JP3278176B2 - Vehicle noise control device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle noise control device for reducing noise in a vehicle cabin.

[0002]

2. Description of the Related Art One of the methods for reducing the noise in the cabin of an automobile is to filter a reference signal related to the vibration of a noise source with an appropriate filter and generate a sound wave of the opposite phase from a speaker. An active noise control method that silences sound due to noise has been well known.

As one of the active noise control methods well known in the related art, there is a method using an LMS algorithm (Patent Application Publication No. 1-501344).
No.).

The LMS algorithm is to grasp a transfer function between the position of the detected microphone and the position of the speaker in advance, and to calculate and output a noise control sound wave having a phase opposite to that of the input sound wave. The filter coefficient vectors F ₁ , F ₂ ... Fm are successively updated using the following equation to obtain optimum noise reduction performance.

[0005]

(Equation 1)

Here, Fm: a filter coefficient vector for determining the output of the m-th speaker Hmn: a filter coefficient vector for giving a transfer characteristic between the m-th speaker and the n-th microphone x: a reference signal (the same frequency as the noise) Component: en: input signal from the n-th microphone μ: coefficient for determining the peripheral speed in equation (1) M: number of speakers N: number of microphones I: tap length of filter coefficient vector Fm J : Tap length of the filter coefficient vector Hmn k: Value at the k-th point Further, the speaker output ym (k) of the m-th speaker is obtained by the following equation.

[0007]

(Equation 2)

If the speaker output determined through the above equations (1) and (2) is generated from each speaker, the noise and the sound generated from the speaker cancel each other out at each microphone position, so that a noise reduction effect can be obtained. It is.

The above-described noise control by the LMS algorithm is also explained by a block diagram shown in FIG.

Here, the case of one speaker and one microphone is described as an example.

From the speaker 3, a control signal z based on a speaker output y obtained by processing the reference signal x by a filter F (specified by the adaptive algorithm A) is output. The synthesized noise d is observed and input to the adaptive algorithm A as an error signal e. Further, the reference signal x is processed by the filter H that gives a speaker / microphone transfer characteristic, and is input to the adaptive algorithm A as a signal r. Therefore, the speaker 3
The control signal z output from and cancels the synthetic noise d to obtain a noise reduction effect. Here, the filter F,
The filter H and the adaptive algorithm A use the speaker output
It is to constitute signal generation means for generating y.

On the other hand, as an algorithm that does not use the transfer characteristic between the speaker and the microphone, for example, an algorithm that applies a simplex method, which is a kind of nonlinear programming, can be considered.

The noise control using the simplex algorithm is performed as follows.

(Step 0) In advance, (I + 1) × M different filter coefficient vectors Fm (1), Fm (2)... Fm (I +
1) (m = 1, 2,... M) is prepared, and first, m = 1.

(Step 1) Evaluation function J [Fm (i)] (i = 1,
2... I + 1). However, the evaluation function J [Fm (i)]
Is a value that evaluates the degree of reduction of the sound collected by the microphone, and is expressed by the following equation when Fm (i) is used as a filter coefficient vector for determining the output of the m-th speaker. It is.

[0016]

(Equation 3)

Here, E [•] represents an average value.

(Step 2) Of J [Fm (1)], J [Fm (2)]... J [Fm (I + 1)] obtained in the above (Step 1), the largest one is J [Fm (i *)], Fm (i *) is updated by the following equation.

[0019]

(Equation 4)

Using the updated Fm (i *), J
[Fm (i *)] is obtained.

(Step 3) As a result of the processing in the above (Step 2), J [Fm (1)], J [Fm (2)]... J [Fm
If (I + 1)] is J [Fm (i *)], Fm (i *) is updated again by the following equation.

[0022]

(Equation 5)

Thereafter, the process proceeds to (Step 5) described later.

(Step 4) As a result of the processing in the above (Step 2), J [Fm (1)], J [Fm (2)]... J [Fm (I +
1)], if the largest one is J [Fm (i *)], Fm (i *) is updated again by the following equation.

[0025]

(Equation 6)

(Step 5) If m ≠ M, m ← m + 1 If m = M, m ← 1 is set and the process returns to (Step 1).

Then, the speaker output ym (k) of the m-th speaker
Is also obtained by the above equation (2). However, in this case, any one of Fm (1), Fm (2)... Fm (I + 1) is used as Fm (k).

The noise control based on the algorithm using the simplex method described above is also explained with reference to a block diagram shown in FIG.

Here, the case of one speaker and one microphone is described as an example.

The speaker 3 outputs a control signal z based on a speaker output y obtained by processing the reference signal x with a filter F (specified by the adaptive algorithm A). The synthesized noise d is observed and input to the adaptive algorithm A as an error signal e. Therefore, the control signal z output from the speaker 3 and the synthetic noise d cancel each other, so that a noise reduction effect can be obtained. Here, the filter F and the adaptive algorithm A are
A signal generating means for generating the speaker output y is configured.

In the noise control using the above algorithm, it takes a certain amount of time to proceed to a state where the noise is sufficiently reduced (in other words, to converge).

However, the noise in the vehicle compartment is not always constant. For example, if the engine speed or the vehicle speed changes, the noise level and the structure of the frequency characteristics also change. Therefore, unless the speaker output in the noise control follows the change in the vehicle interior noise, good noise control cannot be performed.

For the above reasons, it is desirable to minimize the time required for noise control (in other words, the convergence time). However, if the conventional noise control method is used as it is, it becomes a control target. The convergence time only increases with the increase in the number of seats.

[0034]

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problem, and stores a result of adaptive control in noise control using a factor correlated with a noise state as a parameter. The next time the parameter becomes the same, the previously stored intermediate result is called, and by restarting adaptive control from this state,
It is an object of the present invention to reduce the convergence state in noise control as much as possible.

[0035]

To achieve the above object, the present invention has the following configuration. That is, a microphone for collecting sound in the vehicle interior, a speaker provided in the vehicle interior, reference signal detection means for detecting a reference signal related to vibration of a noise source, and a microphone for collecting the sound based on the reference signal. A signal generating means for generating a speaker output to reduce a sound to be sounded, comprising: a correlation factor detecting a correlation factor correlated with a state of noise detected by the microphone. Detecting means; storing means for storing an adaptive filter coefficient vector determined in the course of signal generation by the signal generating means in association with the correlation factor detected by the correlation factor detecting means; When the adaptive filter coefficient vector changes, the adaptive filter coefficient vector corresponding to the changed correlation factor is read out from the storage means. There as the signal generating means and control means for outputting a command to resume the signal generator is attached by, as constructed in accordance with.

[0036]

According to the present invention, the following effects can be obtained by the above means.

That is, during the signal generation in the noise control, the adaptive filter coefficient vector for generating the speaker output signal is stored in the storage means each time in correspondence with the correlation factor correlated with the state of the noise. When the factor has changed by a predetermined amount or more, the adaptive filter coefficient vector corresponding to the changed correlation factor is read from the storage means, and the speaker output signal generation is restarted based on the adaptive filter coefficient vector. Therefore, after the correlation factor (in other words, noise) changes significantly,
This eliminates the need to generate the speaker output signal from the beginning.

[0038]

According to the present invention, a microphone for collecting sound in a vehicle compartment, a speaker provided in the vehicle compartment, reference signal detecting means for detecting a reference signal related to vibration of a noise source, and A signal generating means for generating a speaker output for reducing a sound collected by the microphone based on a reference signal; The adaptive filter coefficient vector for generating the speaker output signal is stored in the storage means in each case in correspondence with the correlated correlation factor, and when the correlation factor changes by a predetermined amount or more, it corresponds to the changed correlation factor. The adaptive filter coefficient vector is read from the storage means, and the generation of the speaker output signal is restarted based on the adaptive filter coefficient vector. Therefore, after the correlation factor (in other words, noise) changes significantly, there is no need to generate the speaker output signal from the beginning, and the excellent effect that the convergence time of the noise control can be greatly reduced. There is.

[0039]

Preferred embodiments of the present invention will be described below with reference to the accompanying drawings. Embodiment 1 FIG. 2 shows a layout of each device in a vehicle interior, wherein reference numeral 1 denotes a microphone that collects sound in the vehicle interior, and reference numeral 2 denotes a reference that detects a reference signal x related to vibration of a noise source. The signal detection means 3 is the reference signal x
, A speaker for outputting a signal to reduce the sound collected by the microphones 1, 1,.
Operation switch for specifying / OFF and operation mode,
Reference numeral 5 denotes a controller for performing noise control, and reference numeral 6 denotes a correlation factor detecting means for detecting a correlation factor correlated with the state of noise.

A total of four microphones 1 are provided, two on each of the driver's seat and the passenger's seat, and two on the rear left and right of the passenger compartment.

As the reference signal detecting means 2, an acceleration sensor for detecting engine vibration which detects the vibration of an engine which is a noise source is used. Note that an ignition pulse detection circuit may be used instead of the acceleration sensor.

One speaker 3 is provided at the center of the front of the passenger compartment.
A total of five audio speakers are provided, one at the left and right at the front of the cabin and two at the left and right at the rear of the cabin.

As the correlation factor detecting means 6, an engine speed sensor for detecting the engine speed Ne is used. Note that a vehicle speed sensor may be used instead of the engine speed sensor.

As shown in FIG. 3, the controller 5 has a CPU 11 comprising a microprocessor such as a digital signal processor.
Is to execute control according to a program stored in the memory circuit 12. Reference numeral 13 denotes an A / D conversion circuit for converting analog signals from the reference signal detecting means 2 and the microphones 1, 1,.
14 is a D / A conversion circuit for converting a digital signal from the CPU 11 into an analog signal, and 15 is a fixed cycle (for example,
A clock circuit for supplying a clock signal to the CPU 11 in 500 μs to 1 ms), and 16 and 17 are amplifiers.

As shown in the function correspondence diagram of FIG. 1, the CPU 11 generates a signal for generating the output signals of the speakers 3, 3,... According to the LMS algorithm [ie, the above-described equations (1) and (2)]. Means 21 and an adaptive filter coefficient vector determined in the course of signal generation by the signal generation means 21. The correlation factor detected by the correlation factor detection means (engine speed sensor in the present embodiment) 6 (this embodiment). In the case of the example, when the correlation factor (ie, the engine speed Ne) changes by a predetermined amount or more, the storage unit 22 that stores the correlation factor in correspondence with the engine speed (Ne), and the changed correlation factor (ie, the engine speed). Ne) and an instruction to read out the adaptive filter coefficient vector corresponding to Ne) from the storage means 22 and restart the signal generation by the signal generation means 21 based on the adaptive filter coefficient vector. And a control unit 23 for outputting.

As shown in FIG. 4, the signal generating means 21 includes a filter F for processing the reference signal x,
The adaptive filter coefficient vector Fm (k) in the filter F
Is determined by the above-described equation (1).
A filter H for giving a speaker / microphone transfer characteristic. The filter H processes the reference signal x and sends it to the adaptive algorithm A as a signal r. Here, the case of one speaker and one microphone is described as an example.

Then, the speaker 3 outputs a control output z based on the speaker output y obtained by processing the reference signal x with the filter F, and the microphone 1 outputs the control output z
And the synthetic noise d are observed and input to the adaptive algorithm A as the error signal e. Further, the reference signal x is processed by the filter H and is used as an adaptive algorithm A as a signal r.
Is input to Therefore, the control output z output from the speaker 3 and the synthetic noise d cancel each other, so that a noise reduction effect can be obtained.

In this embodiment, the engine speed sensor 6
Is input to the CPU 11, and the filter coefficient vector Fm (k) of the adaptive filter F
Has been input to the storage means 22 for storing.

Next, the operation of the vehicle noise control device according to this embodiment will be described with reference to the flowchart shown in FIG.

[0050] The foregoing equation (1), read (2) When noise control based on is started, the instantaneous value Ne of the engine speed detected by the engine speed sensor 6 in step S _1, step S the instantaneous value Ne a reference engine speed Ne ₀ at _2. Then, again reads the instantaneous value Ne of the engine speed in step S _3, the difference between the with the instantaneous value Ne reference engine speed Ne ₀ (i.e., |
Ne-Ne ₀ |) and comparison between the change amount reference value ΔNe of the engine speed is performed in step S _4. Where ΔNe
> | Ne-Ne ₀ | If it is determined (i.e., when the engine speed has not changed significantly), the process proceeds to step S _5, the adaptive noise control in accordance with the LMS algorithm is implemented.

[0051] On the other hand, ΔNe ≦ in step S ₄ | Ne
-Ne ₀ | and when it is (that is, when the engine speed greatly changes), the reference value the engine speed N before the engine speed is largely changed in step S ₆
e ₀ the filter F of the adaptive filter coefficient vector Fm when as an index (k) is stored in the storage means 22. Then, replace the engine speed instantaneous value Ne after the change at step S ₇ to the reference engine speed Ne _0, already the stored adaptive filter coefficient vector Fm the reference engine speed Ne ₀ as an index in step S ₈ (k) is read from the storage means 22. Thus read adaptive filter coefficient vector Fm (k) is transferred to a filter F, adaptive noise control from the adaptive filter coefficient vector Fm (k) is resumed in step S _5.

As described above, in this embodiment, the adaptive control of the noise is performed while observing the engine speed Ne from time to time, and the change in the engine speed Ne exceeding a certain value (a preset reference value ΔNe) is obtained. Occurs, the adaptive filter coefficient vector Fm (k) is stored using the engine speed Ne before change as an index, and when the engine speed Ne returns to the previously stored state, The adaptive filter coefficient vector Fm (k) stored with the engine speed as an index is transferred from the storage unit 22 to the filter F, and the adaptive noise control is restarted from that state.

Therefore, after the engine speed (in other words, noise) has changed significantly, it is not necessary to generate a speaker output signal from the beginning, and the convergence time of noise control can be greatly reduced. -ing

Second Embodiment FIG. 6 is a block diagram of a signal generation unit in a vehicle noise control device according to a second embodiment of the present invention.

The signal generating means 21 of this embodiment includes a filter F for processing the reference signal x and an adaptive algorithm A for determining an adaptive filter coefficient vector Fm (k) in the filter F by the above-mentioned equation (1). Have.

The other configuration, operation, and effect are the same as those of the first embodiment, and the description is omitted.

The invention of the present application is not limited to the configuration of each of the above embodiments, and it is needless to say that the design can be changed as appropriate without departing from the gist of the invention.

[Brief description of the drawings]

FIG. 1 is a function correspondence diagram in a vehicle noise control device of the present invention.

FIG. 2 is a layout diagram illustrating an overall configuration of the vehicle noise control device according to the first embodiment of the present invention.

FIG. 3 is a block diagram of a controller part in the vehicle noise control device according to the first embodiment of the present invention.

FIG. 4 is a block diagram illustrating a signal generation unit in a controller of the vehicle noise control device according to the first embodiment of the present invention.

FIG. 5 is a flowchart illustrating an operation of the vehicle noise control device according to the first embodiment of the present invention;

FIG. 6 is a block diagram illustrating a signal generation unit in a controller of the vehicle noise control device according to the second embodiment of the present invention.

FIG. 7 is a block diagram illustrating an example of a signal generation unit in a controller of a conventional noise control device.

FIG. 8 is a block diagram showing another example of the signal generation means in the controller of the conventional noise control device.

[Explanation of symbols]

1 is a microphone, 2 is a reference signal detector, 3 is a speaker, 5 is a controller, 6 is a correlation factor detector (engine speed sensor), 21 is a signal generator, 22 is a storage, and 23 is a controller.

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G10K 11/178 B60R 11/02 F01N 1/00 F01N 1/06

Claims (1)

(57) [Claims] A microphone for collecting sound in the vehicle interior; a speaker provided in the vehicle interior; reference signal detecting means for detecting a reference signal related to vibration of a noise source; A signal generating means for generating a speaker output for reducing sound collected by a microphone, comprising: a signal generator configured to detect a correlation factor having a correlation with a state of noise detected by the microphone. Storage means for storing an adaptive filter coefficient vector determined in the middle of signal generation by the signal generation means in association with the correlation factor detected by the correlation factor detection means; When the adaptive filter coefficient vector has changed by a predetermined amount or more, the adaptive filter coefficient vector corresponding to the changed correlation factor is read from the storage means. Vehicle noise control apparatus characterized by a control means for outputting a command to resume the signal generation by the signal generation means based on the Le is attached.