control:ANC) concept was proposed by roentgen P.Lueg the thirties, its mixing according to the relevant sound wave of two row can produce the physical phenomenon of interference, artificial stack one row equate and the sound wave of single spin-echo with original magnitudes of acoustic waves, utilize the destructive interference of sound wave to reach the object of eliminating and reducing noise, by sound source, produce secondary sound field and carry out the control of noise noise reduction.Yet because technology realizes, this technology has been abandoned nearly 30 years.Along with the fast development of electronic information technology, active control technology has progressively entered electric equipment products field, and the noise reduction process from industrial premises noise control to central type air-conditioning equipment all part has been introduced ANC technology.Take domestic refrigerator as example, its noise source be mainly casing by totally-enclosed compressor to external diffusion, the people such as the Saruta of Toshiba Corp utilize active technique and resistance noise reduction to combine the noise of household electric refrigerator to be carried out to experimental study, obtain noise reduction 7dB effect; The D S Smith of U.S. Americold company utilizes similar method, with acceleration sensor, detects compressor housing vibrations, with microphone, detects remaining noise, and the anti-noise of driver output, in the noise reduction of the indoor test result acquisition 5dB of small-sized semireverberation.In recent years, also there is the noise reducing apparatus based on ANC in market at home, such as the NANTY series noise controller of Nanjing zero shellfish acoustics Science and Technology Ltd., the agency's of Beijing CA Acoustics Co., Ltd. S-cube series noise reduction controller etc.

What this class specialty noise reducing apparatus was mainly faced is the industrial environments such as Aero-Space, automobile making, large-scale instrument, and equipment is installed complicated, bulky and cost is high, and general user's ability is not purchased and disposed; Another part product only carries out inner noise reduction process for electric equipments such as air-conditioning, lampblack absorber or refrigerator, belongs to the additional function of independent electric equipment.Not only noise reduction is poor, and cannot realize the centralized noise reduction process of environment within doors, cannot reduce the low frequency noise that indoor multiple electronic and electrical equipment sends.

Summary of the invention

The purpose of this utility model is for the deficiencies in the prior art, and a kind of adaptive noise reduction device that can effectively reduce indoor environment medium and low frequency noise is provided, and concentrates the low frequency noise pollution problem solving in home environment with cheap cost.

The purpose of this utility model is achieved by following technical proposals.

A kind of adaptive noise reduction device for indoor environment medium and low frequency noise, comprise elementary microphone, anti-phase sound source and frequency regulation arrangement, it is characterized in that: also comprise a secondary microphone, described secondary microphone is that the high sensitivity of frequency response 20HZ～20KHz is pointed to the secondary microphone of electret entirely; Elementary microphone is connected with 16 A/D converters by audio cable respectively with secondary microphone, 16 A/D converters are connected with the I/O read-write parallel port of frequency regulation arrangement by 16 pin parallel port buses, and anti-phase sound source is connected with the MCU processor of frequency regulation arrangement by serial A/D converting device.

Described elementary microphone and secondary microphone are all used two NPN type triodes to form cascode collection secondary pre-amplification circuit, guarantee undamped transmission of audio signal.

Described frequency regulation arrangement consists of the outer random access memory (SRAM) of a slice CPLD (CPLD), 32 microprocessing units of a slice (MCU) and a slice.

Described anti-phase sound source adopts 20HZ～20KHz full range signal concurrent coaxial loudspeaker.

Principle of work of the present utility model: elementary microphone gathers the sound signal of low frequency noise sound source, exports anti-phase sound source to by frequency regulation arrangement, forms noise reduction audio signal.Secondary microphone gathers the audio frequency error signal that the real-time noise-reducing in sound field is processed, and feeds back to frequency regulation arrangement, forms the compensation sound source to original low frequency noise sound source, makes space environment interior low frequency noise value in airtight house lower than the value of largest tolerable.

With respect to prior art, the utlity model has following advantage:

1, the low frequency noise that the utility model is specially adapted under domestic environment is processed, and guarantees robustness and the applicability of equipment, and volume is small and exquisite, and system powers on and can use, without complicated installation process.Meanwhile, device fabrication is cheap, and general user has the ability to purchase and dispose completely.

2, the utility model can be realized the centralized noise reduction process of environment within doors, reduces preferably the low frequency noise that electronics and electric equipment send within doors, improves living environment, improves the quality of living.

Accompanying drawing explanation

Fig. 1 is the system principle diagram of the utility model low frequency denoiser;

Fig. 2 is the control principle drawing of the utility model low frequency denoiser;

Fig. 3 is the structural representation of the utility model low frequency denoiser;

Fig. 4 is the program flow diagram of control system;

Fig. 5 is enclosure space noise source sampled data figure, and black curve is the noise data while being not activated self-adaptation low frequency denoiser, 300 seconds time spans that coordinate transverse axis is sampling (unit: second), vertical pivot is confined space sound value (unit: decibel);

Fig. 6 is enclosure space noise reduction sampled data figure, 300 seconds time spans that coordinate transverse axis is sampling (unit: second), vertical pivot is confined space sound value (unit: decibel);

Fig. 7 is enclosure space noise reduction comparison diagram, 300 seconds time spans that coordinate transverse axis is sampling (unit: second), vertical pivot is confined space sound value (unit: decibel).Solid black lines is the sampled data that in confined space, noise source produces, and dotted line is sound sampled data in the confined space after noise reduction process.

Embodiment

Below in conjunction with drawings and Examples, the utility model is described in further detail, but they are not to restriction of the present utility model.

Embodiment 1

Terminological interpretation:

A/D(analog to digital converter) be simulating signal to be converted to the circuit of digital signal, be called analog to digital converter, be called for short A/D converter.

D/A(digital to analog converter) be digital signal to be converted to the circuit of simulating signal, be called digital to analog converter, be called for short D/A converter.

CPLD(Complex Programmable Logic Device) be a kind of user according to needing separately the digital integrated circuit of constitutive logic function voluntarily, the abbreviation of Complex PLD, be a kind of compared with PLD, be complicated logic element.

MCU(Micro Controller Unit) be a kind of integrated circuit (IC) chip, adopt very large scale integration technology that central processing unit, random access memory, multiple input/output interface and the interrupt system with data-handling capacity are integrated into the little and perfect computer system forming on a silicon chip.

JTAG(Joint Test Action Group; Combined testing action group) be a kind of international standard test protocol (IEEE1149.1 is compatible), be mainly used in chip internal test.The jtag interface of standard is 4 lines: TMS, TCK, TDI, TDO, is respectively model selection, clock, data input and DOL Data Output Line.

As shown in Figure 1, the principle of the utility model based on " with sound noise reduction ", in the indoor environment of relative closure, utilize electronic equipment to gather the sound pressure level of original noise sound field, utilize frequency regulation arrangement in equipment to produce a sound wave with original acoustic pressure equal and opposite in direction, single spin-echo, it is offseted with original noise sound field in the spatial dimension of sealing.

First, based on FX-LMS algorithm, set up adaptive control system.As shown in Figure 2, the principle of adaptive control system is as follows:

(1) h _r(n) be elementary microphone reference channel transport function, receive low frequency noise sound-source signal p (n), output low frequency noise signal x (n);

(2) W (n) is filter transfer function, receives the residual signals e (n) after low frequency noise signal x (n) and noise reduction process, exports anti-phase de-noising signal y (n);

(3) h _s(n) be anti-phase imparts acoustic energy delivery function, receive anti-phase de-noising signal y (n), the relevant de-noising signal s (n) of output;

(4) h _p(n) be secondary microphone reference channel transport function, it receives relevant noise cancellation signal, output secondary sound source low frequency noise signal d (n);

(5) relevant de-noising signal s (n) exports the residual signals e (n) after noise reduction process with secondary sound source low frequency noise signal d (n) by coherence stack.

For making control system possess adaptive noise reduction, process, W (n) has self-adaptation adjustment for filter transfer function.

Adaptive control system is exported anti-phase de-noising signal y (n) and is seen following formula:

$y\left(n\right)=W^T\left(n\right)X\left(n\right)=\underset{i=0}{\overset{L-1}{\mathrm{\Σ}}}{w}_i\left(n\right)x(n-i)$

Wherein, the length that L is wave filter, X (n) is reference signal vector, W (n) is wave filter weight coefficient vector.

Residual signals e (n) after noise reduction process is shown in following formula:

e(n)=d(n)+h _s(n)*y(n)

Because the residual signals e (n) after noise reduction process exports relevant d (n) with secondary sound source low frequency noise signal, so e (n) is a stochastic variable.Press minimum mean square error criterion, wave filter best weights coefficient is:

W(n+1)=W(n)-2μ _we(n)R(n)

Wherein, μ _wstep factor for wave filter; R (n) is-X (n) signal phasor that its value equals R (n)=X (n) * h _s(n).

Build a kind of adaptive noise reduction device for indoor environment medium and low frequency noise:

As shown in Figure 3, system is single channel feed forward architecture, 32 MCU of take are core, by audio cable, be connected with two 16 A/D converters respectively, 2 16 high precision converters are realized the synchronous acquisition of 2 tunnel simulating signals, I/O read-write parallel port by 16 pin parallel port bus CPLD devices connects, CPLD device is realized latch function, signal is after phase inverse operation, by high speed serialization D/A device, realize the output of antinoise signal, in addition, outside MCU sheet direct-connected a slice SRAM as data space.

The information data of low frequency noise sound source gathers by elementary microphone and secondary microphone, through 16 A/D converters, transfer to CPLD latch, CPLD latch receives A/D audio conversion alias sampled data, when system break, keep latching, by parallel port, elementary microphone audio frequency signal and secondary microphone audio frequency signal are transferred to 32 MCU processors, for system provides sampling, keep function.MCU is read in latch information and be stored to chip external memory SRAM by external interrupt, after the reverse computing of phase place, by serial D/A converter, export anti-phase sound source to, realizes relevant noise reduction process.

System software is the concrete instruction set of the hardware combined enforcement noise reduction process of low frequency denoiser each several part, is the necessary condition of the normal operation of low frequency denoiser.

System software mainly comprises two parts: main system communication program and interrupt service routine.Wherein, main system communication program comprises 2 road microphone signal capture programs and D/A conversion outputting communication program; Storage, the auto adapted filtering based on FX-LMS algorithm that interrupt service routine is responsible for processing two-stage sound source sound signal regulate and export reverse de-noising signal program.System program flow process is shown in Fig. 4.

Software system design is sought structure based on suitable, and main communication program execution sequence is as follows:

1, carry out system initialization, comprise A/D converter, D/A converter, CPLD device and MCU core;

2, A/D device receives noise data and carries out conversion, and MCU waits for data processing;

3, judgement A/D EOC, if conversion does not finish to continue to receive and converting audio data;

4, after A/D EOC, enter interrupt service routine;

5, interrupt service routine is disposed, and carries out D/A Serial Communication Program, output low frequency de-noising signal.

Interrupt service routine execution sequence is as follows:

1, the audio signal data based on the sampling of CPLD device, maintenance 2 road A/D conversions;

2, based on FX-LMS, calculate y (n) low frequency de-noising signal, self application filtering is reconciled, and reduces the residual error signal e (n) obtaining after noise reduction process;

3, MCU calculates the reverse de-noising signal of output;

4, jump out interruption, return to main communication program.

Application Example 1

Self-adaptation low frequency denoiser described in embodiment 1 is carried out to the noise reduction test of separate space.The correlation standard of our country, uptown noise can not surpass 50 decibels daytime, and night should be lower than 45 decibels, and surpassing this standard will produce harm to human body.Accordingly, noise source draws from an automatic washing machine and carries out the lasting noise sending under standard wash state, through decibel instrument test noise, reaches 48 decibels, human body is formed to harm.

Self-adaptation low frequency denoiser is positioned over separate space central floor and carries out noise reduction test.

As shown in Figure 5, the noise data gathering in 3 minutes in enclosure space, the lasting noise that automatic washing machine carries out sending under standard wash state is between 47～49 decibels, and mean value is 48 decibels, and the lasting audible noise scope that noise source sends forms harm to human body.

As shown in Figure 6, start self-adaptation low frequency denoiser, after stable equipment operation, the noise reduction data of carrying out sound sampling in 3 minutes, in figure, noise range drops between 41～43 decibels, and mean value is about 42 decibels, noise has been reduced to below 45 decibels, meets country about the standard regulation of noise.

As shown in Figure 7, the sound sampled data before and after confined space noise reduction, equipment can carry out obvious noise reduction process to confined space, and its noise reduction amplitude is 6 decibels of left and right.

By test result, electronic equipment of the present utility model can effectively reduce the low frequency noise value in confined space, reduces the physical hazard that low frequency noise causes human body.

Claims (4)

1. the adaptive noise reduction device for indoor environment medium and low frequency noise, comprise elementary microphone, anti-phase sound source and frequency regulation arrangement, it is characterized in that: also comprise a secondary microphone, described secondary microphone is that the high sensitivity of frequency response 20HZ～20KHz is pointed to the secondary microphone of electret entirely; Elementary microphone is connected with 16 A/D converters by audio cable respectively with secondary microphone, 16 A/D converters are connected with the I/O read-write parallel port of frequency regulation arrangement by 16 pin parallel port buses, and anti-phase sound source is connected with the MCU processor of frequency regulation arrangement by serial A/D converting device.

2. the adaptive noise reduction device for indoor environment medium and low frequency noise according to claim 1, is characterized in that: described elementary microphone and secondary microphone are all used two NPN type triodes to form cascode collection secondary pre-amplification circuit.

3. the adaptive noise reduction device for indoor environment medium and low frequency noise according to claim 1, is characterized in that: described frequency regulation arrangement consists of the outer random access memory of a slice CPLD, 32 microprocessing units of a slice and a slice.

4. the adaptive noise reduction device for indoor environment medium and low frequency noise according to claim 1, is characterized in that: described anti-phase sound source adopts 20HZ～20KHz full range signal concurrent coaxial loudspeaker.

Images