Summary of the invention
The invention provides and need not to use special-purpose externally measured instrument just can determine the active noise reduction device of the signal transmission characteristics of self-excited oscillation noise absorber till the error signal generation device.And, the cosine corrected value and the sinusoidal corrected value that do not use a computer and just can calculate described signal transmission characteristics, and the cosine corrected value that calculates and sinusoidal corrected value be stored in the memory storage in the means for correcting.The objective of the invention is to, the active noise reduction device that uses this cosine corrected value and sinusoidal corrected value can initiatively reduce vibrating noise is provided.
Active noise reduction device of the present invention possesses following inscape.That is,
(a) mode selector is selected normal mode and measurement pattern,
(b) frequency detecting device, the normal mode according to mode selector is selected detects the vibrating noise frequency that is produced by vibration noise source,
(c) first switchgear, according to the selected measurement pattern of mode selector, in the output signal of the virtual vibrating noise generation device of the signal of the particular frequency range that selection output is corresponding with the vibrating noise frequency that is produced by vibration noise source and the output signal of frequency detecting device any and output
(d) reference cosine wave generation device and reference sine wave generation device are imported the output signal of first switchgear,
(e) the first adaptive notch filter device for offsetting the generation vibrating noise that produces because of the vibrating noise from vibration noise source, is exported first control signal according to the reference cosine wave signal of reference cosine wave generation device output,
(f) the second adaptive notch filter device is exported second control signal according to the reference sine wave signal of being exported by the reference sine wave generation device,
(g) first adder is imported first control signal and second control signal respectively,
(h) second switch device will offer the vibrating noise cancellation element by the signal of first adder output,
(i) the 3rd switchgear offers the vibrating noise cancellation element with reference cosine wave signal or reference sine wave signal,
(j) vibrating noise cancellation element, the generation vibrating noise of the output of elimination input second switch device and the output of the 3rd switchgear,
(k) error signal that error signal detection device, output produce vibrating noise and produced by the interference result of the elimination sound of vibrating noise cancellation element output,
(l) the 4th switchgear offers second adder with the output of first adder,
(m) second adder is imported the output of the 4th switchgear and the output of error signal detection device,
(n) the 5th switchgear exports reference cosine wave signal to the 3rd adder,
(o) the 6th switchgear exports the reference sine wave signal to the 4th adder,
(p) the first filter factor updating device, according to the output signal of second adder and each output signal of the 5th switchgear, calculate the filter factor of the first adaptive notch filter device so that the output signal of second adder minimizes, and this filter factor is upgraded respectively one by one
(q) the second filter factor updating device, according to the output signal of second adder and each output signal of the 6th switchgear, calculate the filter factor of the second adaptive notch filter device so that the output signal of second adder minimizes, and this filter factor is upgraded respectively one by one
(r) correction value calculation device, the filter factor that input is exported by the first and second filter factor updating devices respectively, at least calculate corresponding with the frequency of reference cosine wave signal or reference sine wave signal by the vibrating noise cancellation element to the signal transmission characteristics between error signal detection device the gain characteristics value and the phase characteristic value in the phase characteristic value, and, can calculate respectively cosine corrected value and sinusoidal corrected value and
(s) means for correcting uses cosine corrected value and sinusoidal corrected value calibration reference cosine wave signal and reference sine wave signal respectively, and high-ranking officers' sine and cosine ripple signal outputs to the 5th switchgear and the 6th switchgear with the correction sine wave signal respectively.
Described means for correcting (s) possesses:
(s1) memory storage, storage cosine corrected value and sinusoidal corrected value,
(s2) first multiplier, complementation string corrected value and reference cosine wave signal long-pending,
(s3) second multiplier is asked the long-pending of sinusoidal corrected value and reference sine wave signal,
(s4) the 3rd multiplier, complementation string corrected value and reference sine wave signal long-pending,
(s5) the 4th multiplier is asked the long-pending of sinusoidal corrected value and reference cosine wave signal,
(s6) the 3rd adder is imported the output signal of first multiplier and the output signal of second multiplier respectively, and the output calibration cosine wave signal,
(s7) the 4th adder is imported the output signal of the 3rd multiplier and the output signal of the 4th multiplier respectively, and the output calibration sine wave signal.
According to the structure of above means for correcting, then need not to use the signal transmission characteristics between the error signal generation device that special-purpose externally measured instrument can determine to contain the loudspeaker vibration noise absorber and contain microphone.And, can not use outer computer to calculate the cosine corrected value and the sinusoidal corrected value of signal transmission characteristics.The invention provides the active noise reduction device, described cosine corrected value that calculates and sinusoidal corrected value are stored in the memory storage of being located at means for correcting, can use cosine corrected value and the sinusoidal corrected value preserved to reduce vibrating noise on one's own initiative.
Embodiment
(embodiment 1)
Below, use Fig. 1~Fig. 4 that embodiment 1 is described.Fig. 1 is the block scheme of the formation of explanation embodiments of the present invention 1 described active noise reduction device.Fig. 2 is the block scheme of the operation of the measurement pattern in the explanation active noise reduction device shown in Figure 1.Fig. 3 is the block scheme of the operation of the normal mode in the explanation active noise reduction device shown in Figure 1.Fig. 4 is for illustrating in the active noise reduction device of the present invention shown in Figure 1 the block scheme of the operation when having a plurality of vibrating noise cancellation elements or error signal detection device.
In the active noise reduction device 100 of Fig. 1, can be divided into substantially: revolution speed detecting device 1, contact panel 3, microphone 15, vibrating noise cancellation element 30 and discrete arithmetic processing apparatus 32.Vibrating noise cancellation element 30 possesses power amplifier 13, loudspeaker 14.
Possess in the discrete arithmetic processing apparatus 32: frequency detecting device 2, virtual vibrating noise generation device 4, first switchgear 5, reference cosine wave generation device 6, reference sine wave generation device 7, the first adaptive notch filter devices 8, the second adaptive notch filter device 9, first adder 10, second switch device 11, the three switchgears 12, the 4th switchgear 16, second adder, 17, the five switchgears, 18, the six switchgears 19, the first adaptive control algorithm arithmetic unit 20, the second adaptive control algorithm arithmetic unit 21, correction value calculation device 22 and means for correcting 31.
Also has frequency detecting device 2, virtual vibrating noise generation device 4, first switchgear 5, reference cosine wave generation device 6, reference sine wave generation device 7, the first adaptive notch filter device 8, the second adaptive notch filter device 9, first adder 10, second switch device 11, the 3rd switchgear 12, the 4th switchgear 16, second adder 17, the 5th switchgear 18, the 6th switchgear 19, the first adaptive control algorithm arithmetic unit 20, the second adaptive control algorithm arithmetic unit 21, correction value calculation device 22, first multiplier 24, second multiplier 25, the 3rd multiplier 26, the 4th multiplier 27, the 3rd adder 28 and the 4th adder 29 are by the software service that is built-in with CPU etc. and constitute.
Yet, have at least one can constitute in described first switchgear 5, second switch device 11, the 3rd switchgear 12, the 4th switchgear 16, the 5th switchgear 18 and the 6th switchgear 19 by hardware.
And the revolution speed detecting device 1 of the active noise reduction device 100 shown in Fig. 1 detects the rotation number that is equipped on the engine on the vehicle.The engine pulse that input is detected by revolution speed detecting device 1 in the frequency detecting device 2, and export correspondent frequency signal therewith.Contact panel 3 has the input device that is equipped on the sound system on the vehicle as mode selector.If select measurement pattern then the signal of virtual vibrating noise generation device 4 generation characteristic frequency by contact panel 3.
First switchgear 5 is the output signal of the output signal of output frequency pick-up unit 2 or virtual vibrating noise generation device 4 according to the selection of contact panel 3 indication and optionally.Reference cosine wave generation device 6 produces reference cosine wave signal according to the output signal from first switchgear 5.Reference sine wave generation device 7 produces the reference sine wave signal according to the output signal from first switchgear 5.
The first adaptive notch filter device 8 is according to exporting first control signal from the reference cosine wave signal of reference cosine wave generation device 6 outputs.The second adaptive notch filter device 9 is according to exporting second control signal from the reference sine wave signal of reference sine wave generation device 7 outputs.
Import first control signal and second control signal in first adder 10 respectively.The signal that second switch device 11 is used for being exported from first adder 10 offers vibrating noise cancellation element 30 or interrupts the supply of described signal.Switch shown in Figure 1 11 for the state opened, be interruption status.The 3rd switchgear 12 is used for the reference sine wave signal is offered vibrating noise cancellation element 30 or interrupts the supply of described signal.Switch 12 shown in Figure 1 is open mode, is interruption status.
Import the output signal of second switch device 11, the 3rd switchgear 12 in the power amplifier 13 respectively.The output signal of power input amplifier 13 in the loudspeaker 14.Microphone 15 has the function of error signal detection device, the error signal that the interference result of the elimination sound that the output of described error signal detection device is exported as the vibrating noise that engine produced and the loudspeaker 14 of vibration noise source is produced.
The 4th switchgear 16 offers second adder 17 with the output of first adder 10, or interrupts described output.Import the output of the 4th switchgear 16 and the output of microphone 15 in second adder 17 respectively.The 5th switchgear 18 receives the indication from contact panel 3, and will input to the 3rd adder 28 from the reference cosine wave signal of reference cosine wave generation device 6 outputs.
The 6th switchgear 19 receives the indication of contact panel 3, and the reference sine wave signal is inputed to the 4th adder 29.The first adaptive control algorithm arithmetic unit 20 calculates the filter factor of the first adaptive notch filter device 8, and upgrades this coefficient.The second adaptive control algorithm arithmetic unit 21 calculates the filter factor of the second adaptive notch filter device 9, and upgrades this coefficient.Import filter factor in the correction value calculation device 22 respectively by the first adaptive control algorithm arithmetic unit 20 and 21 outputs of the second adaptive control algorithm arithmetic unit.
Correction value calculation device 22 can calculate from the power amplifier 13 corresponding, loudspeaker 14 with the frequency of reference sine wave signal to the signal transmission characteristics of 15 of microphones the gain characteristics value and the value of phase characteristic at least in the phase characteristic value.And correction value calculation device 22 can calculate cosine corrected value C0 and sinusoidal corrected value C1.Memory storage 23 storage cosine corrected value C0 and sinusoidal corrected value C1.Amassing of first multiplier, 24 complementation string corrected value C0 and reference cosine wave signal.Second multiplier 25 is asked the long-pending of sinusoidal corrected value C1 and reference sine wave signal.Amassing of the 3rd multiplier 26 complementation string corrected value C0 and reference sine wave signal.The 4th multiplier 27 is asked the long-pending of sinusoidal corrected value C1 and reference cosine wave signal.The input side of the 3rd adder 28 is imported the output signal of first multiplier 24 and the output signal of second multiplier 25 respectively, and from the outgoing side output calibration cosine wave signal of the 3rd adder 28.The input side of the 4th adder 29 is imported the output signal of the 3rd multiplier 26 and the output signal of the 4th multiplier 27 respectively, and from the outgoing side output calibration sine wave signal of the 4th adder 29.Vibrating noise cancellation element 30 is made of power amplifier 13 and loudspeaker 14.Comprise in the means for correcting 31: memory storage 23, first multiplier 24, second multiplier 25, the 3rd multiplier 26, the 4th multiplier 27, the 3rd adder 28 and the 4th adder 29.
In addition, use the contact panel 3 possess as the input device of the sound system of vehicle mounted equipment in the mode selector.Be used under the situation of vehicle at the active noise reduction device of the present invention that will adopt described structure, can utilize the truck mounted machine of extensively popularizing, very convenient.
The following describes the situation that adopts sound system as vehicle mounted equipment.But, be not limited only to described situation.For example, also can be used in navigational system etc.
Herein,, the contact panel 3 that has as the input device of the sound system of mobile unit is described, but is not limited only to described content, also can use speech recognition equipment with mechanical switch or microphone etc. as mode selector.Thus, not only can freely select measurement pattern and normal mode, can also constitute and need not manually operated mode selector.
Secondly, below use Fig. 2 that the operation of measurement pattern is described.Use identical symbol for the position identical with Fig. 1.
Select measurement pattern by contact panel 3, then virtual vibrating noise generation device 4 brings into operation.The output signal that has characteristic frequency by virtual vibrating noise generation device 4 outputs.Described output signal is selected by first switchgear 5, and inputs to reference cosine wave generation device 6 and reference sine wave generation device 7 respectively.
Reference sine wave generation device 7 by the 3rd switchgear 12 will with offer power amplifier 13 by the synchronous reference sine wave signal of the output signal frequency of virtual vibrating noise generation device 4 outputs.The output of power amplifier 13 inputs in the loudspeaker 14.The reference sine wave signal is launched by loudspeaker 14, and microphone 15 should be launched sound and detect as error signal e (n).Detected error signal e (n) is inputed in second adder 17.
Reference cosine wave signal by 6 outputs of reference cosine wave generation device is multiplied each other with filter factor W0 (n) by the first adaptive notch filter device 8.Reference sine wave signal by 7 outputs of reference sine wave generation device is multiplied each other with filter factor W1 (n) by the second adaptive notch filter device 9.The output signal of the output signal of the first adaptive notch filter device 8 and the second adaptive notch filter device 9 is by 10 additions of first adder.The output signal of first adder 10 inputs to second adder 17 by the 4th switchgear 16.From the error signal e (n) that detects by microphone 15, deduct the output signal of exporting by first adder 10 by second adder 17.The signal of described process subtraction is as error signal e ' (n) output.Described error signal e ' (n) input to respectively in the first adaptive control algorithm arithmetic unit 20 and the second adaptive control algorithm arithmetic unit 21.
Below explanation is according to the mechanism of adaptive control algorithm with filter factor W1 (n) renewal of the filter factor W0 (n) of the first adaptive notch filter device 8 and the second adaptive notch filter device 9.As this adaptive control algorithm, for example, as well-known a kind of LMS (Least Mean Square, the lowest mean square) algorithm that has in the steepest descent method.According to described algorithm, the filter factor W0 (n) of the first adaptive notch filter device 8 is upgraded by the first adaptive control algorithm arithmetic unit 20.The filter factor W1 (n) of the second adaptive notch filter device 9 is upgraded by the second adaptive control algorithm arithmetic unit 21.The filter factor W0 (n+1) of the first adaptive notch filter device 8 and the filter factor W1 (n+1) of the second adaptive notch filter device 9, can use reference cosine wave signal r0 ' that the filter factor W1 (n), error signal e of filter factor W0 (n), the second adaptive notch filter device 9 of the first adaptive notch filter device 8 before upgrading ' (n) and reference cosine wave generation device 6 exported (n), the reference sine wave signal r1 ' that exported of reference sine wave generation device 7 (n) and step-size parameter mu, try to achieve by formula (1), formula (2).Step-size parameter mu can be used steepest descent method decision speed of convergence.
[formula 1]
W0(n+1)=W0(n)+μ·e′(n)·r0′(n)
[formula 2]
W1(n+1)=W1(n)+μ·e′(n)·r1′(n)
Like this, the filter factor W0 (n) of the first adaptive notch filter device 8 and the filter factor W1 (n) of the second adaptive notch filter device 9 are updated to error signal e ' (n) and approach zero, and converge to optimum value.Herein, converge to optimum value and be meant employing threshold value e0, e1 represents then to satisfy the state of formula (3), formula (4).
[formula 3]
W0(n+1)-W0(n)<e0
[formula 4]
W1(n+1)-W1(n)<e1
The filter factor W0 (n) of the first adaptive notch filter device 8 and the filter factor W1 (n) of the second adaptive notch filter device 9 make the output signal of first adder 10 equate with the error signal e (n) that detects through microphone 15 by converging to optimum value as mentioned above.That is, the output signal of first adder 10 and error signal e (n) can embody power amplifier 13, loudspeaker 14 signal transmission characteristics to 15 of microphones really.
And, the filter factor that converges to the first adaptive notch filter device 8 after the optimum value is made as W0 ', the filter factor that converges to the second adaptive notch filter device 9 after the optimum value is made as W1 ', and then error signal e (n) is suc as formula shown in (5), the formula (6).
[formula 5]
e(n)=R·sin(ωt+a)
[formula 6]
e(n)=W0′·cos(ωt)+W1′·sin(ωt)
W0 ', W1 ' are inputed to correction value calculation device 22, utilize correction value calculation device 22 to carry out the computing shown in formula (7), the formula (8), obtain gain characteristics value G7 and phase characteristic value f7 in the described signal transmission characteristics with this.
[formula 7]
G7=v(W0′
2+W1′
2)
[formula 8]
f7=-arctan(W0′/W1′)
In addition, filter factor W0 ' and filter factor W1 ' are inputed to correction value calculation device 22, utilize correction value calculation device 22 to carry out the computing shown in formula (9), the formula (10) complementation string corrected value C0 and sinusoidal corrected value C1 respectively.
[formula 9]
C0=v(W0′
2+W1′
2)cos{-arctan(W0′/W1′)}
[formula 10]
C1=v(W0′
2+W1′
2)sin{-arctan(W0′/W1′)}
Cosine corrected value C0 and sinusoidal corrected value C1 are saved in the memory storage 23, thereby, the series of steps in the measurement pattern finished.
According to described calculation step, need not to use special-purpose externally measured instrument can ask the signal transmission characteristics to 15 of microphones from power amplifier 13, loudspeaker 14.And, can not use outer computer and calculate cosine corrected value C0 and sinusoidal corrected value C1.Described cosine corrected value C0 and sinusoidal corrected value C1 are stored in the memory storage 23 that is arranged at means for correcting 31.
In addition, in discrete arithmetic processing apparatus 32 shown in Figure 2, be provided with the gain characteristics value of storage correction value calculation device 22 calculating and second memory storage (not shown) of phase characteristic value.And be provided with comparer (not shown), described comparer be used for gain characteristics value that comparison correction value calculation device 22 first calculated go out and phase characteristic value, with the difference of the value of phase characteristic at least of gain characteristics value that calculates afterwards and phase characteristic value whether in particular value.By described formation, can bring into play new as described below function.
That is comparer gives a warning when, can surpass particular value in the difference of phase characteristic value.That is the signal transmission characteristics to 15 of microphones has produced variation from loudspeaker 14, can to notify the driver of vehicle.
In addition, in the comparer, difference in the phase characteristic value surpasses under the situation of particular value, can utilize by as first adaptive control algorithm arithmetic unit 20 of the first filter factor updating device and the filter factor of exporting respectively as the second adaptive control algorithm arithmetic unit 21 of the second filter factor updating device, use correction value calculation device 22 to calculate cosine corrected value and sinusoidal corrected value once more.After calculating, described cosine corrected value and sinusoidal corrected value are stored in the memory storage 23.Thus, the signal transmission characteristics to 15 of microphones has produced variation even the present invention is from loudspeaker 14, also can fully eliminate vibrating noise once more.
In addition, when having selected measurement pattern by contact panel 3, if engine is out of service, can suppress then that passenger to vehicle transmits that loudspeaker 14 sent be used to test make us uncomfortable emission sound.
Below, the operation of normal mode is described with reference to Fig. 3.Use identical symbol for the part identical with Fig. 1, Fig. 2.When selecting normal modes by contact panel 3, the rotation number of the engine that is detected by revolution speed detecting device 1 can convert the signal of pulse waveform to and offer frequency detecting device 2.In addition, the output signal of frequency detecting device 2 is to select and input to reference cosine wave generation device 6 and reference sine wave generation device 7 by first switchgear 5.
Reference cosine wave generation device 6 and reference sine wave generation device 7 produce respectively and synchronous reference cosine wave signal and the reference sine wave signal of exporting from frequency pick-up unit 2 of output signal frequency.
The reference cosine wave signal that reference cosine wave generation device 6 is exported multiplies each other with filter factor W0 (n) in the first adaptive notch filter device 8.The reference sine wave signal that reference sine wave generation device 7 is exported multiplies each other with filter factor W1 (n) in the second adaptive notch filter device 9.The output signal of the first adaptive notch filter device 8, with output signal addition in first adder 10 of the second adaptive notch filter device 9.The output signal of first adder 10 offers power amplifier 13 by second switch device 11.The output of power amplifier 13 inputs in the loudspeaker 14.Loudspeaker 14 is launched the elimination sound that is used to eliminate the vibrating noise that engine sends.
In addition, even select normal modes by contact panel 3, the vibrating noise that may not thoroughly eliminate engine and produced from the initial elimination sound of loudspeaker 14.
Therefore, below explanation uses the present invention thoroughly to eliminate the signal Processing step of vibrating noise.At first, the interference of the initial elimination sound of being launched by vibrating noise that engine produced, with loudspeaker 14.At this moment, utilize 15 pairs of microphones not fully the residual sound of noise reduction detect.
The residual sound that microphone 15 detected detects as error signal e (n).Described detected error signal e (n) inputs in the first adaptive control algorithm arithmetic unit 20 and the second adaptive control algorithm arithmetic unit 21 as error signal e (n) by second adder 17.Error signal e (n) is used for adaptive control algorithm, and described adaptive control algorithm is used for upgrading respectively the filter factor W0 (n) of the first adaptive notch filter device 8 and the filter factor W1 (n) of the second adaptive notch filter device 9.
Secondly, utilize first multiplier 24 that the cosine corrected value C0 that is stored in reference cosine wave signal (cos ω t) and the memory storage 23 is multiplied each other.In addition, utilize second multiplier 25 that the sinusoidal corrected value C1 that is stored in reference sine wave signal (sin ω t) and the memory storage 23 is multiplied each other.The output signal of the output signal of first multiplier 24 and second multiplier 25 inputs in the 3rd adder 28.Utilize the 3rd multiplier 26 that the cosine corrected value C0 that is stored in reference sine wave signal (sin ω t) and the memory storage 23 is multiplied each other.Utilize the 4th multiplier 27 that the sinusoidal corrected value C1 that is stored in reference cosine wave signal (cos ω t) and the memory storage 23 is multiplied each other.The output signal of the output signal of the 3rd multiplier 26 and the 4th multiplier 27 inputs in the 4th adder 29.Therefore, by the 3rd adder 26 and the 4th adder 27, can export respectively suc as formula correction cosine wave signal r0 (n) shown in (11), the formula (12) and correction sine wave signal r1 (n).
[formula 11]
r0(n)=C0·cosωt+C1·sinωt
[formula 12]
r1(n)=C0·sinωt-C1·cosωt
Proofread and correct cosine wave signal r0 (n) and proofread and correct sine wave signal r1 (n) and input to respectively in the first adaptive control algorithm arithmetic unit 20, the second adaptive control algorithm arithmetic unit 21.Described correction cosine wave signal r0 (n) and proofread and correct sine wave signal r1 (n) and be used in order to the filter factor W0 (n) that upgrades the first adaptive notch filter device 8, and the adaptive control algorithm of the filter factor W1 (n) of the second adaptive notch filter device 9.
Below explanation uses adaptive control algorithm that the signal Processing step that the filter factor W1 (n) of the filter factor W0 (n) of the first adaptive notch filter device 8 and the second adaptive notch filter device 9 upgrades is carried out following explanation.Identical with the situation of measurement pattern, according to the LMS algorithm, for the filter factor W0 (n) of the first adaptive notch filter device 8, the filter factor W1 (n) of the second adaptive notch filter device 9, use the first adaptive control algorithm arithmetic unit 20 and the second adaptive control algorithm arithmetic unit 21 to upgrade respectively.
Secondly, the filter factor W0 (n+1) of the first adaptive notch filter device 8 that upgrades by the first adaptive control algorithm arithmetic unit 20 and the second adaptive control algorithm arithmetic unit 21 respectively, filter factor W1 (n+1) with the second adaptive notch filter device 9, can use the filter factor W0 (n) of the first adaptive notch filter device 8 before upgrading, the filter factor W1 (n) of the second adaptive notch filter device 9, error signal e (n), the correction sine wave signal r1 (n) that the correction cosine wave signal r0 (n) of the 3rd adder 28 outputs and the 4th adder 29 are exported, and step-size parameter mu, according to formula (13), formula (14) is obtained.In addition, step-size parameter mu is used steepest descent method decision speed of convergence as mentioned above.
[formula 13]
W0(n+1)=W0(n)-μ·e(n)·r0(n)
[formula 14]
W1(n+1)=W1(n)-μ·e(n)·r1(n)
As mentioned above, the filter factor W0 (n) of the first adaptive notch filter device 8, and the filter factor W1 (n) of the second adaptive notch filter device 9 obtain upgrading, make error signal e (n) approach zero, and converge to optimum value.This means that the vibrating noise that engine produced can thoroughly be eliminated by the elimination sound that the loudspeaker 14 that constitutes vibrating noise cancellation element 30 is launched.
Below with reference to Fig. 4 explanation for having a plurality of vibrating noise cancellation elements 30 that constitute by power amplifier 13 and loudspeaker 14 or the operation of a plurality of microphones 15 as error signal detection device being arranged.
Conventionally, in common vehicle, loudspeaker is arranged on Qianmen and the back door, and microphone is arranged near the driver's seat.Therefore, loudspeaker 14 to the signal transmission characteristics of 15 of microphones is fixing (qualification) to a certain extent.Yet, recently, be accompanied by popularizing of real entertainment (true amusement) technology, there have the multitone case surrounding system of the loudspeaker more than six or microphone that hand-free call is used to be located at second of vehicle or the 3rd this situation use to be not rare.Therefore, loudspeaker to the selection degree of freedom of the signal transmission characteristics between microphone increases.Therefore, under measurement pattern, select better signal transmission characteristics, it is stored, and under normal mode, be used, thereby can obtain better noise reduction.
Therefore, under the situation that has a plurality of loudspeakers 14 and microphone 15, it is represented with SPK (i) and MIC (j) respectively.Wherein, loudspeaker is M, and microphone is N, is located in the vehicle, and i is that integer, the j of 1~M is the integer of 1~N.
In the operation instructions of before measurement pattern, be the explanation of carrying out for fixing situation for SPK (i) and MIC (j).Under these circumstances, even loudspeaker always is located at identical position with microphone, when the signal transmission gain characteristic of SPK (i) between MIC (j) also can not exist energy level reduction or depression, can not produce any fault, therefore can carry out noise reduction control more simply.Yet, be provided with in the vehicle of active noise reduction device, exist the situation of interior distinctive peak value of vehicle or depression much on the signal transmission gain characteristic.Therefore, the noise reduction control near the frequency band the depression becomes unstable.Perhaps, in the lower frequency band of the energy level of signal transmission gain characteristic, the elimination sound of launching as the loudspeaker of vibrating noise cancellation element will inevitably become greatly, thereby can produce the such problem of sound that loudspeaker sends distortion.
Therefore, under measurement pattern, select SPK (i) the loudspeaker of the M in being located at vehicle, and from N microphone, select MIC (j).Obtain to the gain characteristics value of the signal transmission characteristics between MIC (j) for SPK (i) that (M * N) plants, and it is stored in the 3rd memory storage.In second comparer, for be stored in the 3rd memory storage (M * N) plants the gain characteristics value and compares, and selects the combination of dark depression is less and the gain energy level is higher SPK (i) and MIC (j).To be stored in the memory storage 23 according to cosine corrected value and the sinusoidal corrected value that selected SPK (i) calculates to the signal transmission characteristics between MIC (j).Cosine corrected value and the sinusoidal corrected value stored in the memory storage 23 under the normal mode operation by using can be provided, can more effectively reduce the active noise reduction device of noise.
And, at each frequency, in second comparer to (M * N) plants the gain characteristics value and compares, and selects dark depression still less and the combination of the higher SPK (i) of gain energy level and MIC (j).Will according at the selected SPK of each frequency (i) to cosine corrected value that signal transmission characteristics calculated and sinusoidal corrected value between MIC (j), be stored in the memory storage 23.The normal mode operation is used cosine corrected value and the sinusoidal corrected value stored in the memory storage 23 down.Thereby, in the full range band of noise reduction controlling object,, SPK (i) caves in or the lower part of gain energy level even if all existing to the arbitrary signal transmission characteristics between MIC (j), and also can provide noise reduction higher active noise reduction device.
(embodiment 2)
Fig. 5 is the block diagram of the active noise reduction device of embodiment 2, and Fig. 6 is the block scheme of expression normal manipulation mode for the block scheme of expression measurement pattern operation, Fig. 7.In addition, for the structure identical, use same-sign and omit its related description with embodiment 1.
Active noise reduction device 100 possesses first means for correcting 40 that the reference sine wave signal exported for the reference sine wave generation device is proofreaied and correct.When selecting measurement pattern, the signal of proofreading and correct through first means for correcting 40 inputs in the power amplifier 13 by the 3rd switchgear 12.Minion close that device 41 receives from the indication of contact panel 3 and reference cosine wave signal that the reference cosine wave generation device is exported, and the first adaptive notch filter device 8 in signal after multiplying each other with filter factor W0 offer a input terminal in first adder.Octavo close that device 42 receives the indication of contact panel device 3 and reference sine wave signal that the reference sine wave generation device is exported, with the second adaptive notch filter device 9 in signal after multiplying each other with filter factor W1 offer another input terminal in first adder.Second means for correcting 43 closes the signal of being exported to minion and proofreaies and correct and it is inputed in first adder 10 when measurement pattern is moved.The 3rd means for correcting 44 is proofreaied and correct and it is inputed in first adder 10 signal that octavo pass device is exported when measurement pattern is moved.
Be used for illustrating Fig. 5~Fig. 7 of embodiment 2, be with the difference of embodiment 1: except structure, have in addition that first means for correcting 40, minion close device 41, octavo is closed device 42, second means for correcting 43 and the 3rd means for correcting 44 with Fig. 1~shown in Figure 3.
The method of trying to achieve signal transmission characteristics herein, when moving for measurement pattern describes.For example, when loudspeaker 14 when the gain characteristic of the signal transmission characteristics of 15 of microphones substantially exceeds 0dB etc., the error signal e (n) that microphone 15 is detected also can become big.Yet there is the upper limit in the amplitude of microphone 15 detectable signals.Therefore, when the amplitude of the locational transmission signals of microphone 15 surpasses the higher limit that microphone 15 can detect, can't obtain the error signal e (n) of right value.
That is,, also can't obtain correct value according to the filter factor W0 ' of the first adaptive notch filter device 8 that convergency value obtained of adaptive control algorithm computing and the filter factor W1 ' of the second adaptive notch filter device 9.Therefore, there is the also incorrect such problem of gain characteristics value that is obtained according to formula (7).
Therefore, if add first means for correcting 40 that is used for the calibration reference sine wave signal, then can reduce the absolute value of corrected value ρ.Thereby, can reduce the amplitude of transmission signals on the position of microphone 15.The error signal e (n) of right value can be obtained, and correct gain characteristics value can be obtained.The gain characteristics value of this moment as the formula (15).
[formula 15]
G15=1/ρ·v(W0′
2+W1′
2)
And, when even the amplitude of the locational transmission signals of microphone 15 surpasses microphone 15 detectable higher limits, when the numerical range of filter factor W0 ' and filter factor W1 ' also is restricted, can't show the problem of the gain characteristics value more than the 0dB in the time of for example can producing the definition of carrying out as the Q7 form.Herein, the Q7 form is meant, in the fixed point representation of 8 bits one, that is and, the information distribution that radix point is later is given 7 bits down.Therefore, minion is closed device 41, octavo is closed device 42, second means for correcting 43 and the 3rd means for correcting 44 if add, and then the gain characteristics value can utilize corrected value s to be represented by formula (16).
[formula 16]
G16=s·v(W0′
2+W1′
2)
Therefore,, W0 ' and W1 ' can be in the numerical range that can show, represented, therefore correct gain characteristics value can be tried to achieve by increasing the absolute value of corrected value s.
And, even the amplitude of the locational transmission signals of microphone 15 surpasses microphone 15 detectable higher limits, and the numerical range of filter factor W0 ' and filter factor W1 ' is restricted, also can utilize formula (17) to obtain the gain characteristics value by the absolute value that reduces corrected value ρ, the absolute value that enlarges corrected value s.
[formula 17]
G17=s/ρ·v(W0′
2+W1′
2)
Secondly, the situation that is significantly less than 0dB for loudspeaker 14 to the gain characteristic of the signal transmission characteristics of 15 of microphones describes.At this moment, the value of the filter factor W1 ' of the filter factor W0 ' of the first adaptive notch filter device 8 of trying to achieve according to formula (5), formula (6) and formula (7) and by the convergency value that adapts to the control algolithm computing and the second adaptive notch filter device 9 diminishes.When the value of filter factor W0 ' and filter factor W1 ' diminished, the relative error that can produce 1LSB became big this problem.
For example, when filter factor W0 ', filter factor W1 ' are 8 bit value of signed, the value of gained is made as W0 '=1, W1 '=2.Suppose that error that it contains 1LSB and true approximate value are W0 '=2, W1 '=2, then according to formula (8), the true approximate value of phase characteristic value is 45 degree, is 26.6 degree according to the phase characteristic value that W0 ', W1 ' calculated of gained.Therefore, the error of phase characteristic value becomes 29% ((45-26.6)/45).
On the other hand, when trying to achieve filter factor W0 ' and filter factor W1 ' for bigger value, for example, W0 '=99, W1 '=100.Suppose that error that it contains 1LSB and true approximate value are W0 '=100, W1 '=100, then the true approximate value of trying to achieve the phase characteristic value according to formula (8) is 45 degree, is 44.7 degree according to the phase characteristic value that W0 ', W1 ' calculated of gained.Thereby the error of phase characteristic value becomes 0.7% ((45-44.7)/45).
Therefore, if add first means for correcting 40 that is used for the calibration reference sine wave signal, then, can increase the amplitude of transmission signals on the microphone position by increasing the absolute value of corrected value ρ.Thereby, can try to achieve filter factor W0 ' and filter factor W1 ' and be bigger value.With this, can reduce the error of phase characteristic value.And, be added with minion and close device 41, octavo pass device 42, second means for correcting 43 and the 3rd means for correcting 44.Therefore, the filter factor of the filter factor of the first adaptive notch filter device 8 of trying to achieve according to the convergency value of adaptive control algorithm computing and the second adaptive notch filter device 9 can be respectively by sW0 ' and sW1 ' expression.
By reducing the absolute value of corrected value s, can obtain W0 ' and W1 ' and be bigger value, therefore can reduce the error of phase characteristic value equally.Therefore, be added with first means for correcting 40, the minion that are used for the calibration reference sine wave signal and close device 41, octavo pass device 42, second means for correcting 43 and the 3rd means for correcting 44.According to such structure, try to achieve the filter factor W0 ' of the first adaptive notch filter device 8 and the filter factor W1 ' of the second adaptive notch filter device 9 and be bigger value, so can further dwindle the error of phase characteristic value.
(embodiment 3)
Use Fig. 8 that embodiment 3 is described.Fig. 8 is the simple block scheme of representing the block scheme (Fig. 3) that embodiment 1 described active noise reduction device is operated under normal mode.Fig. 9 is the block scheme of simple presentation graphs 8, wherein the vibrating noise cancellation element 30 that will be made of power amplifier 13 and loudspeaker 14 to the signal transmission characteristics of 15 of microphones is made as β, the signal transmission characteristics of adaptive notch filter is made as γ, and described signal transmission characteristics is meant the signal transmission characteristics of the output of reference sine wave signal to the first adder 10 that the reference cosine wave signal exported from reference cosine wave generation device 6 or reference sine wave generation device 7 are exported.Structure according to Fig. 9, generation vibrating noise Vn, error signal Ve in the vehicle, output Vout, vibrating noise cancellation element 30 to the relation between the signal transmission characteristics γ of the signal transmission characteristics β of 15 of microphones and auto-adaptive filter device can be suc as formula shown in (18), the formula (19), and according to formula (18), formula (19), Ve/Vn can be represented by formula (20).
[formula 18]
Ve·γ=Vout
[formula 19]
β·Vout=Ve-Vn
[formula 20]
Ve/Vn=1/(1-β·γ)
Figure 10 represents the Ve/Vn characteristic, that is, and and the situation the when frequency of reference cosine wave signal and reference sine wave signal is 50Hz.It is exactly the noise reduction of expression active noise reduction device.When design active noise reduction device 100, emphasis is to consider to keep aforesaid characteristic.That is, holding signal transport property β and γ's is long-pending, that is, β γ is constant, can help keeping the performance of active noise reduction device.
For example, when the user of the vehicle that is assembled with the active noise reduction device with in active noise reduction device and vehicle input amount postpartum of being equipped with described device, replace power amplifier 13 or loudspeaker 14 with existing power amplifier or loudspeaker, will cause noise absorber 30 to the signal transmission characteristics of 15 of microphones that bigger variation is arranged.That is, signal transmission characteristics β changes, but as discussed previously, and the variation of signal transmission characteristics β has harmful effect to the performance of active noise reduction device.Described way to solve the problem is as described below.
Figure 11 makes an addition to the 4th means for correcting 50 deferent segment of first adder in the normal manipulation mode block scheme of Fig. 3.Will and vibrating noise cancellation element 30 to 15 of microphones produced the inversely proportional corrected value of the gain characteristics value of the signal transmission characteristics that changes and be used in the 4th means for correcting 50, thereby, but the amassing of holding signal transport property γ and β, promptly γ β is constant.
At this moment, long-pending for signal transmission characteristics γ and β, that is, it is as described below that γ β remains constant additive method.At first, it is better qualitative to require signal transmission characteristics γ to have.The filter factor W1 of the filter factor W0 of the first adaptive notch filter device 8 and the second adaptive notch filter device 9 is represented with Δ W0, Δ W1 respectively by the renewal amount that an adaptive control computing is upgraded, the frequency of reference cosine wave signal and reference sine wave signal is with ω 0 expression, the frequency of vibrating noise is represented with ω, and can be represented by formula (21), formula (22) according to formula (13), formula (14).
[formula 21]
ΔW0=-(exp(jω0t)+exp(-jω0t)/2·(exp(j(ωt+a))+exp(-j(ωt+a))/2·μ
[formula 22]
ΔW1=-(exp(jω0t)-exp(-jω0t)/2j·(exp(j(ωt+a))+exp(-j(ωt+a))/2·μ
, establish ω x=ω 0+ ω, ω y=ω 0-ω herein, A0, A1 are constant arbitrarily, but then through type (23), formula (24) expression.
[formula 23]
∫ΔW0=-μ/4·{exp(j(ωyt-a))/jωy-exp(-j(ωyt-a))/jωy}+A0
[formula 24]
∫ΔW1=-μ/4j·{exp(j(ωyt-a))/jωy+exp(-j(ωyt-a))/jωy}+A1
And signal transmission characteristics γ can represent with formula (25).
[formula 25]
γ=∫ΔW0·(exp(jω0t)+exp(-jω0t)/2+∫ΔW1·(exp(jω0t)-exp(-jω0t)/2j
And then use formula (23), formula (24) are similar to, but then signal transmission characteristics γ through type (26) expression.
[formula 26]
γ=μ/2(ω0-ω)·sin(ωt+a)
Therefore, be applied to the step-size parameter mu of adaptive control algorithm, calibrated be and change after vibrating noise cancellation element 30 to the inversely proportional value of gain characteristics value of the signal transmission characteristics of 15 of microphones, can make amassing of signal transmission characteristics γ and β, promptly, it is constant that γ β keeps, and can keep the performance of active noise reduction device.