close

Вход

Забыли?

вход по аккаунту

?

JPS61212996

код для вставкиСкачать
Patent Translate
Powered by EPO and Google
Notice
This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate,
complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or
financial decisions, should not be based on machine-translation output.
DESCRIPTION JPS61212996
[0001]
"Industrial field of application" The invention of the present invention is that the sound pressure
of one or two or more points set in an arbitrary three-dimensional space identical to this with
respect to the sound wave emitted from the speaker element The present invention relates to a
multipoint radiation type sound pressure control apparatus that produces a desired sound
pressure distribution by controlling an input signal that supplies a waveform to the speaker
element. [Conventional technology-1 From recent research on "sound image localization", control
the sound pressure at multiple points using multiple speaker elements based on the evaluation
error of "square error minimum" Concept of the method of constructing the device [Yanagi et al .:
J. Eight coust, soc, Jpn (E), 4, 2, 107-109 (1983)). Here, the concept of this method will be
described by developing it on a 1-time axis ?L ?. The following description is based on the
simple model shown in FIG. 1I. Although FIG. 11 shows the conventional method of controlling
the sound pressure at one point for the sake of simplicity, the method for two or more points is
based on the same principle. 11. In FIG. 11, the microphone element 11. placed in the sound field
where the room reflection sound exists. The sound pressure of the virtual speaker element
S.degree. Received at step S.sub.0 is determined using the speaker element 1.sub.1 without using
the virtual speaker element S ". Can reproduce the situation as if the acoustic signal is being
emitted from the virtual speaker element S 'despite the fact that the speaker element 11 emits
the acoustic signal. In this specification, speaker control for creating such a situation is called
"sound pressure control". In order to create the above situation, the signal transfer characteristic
between the virtual speaker element S and the microphone element 111, that is, the impulse
response, and the speaker element 1. The coefficients of the FIR filter (transversal filter) 21 ?
are appropriately determined so that the impulse response between the microphone element 113
and the microphone element 113 becomes equal, and the signal is transmitted to the speaker
phase element 1 through the FIR filter 211. It may be supplied to the Now, let the impulse
08-05-2019
1
response between the speaker element S 'and the microphone element 11 ░ be r (k) (k = 1.2.-=-,
N), and the speaker element 11 and the microphone element 11. Filter coefficient h (k) (k -1, 2----, T) satisfying the following equation, where g (k) (k = 1, 2,-, m) between The desired sound field
control is realized by applying,, 'to the FIR filter 211. r (k) = g (k) h (k)-------------------- Represents
an operation.
In the formula (1), r (k) is (r, rz-"-rn)" (T represents transposition), g (k) is Cg + gz-gm) ", h (k) is (h
+ hz-ht) If it is represented by T and a vector represented by each discrete value, (1) expression%
expression% is rewritten. However, when the filter coefficient hI + 2 +-'-+ hl is obtained from the
(I) equation, the number n of equations is larger than the number of unknowns, so the equation
(1) is usually an impossible equation Understanding that this expression is satisfied, h2. It is
impossible to obtain ?, h,. In the prior art, therefore, a filter coefficient that minimizes the
squared error represented by 11e (k) 12-? giant (k)-g (k) h (k)-+21, using "minimum squared
error" as the evaluation scale Sound pressure control is performed to obtain h (k) and to
approximately realize a desired impulse response r (k) using this. This h (k) is obtained by the
following equation. However, this method is intended to minimize the (al 2 error, and
theoretically includes an error due to the inability to satisfy the equation (1). The error of (blta +)
is greatly influenced by the positional relationship between each speaker element, each
controlled point, and the surrounding environment. There was a drawback of that. An object of
the present invention is to set an arbitrary signal transfer characteristic, ie, an impulse response,
determined for each controlled point at one or more controlled points set in any threedimensional space. An object of the present invention is to provide a sound pressure control
device capable of accurately creating a desired sound pressure distribution by realizing the
method without approximation errors as in the conventional method. [Means for Solving the
Problems] According to the present invention, the speaker elements are arranged at a plurality of
points in the space, the same human power signal is supplied to these speaker elements through
the respective different filters, and the same space as these speaker elements The desired signal
transfer characteristics defined for each of the points (hereinafter referred to as controlled
points) at least one less than the number of speaker elements, and the signal transfer between
each of the speaker elements and the controlled points The coefficients of the respective filters
are calculated using the characteristics, and the coefficients are set as the corresponding filters,
whereby the manual signal is transmitted to the desired points at the controlled points at the
desired signal transmission. It is controlled to satisfy the characteristics. Conventionally, the
number of loudspeaker elements and the number of controlled points are the same, or the latter
is large. However, in the present invention, the number of controlled points is always smaller
than the number of loudspeaker elements.
Principle of the Invention First, the present invention will be described in detail with reference to
FIG. In FIG. 1, eight (S ? 2) speaker elements 11 + 12 +-----+ + I S are disposed in space. Also, N
08-05-2019
2
controlled points in the space 11. ,-, 11N are provided. In FIG. 1, when the microphone elements
are disposed at the controlled points 11 ░,-, 11N, respectively, the controlled points LL,----11u
may be referred to as microphone elements. ???????????? The same input signal is
applied from the input terminal 12 to the speaker elements 11.1 ? and -1s through the FIR
filters 211, 212 ░-and 213, respectively. By controlling the filter coefficients of these FIR filters
21+, 21 ? and -21 ░, the input signals applied to the speaker elements 11, 1 ? and -'- 1s can
be controlled separately. First assume the following conditions. This is generally fulfilled in a
normal acoustic environment. (Condition-1) Eight speaker elements 11 + 1z + -31s and N
controlled points 11. , And 11 N can all be represented by discrete time series having the same
finite length m (hereinafter, such discrete time series will be referred to as "hector"). (Condition2) Transfer functions obtained by Z-transforming each of the impulse responses described in
(Condition-1) do not have the same zero point. Now, N desired impulse responses set for each of
the controlled points,-, and IN are represented by r, (k) U = 1.2. ? N) and eight speaker elements
1. , And S-N impulse responses between I S and N microphone elements 11+ and '-' II N are gij (k)
(i = 1.2.-3; j = 1, 2. -, N), and filters 21. provided in front of each of the eight speaker elements. ,21. Each filter coefficient of h + (k) (i = 1.2. If it is set as ---, S, then filter coefficients hi (k) which
satisfy rJ (k)-. SIGMA.gtJ (k) .hi (k) ---- m =-141 are accurately determined. If so, the
disadvantages of the prior art can be overcome. ??????????????????
??????????
???????????????????????????????????
???????????????????????????
???????????? Equation (4) becomes n = m + 1--1 if it is expressed by the vector-, N).
In order to facilitate the description below (the left side of the 4F equation is 1 (N-nxl vector), the
matrix on the right side is G (N-nXs-L matrix), the vector on the right side is ? (vector of S-LXI) If
it does, (4) type becomes (4 bit type. ??? = G-111-------+ 41 "(noting the matrix on the right
side of the 4-bit equation, the condition that the number of columns is larger than the number of
rows, that is, the unknown is the same as the number of equations or- The condition for
obtaining more solutions and obtaining a solution, S?L ? N?n = N (m +! ?1) 4 ░, L ? N
(m?1) / (S?N 2) ?????? [51 If the filter length satisfying the condition is determined,
(Condition-2) The rank of the matrix G is N и (m ? ?L?1). Therefore, the equation (41) can be a
normal equation% equation% equation when fal L = N (m-1) / (S-N), and can be solved as follows.
(In the case of the normal equation) Since the inverse matrix of the matrix G exists, the equation
(6) may be calculated. IT = Gl; published и I'R---------------------(6) (in the case of an indeterminate
equation) The inverse matrix of the matrix G It can not be asked as it is. (The vector BI satisfying
the equation 41 is not unique. For example, 1-norm minimal general inverse matrix) (Yanagi et al
.: ?projection matrix, general inverse matrix, singular value decomposition?. Using the Todai
Press, 65-68 (1983)), 11 (-[1; T ((G-(1; D)-1 и Il !! The solution of --- --- --- (6) 'can be obtained.
That is, the most important feature of the present invention is the basic equation (4) or (4) for
sound pressure control by using control speaker elements and filters whose number is greater
08-05-2019
3
than the number N of sound pressure controlled points. A filter coefficient satisfying the above is
obtained, and as a result, the impulse response determined for each of the controlled points can
be accurately reproduced by the eight speakers, not by approximation. Therefore, the present
invention, in principle, overcomes the drawbacks of the conventional method, which can
reproduce only an approximate impulse response at each controlled point. Embodiment 1 FIG. 2
shows an embodiment of the present invention, which is an example in which the whole system
is constituted by a digital signal processing system. In FIG. 2, an input signal U (t) to the
multipoint radiation type sound pressure controller (the device of the present invention) is
supplied from the input terminal 12 to the A / D converter 30, and the A / D converter 30 The
signal U (k) obtained by discretizing the input signal U (t) is output, and these are branched into
eight identical signals U + (k) (+ = 1, 2,-, S) and eight Fll?
Filter 21. , 21 ?, ?, 21 ░, respectively, these filters 21. , 21 ?. The outputs of 213 are D / A
converters 50I respectively. 50 g,---, 8 speaker elements 1 through 50 s. It is supplied to 1 ?, ? и
?1, (S ? 2). 11. One or more number N ('N ? 5-1) controlled points In this example, N
microphone elements are used as,. , -11. Of the eight speaker elements 18.1 ? + '-"+ I S and N
controlled points 111,-, 11 via the switch 200 respectively. And S-N impulse response vectors g, j
(i = L2,-,). ???????? The signal is supplied to a waveform recorder 60 which stores-, N).
The output of the waveform recorder 60 and each controlled point 11. ,-. The outputs of the
waveform recorder 61 storing desired N impulse response vectors (hereinafter referred to as
target impulse response hectles) determined for each 11N are supplied to the waveform
manipulator 70, respectively. The output of the waveform operator 70 is supplied to the
calculation setter 80. The output of the operation setter 80 is supplied to the filter coefficient
determiner 100, which outputs eight FIR filters 21+, 21z,-'. 21 is supplied to the corresponding
setting input side. The operation of this embodiment will now be described. In this example, eight
FIR filters 21.1, 21z,-. 21, the first input signal U, (k) (+ = 1, 2, '-и и. It is assumed that S) is set to
output immediately without processing. In the waveform recorder 60, an impulse response
vector (hereinafter referred to as a room impulse response vector) between the speaker elements
II, ?, 1 and 1s and the controlled point 11 ? ? ? + 11N is opened and closed as necessary. It is
assumed that N target impulse response vectors, which are stored by closing the unit 200 and
stored in the waveform recorder 61 for each of the controlled points IL,-, and llo, are stored in
advance. The waveform manipulator 70 performs the following processing [1] to [4] using the
room impulse response vector and the target impulse response vector stored in the waveform
recorder records 60 and 61, respectively. [1] Indoor impulse response vector (gij (-L2. ??
???????? ????? For each of g rJ-[gr J Ig i, + ?-g = ;,] "(m" 2 m)), a component of a
predetermined reference value ? (g 0 0) and the room impulse response vector g ij From the
beginning, grjp> V (P = 1, 2,-)---------171, the value g that satisfies the first until it is 42 is
considered as noise, and from the beginning to P-1 The element of is replaced with 0, and while
the group delay D i j is D 1,,, = P -1------------------- ------+ + 81 to Seso 1.
08-05-2019
4
[2] For each of the target impulse response vectors (fJ (j-1, 2,-, N), ti = [rs ++ rj 2 + 1 rjn '] "(n' ?
n)), the same processing as in [1] is performed. Then, the group delay D'4 is obtained. Not only
when the target impulse response vector is given by calculation in advance, for example, when it
is desired to cancel an acoustic signal from a certain sound source, the impulse response between
each sound source and each controlled point is the target impulse response vector. Ask. [3]
Perform the processing of equation (9) for the group delay obtained in [1] and ? J. D s = MIN (D
I j + D z j,---, D s J)----------(91 MIN (at l z,-1 aq) i a Q (q-1, 2 + 1----- , And> select the smallest value
a9 among the functions [4] j (j = 1.2. N) Remove one element from the head of each of the
impulse response vectors for the target impulse response vector f determined at the controlled
point and the room impulse response vector g. That is, the same delay amount common to each
impulse response vector is removed. However, if D d ? D is not satisfied, accurate target impulse
response can not be reproduced in the control described later. It is because of the physical
reason that it can not "control a signal which is advanced with a signal delayed in time". The
processes of [1] to [4] described above are devices for determining the rank of the equation (4)
? ? described above by the number of rows of the matrix G. Next, the waveform manipulator
70 outputs the target impulse response vector and the indoor impulse response vector which
have been subjected to the above processes [1] to [4] to the operation setting unit 80. Here, the
room impulse response can be represented by a vector of length m, as for the two conditions
(condition-1) mentioned above. (Condition-2) Transfer functions obtained by Z-transforming the
room impulse response are the same and do not have zeros. Shall meet The calculation setter 80
uses the input from the waveform manipulator 70 to perform the following process [61 to [9]. [6]
Using the predetermined length m of the room impulse response vector, the length n of the
target impulse response hector, the number S of speaker elements, and the number N of
controlled points, Fil? Filter 21. , 21 ?, ?, 21. Determine the required tap length.
L ?? ? N (m-1) / (S-N)---------40) n = m + L-1------ Find a satisfactory tap length. However, in the
formula 001, in the case where it is only required to obtain n <m О L ? ? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? Then, the target impulse response vector is
appropriately zeroed to create a vector of such a length as to satisfy the equation (00), and this is
made into the target impulse response hektor again. [71N target impulse response vectors are
used to create a target impulse response vector R having a length of N и (m + L?1) as shown on
the left side of equation (4). [8] Using S-N room impulse response vectors, find the matrix of (N.
(m + L-1), 5-L) as shown on the right side of equation (4). [9] For L defined in [6], (when L = N-(m1) / (S-N)) (N и (m ? L-1), N и (m + L-1)) Create an identity matrix of (When L> N и (m ?1) /
(S?N)) Transpose matrices G and G of G are formed. After performing the processes of [61 to [9]
above, the calculation setter 80 outputs I?R, CG, GIFT (or ?) to the filter coefficient determiner
100. As shown in FIG. 3, the filter coefficient determiner 1oo comprises three matrix multipliers
101, 102, 103, one inverse matrix calculator 110, and a coefficient distributor 120. The
operation of the filter coefficient determiner 100 will be described below with reference to FIG.
08-05-2019
5
[10] Matrix multiplier 101 (7) CT (or ?) is input to input terminals 101-11m [1 ;, 102-2], their
product is calculated, and c-, cT (or G) are output Ru. [11] (([;-([T)-1 (or G-1) is obtained from the
output of the matrix multiplier 101 by the inverse matrix calculator 110 and is output to the
matrix multiplier 102). [GT (or ?) is input to the input terminal 102-1 of the 12-co matrix
multiplier 102, and cT и (G и (CT) ?1 (or (G ?1) is converted as an output, which is a matrix
multiplier It is output to 103. [13] R is input to the input terminal 103-1 of the matrix multiplier
103, and as a result, a filter coefficient ? such as clothing is obtained. ??????? ----- S, [14]
The coefficient distributor 120 has eight filters 21. .
The filter coefficients Ih, Ihz,-, Ihs obtained by the equation Q2) are distributed to 21 ?,-, 213. By
performing the processes of [1] to [14] described above, the FIR filter 21. , 21 ?, and 21S, if an
input signal U (t) is applied to the input terminal 12 of this device, the speaker element 1. . The
acoustic signals emitted from 1 ?, ?, 1s completely reproduce the characteristics of the target
impulse response defined for these points at N controlled points 111 ░ ? ? ? 11 It will be
done. By the way, for the calculation of the filter coefficient, various successive approximation
methods can be used other than the above-mentioned processes of [1] to [14]. When using the
successive approximation method, it is also necessary to pay attention to the convergence of the
algorithm, but the inverse matrix as described above or the norm minimum generalized inverse
matrix [z? Compared to the method of directly obtaining ((1;-(CT)-'thing), it is advantageous in
terms of computational complexity and memory capacity. In the case of the successive
approximation method, the filter coefficient determiner 100 in FIG. 2 may be replaced with a
sequential filter coefficient determiner 130 as shown in FIG. The successive filter coefficient
determiner 130 is composed of a sequential operation circuit 131 having a sequential calculation
and a coefficient distributor 120. In the sequential operation circuit 131, for example, the
following steepest descent algorithm is used [Fazeev et al .: Calculation method of linear algebra,
industrial book. (1976)) filter coefficients can be determined. (In the case of a normal equation or
indeterminate equation) ?; a parameter ? (?) representing the number of iterations of the
algorithm; a convergence coefficient (in the case of a normal equation) ? (?) 5 convergence
coefficient , Eight successive filter coefficient determination circuits 141 ? 1412 as shown in
FIG. It is also possible to configure this with-, 141s and the error operation circuit L0. The error
calculation circuit L0 calculates an error Hector according to the following equation. ? ? ? ?
? ?; the number of iterations of the algorithm is set to a table parameter [ii impulse response
vector gij between the first speaker and N control points (j = 1.2. The following matrix created
using-, N). 1h, (?); Fll connected to the first speaker element? Filter coefficient vector and eight
successive filter coefficient determination circuits 141. . ---, 141, finds the filter coefficient by the
following algorithm. (In the case of a normal equation and an indeterminate equation) 1 h, (? ?
?1) = 1 h, (?) + ? (?) и g + T?E (?) ~ ????????? 051 (of the normal equation Use it
in case.
08-05-2019
6
) Ih 1 (? + 1) ?1h, (?) + ? (?) и st (?> ?? ? ? ? ? ? ? ? ? ? ? ? ? 05) (E (? ) [El
(?) ll! 2 (? ? ?) -'- e N, Im + L-11 (?)] 7, N- (m ? L-1) -3-L, so E '(?) L from the beginning It
can be separated. If this is [ffi, T (?) @, "(?> ? ? T (?)], then ei (?) is the first element vector
of this vector. In this method, eight arithmetic circuits 141. , 1413 calculate in parallel, the
amount of calculation required for the determination of the filter coefficient is Q31. Compared to
the case of calculating ?Q3)? equation, it is apparently 1 / S. Therefore, the time required to
calculate the filter coefficient can be significantly reduced. [Embodiment 2] The above-described
embodiment 1 has acoustic signals radiated from eight (S ? 2) speaker elements, at N (N ? 5-1)
controlled points. Arbitrary sound pressure distribution is realized by controlling so as to satisfy
an arbitrary signal transfer characteristic (target in-plus response) defined for each controlled
point. However, when performing control to "set the sound pressure of all N controlled points to
zero", it can be understood from the fact that R in ?2) becomes a zero vector, the eight FT11
filters 21+,- , 21 and the tap coefficients are all zero, practically meaningless control is
performed. However, in the control such that the sound pressure of all the N controlled points is
kept at zero and the acoustic signal is present at a place other than the controlled point,
especially in the teleconference system etc. It is extremely demanding as a means to prevent
"howing" caused by acoustic coupling. Therefore, another configuration example of the present
invention which enables the above control by slightly modifying the apparatus shown in Example
1 is shown in FIG. In FIG. 6, parts corresponding to those in FIG. 2 are assigned the same
reference numerals and duplicate explanations are omitted. In FIG. 6, the input signal U (t) from
the input terminal 12 is also supplied to the delay circuit 160 and its output is supplied to the
speaker element 9. The delay circuit 160, N controlled points from the speaker element 9; , And
11N are respectively emitted from eight speaker elements 11.1.quadrature. , And IIN are
provided to provide an appropriate amount of delay to arrive later than any of the acoustic
signals emitted respectively.
In other words, the speaker element 9 and the N controlled points 111,----, 11. Speaker element
1. . 1 ?. It means that the delay circuit 160 is unnecessary if it is set so as to be farther than the
distances of-, i, and each of the controlled points 11 ░,-, 11N. The use of the delay circuit 160 is
to avoid contradictory situations such as "a signal delayed in time and a signal advanced beyond
this" is controlled. Also, a waveform recorder 62 is added, and is connected to the subtractor 170
together with the waveform recorder 61, and the output of the subtractor 170 is supplied to the
waveform operator 70. Next, the operation of this embodiment will be described. In this
embodiment, the waveform recorder 62 already contains the loudspeaker elements 9 and the
respective controlled points 1.1. , ?, 'I N N impulse response vectors r s (j = 1. 2. ---9N) shall be
memorized. The other parts are initialized as in the first embodiment. First, the impulse response
vectors rj and rj stored in the waveform recorder 61.62 (j = 1.2. ?? N) is subjected to the
operation of S, = ('j-rj----------------061) by the subtractor 170, and as a target impulse response
vector, C4 is input to the waveform controller 70. 06) In the equation, if rj is all zero vectors,
then the target impulse response vector is as follows, which corresponds to each controlled point
08-05-2019
7
from the speaker element 9], 11. -. The phases of N impulse response vectors rj up to 11N are
reversed. That is, this process obtains an impulse response with the phase of the vector 74
inverted. Thereafter, the same processing as in [1] to [14] described in the embodiment 1 is
performed, and eight FIR filters 21. ,-21. The coefficient of should be determined. ???????
????????? After the setting of the filter coefficients for the speaker elements 11 and 12
is completed. If an input signal is given from the input terminal 12 to-, 1, N controlled points 11. ,
-11. The acoustic signal at the point satisfies the target impulse response given by the surface
equation, and a situation is created as if the speaker element,-, 1, s are sounding in opposite
phase to the speaker element 9. Therefore, the speaker element 9 and the speaker element 1. ,
?, and 1s, the sound pressure at N controlled points IL, ? and IIN becomes all zero and the
controlled points 11. .
--- An acoustic signal will exist in places other than IIN. That is, by adding one or more speaker
elements configured to always emit an acoustic signal of finite amplitude to the apparatus
described in Example 1, N controlled points IL in the sound field,-, 11 N It is possible to control
so as to make only the sound pressure of the sound pressure zero, and emit an acoustic signal of
sufficient volume at other points. This indicates that the device of the present invention can also
be used as one of the means for preventing howling. Embodiment 3 In Embodiment 2, one or
more of the speaker elements 9 configured to emit acoustic signals of finite amplitude all the
time, as a method of preventing the ringing of the present invention as a method It explained that
it was effective. In the embodiment 3, another embodiment of the present invention will be
described in which howling is suppressed by nulling the output signal of each of the microphone
elements installed at N controlled points. FIG. 7 shows this embodiment. In FIG. 7, parts
corresponding to FIG. 2 are given the same numbers. In FIG. 7, a target impulse response
convolution circuit 180 is newly added. This circuit 180 represents a target impulse response
vector provided for each of N controlled points, ie, each microphone element, using an I'IR filter
or the like, and executes a convolution operation with a time series in real time. It is a thing. The
input signal 11 (k) is supplied to the input side of the target impulse response convolutional
circuit 1.80, and the output is supplied to the subtractor 190 together with the outputs of the
respective microphones 111,-, 11N. There is. Next, the operation of this embodiment will be
described. First, it is assumed that the filter coefficients for eight Fill filters 21+,-, 21s are set by
performing the processes of [1] to [] 4J shown in the first embodiment. At this time, if the same
input signal is applied to the target impulse response convolution circuit 180 and the speaker
elements 11, 1 ?, ? ? ?, 1 s, the output signals of the microphone elements 111, ?, lIN
reproduce the target impulse response. Therefore, the same signal is input to the subtractor 190
exactly. Therefore, the output signal of the subtractor 190 is zero, and the effect of suppressing
howling can be obtained as in the inventive device described in the second embodiment. In
addition, in this embodiment, N controlled points 111,-. There is no need to make the sound
pressure at 11 zero at all, so for the person in the space (controlled point 111,-, IIN) the speaker
element, the sound signal from-, 1, is sufficiently clear 11. Each controlled point to be given by
08-05-2019
8
gender, that is, to provide a human-friendly environment. , -11.
Has an extremely superior feature that it can suppress howling while controlling the This is a
great advantage that can not be achieved without the device according to the invention. At each
side described above, controlled points 11. ,-. 11N and the speaker element 1. If each impulse
response between,, and 1 is measured once, if there is no change in the acoustic environment
around these arrangement relationship 1, the controlled point is not necessarily installed. An
acoustic signal having a desired impulse response characteristic can be obtained. Next, the
results of experiments conducted under the arrangement of the speaker element and the
microphone element as shown in FIG. 8 will be described in order to confirm the effectiveness of
the device of the present invention. The experiment includes (experimental data) speaker
elements; three full-range speakers with a diameter of 10 cm, speaker spacing: 11 cm, one
microphone element; one shortest element distance between the speaker element and the
microphone element: 90 cm, room volume: 1.2. M, room average sound absorption coefficient;
0.8 + reverberation time; 75 m 5 ec, target impulse response Hector length; 400 points, indoor
impulse response Hector length; 400 points, impulse response measurement method, 10'OO
times synchronous addition 1 frequency band, 31 .5-3 L0 Hz, sampling period; 8 KHz, algorithm
used; 05) Equation 200 times repeated FIR filter; 399 taps. Under such circumstances, the sound
radiated from the speaker element 9 is erased using the speaker element 11.1. In the
conventional apparatus of this type, it is general to use one speaker element used to extinguish
the sound, that is, one speaker element for sound pressure control, but in this experiment, for the
sound pressure control of one point Two control speaker elements 11.1 ? are used. This is to
construct an apparatus satisfying the equation (5), which is a necessary matter as an
embodiment of the present invention. Further, in FIG. 9, the frequency characteristics of the
impulse response of the speaker element 11 and the microphone element 11, between the
speaker element 1 ? and the microphone element 11, and between the speaker element 9 and
the microphone element 11 are respectively curves A and B. Indicated by C. Here, an impulse
response (curve C) between the speaker element 9 and the microphone element II is taken as a
target impulse response. The frequency characteristic of FIG. 10 is an error waveform (curve E1
?) representing the difference between the impulse response synthesized using the speaker
element II and the speaker element 12 and the target impulse response, and the frequency
characteristic (curve C) of the target impulse response. Drawn on the basis of).
Also, in this figure, the conventional method shown in equation (2) is used as the speaker
element 1. . The frequency characteristic of the error waveform obtained in the experiment in
which each of the speaker elements 12 was used (an FIR filter of 1000 taps was used
respectively) ) Is also drawn simultaneously as curve Ell Bg. The power of each error waveform is
-11 dB (El), -19 dB (E2) when the conventional method is used, and one 256 B (Brz) when the
apparatus of the present invention is used, with respect to the power of the target impulse
08-05-2019
9
response. Met. From the above results, according to the present invention device, it was
confirmed by experiments that sound pressure control to the position of the microphone element
is realized with sufficient performance (performance) over the conventional method. In
particular, FIG. 10 also shows the howling margin as it is, and it has also become clear that this
device can be applied for howling prevention and that a very large howling margin can be
obtained. Moreover, it was also confirmed that the calculation amount required for filter
coefficient calculation can save the apparent calculation amount by the method of making the
filter coefficient determiner work in parallel by confirming the algorithm of the equation (05).
[Effects of the Invention] As described above, according to the device of the present invention,
the coefficient of the filter that performs appropriate control is accurately determined for the
input signal of each of the plurality of speaker elements, and It is possible to set the sound
pressure to be given to an arbitrary target value, and experiments have also confirmed that this
device exhibits sufficient performance in practical use. Also, since the device according to the
present invention can freely set the signal corresponding to the target impulse response, (1) it
differs from the space other than the sound pressure distribution by other acoustic devices in the
three-dimensional space where this device is installed. The sound pressure distribution in
another space (for example, another room) can also be reproduced. (2) A completely new sound
pressure distribution can be created. (3) A signal supplied to a speaker element which is in the
same space as this device and gives a target impulse response. The sound pressure of one or
more microphone elements set in an arbitrary three-dimensional space can be kept substantially
zero by making the signal of the phase opposite to that of the input signal of this device.
Therefore, it has the feature that it can provide one method of effectively preventing howling. In
addition, in configuring the device of the present invention, using a plurality of speaker units (for
example, woofer and tweeter) as speaker elements, combining a plurality of microphone units as
microphone elements used as controlled points (for example, directivity Using a microphone is
considered to be useful for improving the performance of the device of the present invention.
The signal processing system of the present invention can also be configured by a DSP or a
microprocessor that performs operations, delay procedures, and the like based on the
configuration described above. In addition, although this specification has been described on the
premise that an FIR filter is used as a filter, it is also possible to use an IIR filter or the like
instead of the FIR filter.
[0002]
Brief description of the drawings
[0003]
08-05-2019
10
FIG. 1 is a diagram for explaining the principle of the device of the present invention, FIG. 2 is a
block diagram showing one embodiment of the present invention device, and FIG. 3 is a block
diagram showing one embodiment of the filter coefficient determiner 100. Fig. 4 is a block
diagram showing one embodiment of a system for calculating filter coefficients by successive
approximation method, and Fig. 5 is a block diagram of the filter coefficient determiner of Fig. 4
by comprising a plurality of filter coefficient determination circuits. A block diagram showing one
embodiment of a system for calculating filter coefficients by the successive approximation
method with reduced amount of calculation above, FIG. 6 is a case where the sound pressure at
the controlled point is all zero and at a place other than the controlled point FIG. 7 is a block
diagram showing an embodiment of the device according to the present invention for emitting an
acoustic signal of a sufficient volume, and FIG. 7 shows that all output signals of the sound
receiving element placed at the controlled point are zero and the controlled point is Give an
arbitrary sound pressure distribution for the position of FIG. 8 is a block diagram showing an
embodiment of the present invention which is made possible, FIG. 8 is a diagram showing the
positional relationship between the speaker element and the microphone element in the
experiment, and FIG. 9 is between the speaker element and the microphone element used in the
experiment. FIG. 10 shows the frequency characteristics of the impulse response of the present
invention, FIG. 10 shows the results of experiments showing the frequency characteristics of
sound pressure control performance when using the device pertaining to the present invention
and the prior art, and FIG. It is a figure for demonstrating the principle of an apparatus.
?????? -, I S: two or more speaker elements, 9: speaker elements configured to emit a finite
amplitude signal, 11. ,-: 118: one or more controlled points, 21. , 21 ?, 21 :: FIR filter, 3 (1: A / D
converter, 50: D / A converter, 60, 61.62 .: waveform recorder, 70: waveform controller, 80:
arithmetic Setter, 100: filter coefficient determiner, 101, 102, 103: matrix multiplier, 110:
inverse matrix calculator, 120: coefficient distributor, 130: sequential arithmetic circuit, 131:
sequential arithmetic circuit, 141., 141 g ,-. 141,: Sequential filter coefficient determination
circuit, L0: Error operation circuit, 160: Delay circuit, 170, 190: Subtractor, 180: Circuit included
before target impulse response, 200: Switch. Patent applicant Nippon Telegraph and Telephone
Public Corporation Chief Kuso's correction of procedure (Spontaneous) Indication of 1 case on
September 26, 1985 Name of the Japanese Patent Application No. 60-538862 Invention multipoint radiation type sound pressure control device 3 correction Relationship with the Case Case
Patent Applicant Nippon Telegraph and Telephone Corporation 4 Representative Person
Shinjuku City, Tokyo 4-2-21, Sumo Building 5 Target of correction Details of the mid-summer
invention, column of claims, and 6 corrections Contents (1) The title of the invention "Tun-point
emission type sound pressure control device" is corrected as "multi-channel control device". (2,
Correct the claims as attached. (3) Specification, page 3, lines L to 20 "This invention relates to a
speaker element. "As below. "The two inventions have an impulse response that can be regarded
as finite, and seven linear systems can be obtained by controlling input signals applied to a
08-05-2019
11
plurality of output elements to a linear system that is theoretically multi-input multi-output. The
present invention relates to a multi-channel control device which makes it possible to set one
characteristic of an output signal at one or two or more controlled points to be a desired
characteristic. Examples of the linear system include an acoustic system including a plurality of
speaker elements and one or more sound receiving points, a radio wave system including a
plurality of transmitting antennas and one or more receiving antennas, etc. There is. In the case
where the present invention is applied to the above-mentioned acoustic system, the abovementioned plurality of output elements correspond to a plurality of speaker elements, and the
above-mentioned controlled point corresponds to a sound receiving point. The sound pressure
control to realize an arbitrary sound pressure distribution is performed at a point (hereinafter
referred to as a controlled point).
Hereinafter, the present invention will be described with reference to "one note pressure control"
as an example, but also for a system of music "one note", the corresponding relationship between
a plurality of input ends and a controlled point By defining it, exactly the same explanation can
be made. (4) Correct ?Speaker (to eye element 11)? to ?To speaker element 1 ?? on page 5,
line 9 of the document. (5) The same document, page 5, line 17 "sound field control is realized.
"The sound pressure control is realized. And correct. (6) In the same document, page 7, line 19
?The purpose of the second invention is arbitrary? 1. The purpose of the present invention is
corrected as ?arbitrary when stated for an acoustic system?. (7) The same document, page 8,
line 5 ?provides a sound pressure control device. "To provide a multi-channel controller. And
correct. (8) Line 8: lines 7-8 "each according to the invention ... each" "If the second invention is
described by taking an acoustic system as an example, speaker elements are arranged at a
plurality of points in space. , Correct each of these speaker elements. (9) In the same document,
page 8, line L, ?Correspondence of coefficient of heat? is corrected to ?1?. (10) In the same
document, page 9, line 5 "Speaker element" is corrected as "a speaker element which is an output
element." (11) ?The number of sound pressure controlled points... By eight speakers? in the
same document, page 13, lines 6 to 11 ?The number of output elements larger than the number
N of controlled points (in this case, the speakers By using the element and the filter, a filter
coefficient satisfying the basic equation (4) or (4) ? ? for output control can be obtained, and as
a result, the impulse response determined for each of the controlled points can be calculated. ,
According to the signals from the eight output elements. (12) In the same document, page 13,
line 19 ?Multipoint Radiation Type Sound Pressure Controller? is corrected as ?Multichannel
Controller?. (13) same document, page 14, line 7 "eight speaker elements 11. Speaker element
with 1S output elements. And correct. (L) p.17, lines 3 and 4 in the same document, ?this is a
vector, and this is determined. ??????????? And correct. (16) The 18th line, the 19th
line, the 20th line, the 2nd line, and the 8th line ['6'l] J of the same document are corrected as r
[5] J, respectively. (17) at page 18: line 18; page 20; line 2; line 8 "[9]" is corrected to r [8] 1
respectively. (18) In the same document, page 19, line [71J is corrected to [6 ':] J. (19) In the
same document, page 19, line 18 "[8]" is corrected as "[7]". (20) Correct the same document on
08-05-2019
12
page 20, line 16 [101 J as "[9]".
(21) In the same document, page 20, line 19 [[11]] is corrected to [10] J. (22) Correct the same
document on page 21, line 6 [12] as rclllJ. (23) In the same document, page 21, line 6 [1: D, j is
corrected as r, [] l) J. (24) The same document, page 21, line 13, line 1, line 22, page 5, line 28,
page 17, line 31, page 4, line [1, 4] J is corrected to [[13, 1) J, respectively. (25) In the same
document, p.31, line 19 ?correct by clarity? is corrected as (1) to ensure clarity. (26) ?The
Arashi?, page 35, pages 9 to 17 in the same document ? as explained иииииииииииииииииииииииииииииииииии ? As
described above for one or more of the present inventions, the device of the present invention
strictly controls the coefficients of the filter that appropriately controls the input signals of the
plurality of output elements of the multi-input multi-output linear system (i.e. the speaker
elements in the embodiment). This makes it possible to set one characteristic (sound pressure in
the embodiment) of the output signal from the controlled point less than the above-mentioned
number of output elements to an arbitrary target value, and this device is practically sufficient. It
has been confirmed by experiments that sound pressure control is also performed to
demonstrate good performance. Further, since the device according to the present invention can
freely set a signal corresponding to the target impulse response, particularly when used in a
sound pressure control system, "(27) ibid 36, line 11 [In addition, in configuring the device of the
present invention, ?When using the device of the present invention for sound pressure
control,? is corrected as ?.? (28) ibid., P. 36, line 17 [It is considered to be useful. Also, this is
considered to be useful. Father corrects this. (29) It is possible on page 37, line 4 ?minute. Then
join the following. In the specification H # of ":", the present invention has been described taking
sound pressure control as an example. The present invention is intended to control a linear
system having an impulse response that can be regarded as finite and having multiple inputs and
outputs in principle, and in the sound pressure control system described above, plural output
elements are plural. And one or more of the two controlled points are embodied as sound
receiving elements. Therefore, the same description as that described for the sound pressure
control system can be made by appropriately embodying the output element and the controlled
point with respect to the same (this also applies to the multi-input multi-output linear system.
That is, according to the present invention, it is possible to control the characteristics of the
signal at the controlled point less than the number of output elements arbitrarily and strictly with
respect to the linear system. Claims (1) A plurality of output elements connected to a linear
system which has multiple inputs and multiple outputs, and a plurality of filters which apply
outputs to each of these output elements and to which the same input signal is supplied, 1)
Desired signal transfer characteristics defined for elements less than the number of the plurality
of output elements connected to the linear system (hereinafter referred to as controlled points),
each output element, and each controlled point, and And calculating means for calculating and
setting the coefficients of the respective filters by calculation using the signal transfer
characteristics between the two, and the same input signal is given to the respective controlled
points at the respective controlled points. And a multi-channel controller for controlling to satisfy
08-05-2019
13
the desired signal transfer characteristics.
(2) Each of the plurality of output elements and the plurality of filters is S, the number of
controlled points is N, and each of the signal transfer characteristics between the output
elements and the controlled points is m Configured to handle as discrete signals, and each of the
S filters is configured by a discrete FIR filter having L taps given by L ? ? N и (m?1) / (S?N)
The multi-day controller according to claim 1, characterized in that: (3) A method for controlling
an acoustic system including S speaker elements and N controlled points using S speaker
elements instead of the output elements of the plurality of filters. Multi-channel controller
according to item 1.
08-05-2019
14
Документ
Категория
Без категории
Просмотров
0
Размер файла
35 Кб
Теги
jps61212996
1/--страниц
Пожаловаться на содержимое документа