JP3102229B2 - Signal processing 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 signal processing apparatus for processing a tone signal or the like and adding an effect such as reverberation.

[0002]

2. Description of the Related Art In recent years, a DSP (Digital Signal Processor) has often been used in order to impart an effect such as reverberation to a tone signal. The DSP can be configured into various filters by a microprogram, and has a feature that a high quality effect can be freely provided.

[0003]

By the way, in order to operate the DSP properly, it is necessary to give appropriate coefficients to each processing step, and if these are not appropriate, the desired effects cannot be sufficiently obtained. . However, the coefficients necessary for operating the DSP are enormous, and it is very troublesome for the operator to input all of them, and not only each coefficient is appropriate alone,
Unless the relations of all coefficients are properly set, desired operations are not performed, and it is very difficult to perform all of these operations at the discretion of the operator.

[0004] This is also true not only when the DSP is used, but also when a complicated filter is constituted by an analog circuit or a hard logic circuit.

An object of the present invention is to facilitate setting of coefficients in a complicated filter system.

[0006]

According to the present invention, there is provided a first signal path having first signal processing means for performing signal processing of a first processing characteristic, and a second signal path for performing signal processing of a second processing characteristic. A plurality of waveguides having a second signal path having processing means, an input terminal for inputting a signal to the waveguide, and a first signal path obtained by synthesizing a signal output from a first signal path of each waveguide; And a first feedback unit that outputs the combined signal to the second signal path of each waveguide and combines the signal output from the second signal path of each waveguide to output the combined signal to the second signal path of each waveguide. A second feedback means for outputting the combined signal to the first signal path of each waveguide and outputting the combined signal to the first signal path of each waveguide; and the characteristic of the signal output from the first output terminal and the second feedback means. of the signal outputted from the output terminal Wherein the sex first processing characteristics and pre
Signal processing means for an output signal to be balanced in accordance with the second processing characteristic .

[0007] The present invention is the above-mentioned invention, wherein
One of the signal processing means and the second signal processing means is a low-pass filter, and the other is an attenuator.

According to the present invention, in the above invention, the processing characteristics of the signal processing means for the output signal are obtained by averaging the processing characteristics of the first signal processing means of the plurality of waveguides.
Or at least one of characteristics obtained by averaging the processing characteristics of the second signal processing means. Further, according to the present invention, in the above-mentioned invention, the first signal path and the second signal path of the plurality of waveguides each have a delay unit, and the delay times of the respective delay units are set to a prime relationship. It is characterized by being.

[0009]

According to the present invention, an input signal is input to a plurality of first signal processing means and a plurality of second signal processing means,
Processing Japanese which is set in each of the signal processing means
Receive the signal processing in sex. Each of the first signal processing means and the second
The signal processing means is connected in a closed loop to form a waveguide.
Signal of each waveguide is
Returned to Ido. This allows the input signal to continue
And output from the output terminal. Each wave guy
Signal processed by the de is output after undergoing processing in signal processing means for the output signal. <br/> processing characteristics of the signal processing means for the output signal is determined based on the processing <br/> hydraulic characteristics of the plurality of first and second signal processing means (for example, is set to an average value There). Thereby, the first output terminal and
Never balance of signals output from the second output terminal is broken.

Further, as described above, having a first signal path and the second signal processing unit having said first signal processing means
The second signal paths are connected one by one in a closed loop . Thus, the input signal can be circulated through the first signal processing means-the second signal processing means and subjected to repeated processing, thereby realizing characteristics such as an impulse response. In addition, only the first signal processing means is provided on one side and the second signal processing means is provided on the other side, and both processes can be performed on the signal.

According to the present invention, the input signal includes a plurality of signals .
閉Ru while being delayed by the delay circuit in the waveguide
Circulated through the loop and transmitted to other waveguides
You. Here, the delay times of the plurality of delay circuits are set in prime relation by the delay time setting means. The prime relation refers to a relation having no common divisor other than 1, and this relation is all satisfied in two arbitrarily extracted delay times. If there is a delay time for a non-prime relationship,
Resonance occurs at a frequency whose cycle is the common divisor of the delay time, and the output signal becomes a spectrum in which the frequency is emphasized. On the other hand, by making the delay times relatively prime, such uneven signal processing is not performed.

[0012]

FIG. 1 is a block diagram of a signal processing apparatus according to an embodiment of the present invention. This signal processing device is a device which receives an analog tone signal, converts the analog signal into a digital signal, gives an effect by a DSP, and outputs the signal as a stereo analog signal. In this embodiment, it is assumed that a program is set in the DSP so as to function as a reverberation imparting device having a plurality of resonance systems, ie, waveguides. In addition, there are a large number of parameters (operation parameters) for determining the operation characteristics of the DSP. In this device, however, the large number of operation parameters are determined based on a small number of setting parameters for intuitively specifying the effect (reverberation) characteristics. The parameters are determined and set in the DSP.

C for controlling the operation of the entire signal processing apparatus
A memory 12, a reverberation characteristic setting unit 13, and a DSP 14 are connected to the PU 10 via a bus 11. The memory 12 stores an operation program of the CPU 10,
A storage area used for arithmetic processing performed by the CPU 10 and an area for storing setting contents of the reverberation characteristic setting unit 13 are set. The reverberation identification setting unit 13 includes a plurality of control operators, and can input setting parameters for setting reverberation characteristics described later. The DSP 14 is connected to a signal processing RAM 15, an A / D converter 16, and a D / A converter 17. The signal processing RAM 15 stores a microprogram for controlling the operation of the DSP 14 and operation parameters such as coefficients used in each step of the microprogram and the number of delay stages. These programs and parameters are written from the CPU 10 via the DSP 14 based on the setting contents of the reverberation characteristic setting unit 13 and the like. An analog tone signal is input to the A / D converter 16 from the outside. The A / D converter 16 converts this analog signal into a 16-bit digital signal and inputs it to the DSP 14. The DSP 14 performs predetermined processing on the digital signal, and outputs the digital signal to the D / A converter 17. D /
The signal output to the A converter 17 is a two-channel (stereo) signal. Although two signal lines from the DSP 14 to the D / A converter 17 are shown in the figure, the signal may be transmitted in a time-division manner using one signal line. D / A converter 1
Reference numeral 7 converts the input digital signal into a stereo analog signal and outputs it externally. For example, P
An analog audio device such as A is connected.

FIG. 2 is a circuit diagram showing the function of the DSP 14. That is, the DSP 14 is provided in the musical tone control RAM 15.
Functions as a digital circuit as shown in FIG. This circuit is composed of five waveguides WG1 to WG5. Each system is composed of a traveling system circuit and a reflection system circuit. To explain using the waveguide WG1 as an example, the traveling system circuit includes an adder AD11 to which a signal is input from the adder ADL of the left coupling unit, a delay circuit DELAYF1, a low-pass filter LPF1, a coefficient multiplier A11, and a signal multiplier. It comprises an adder ADF1 serving as an output terminal, and is connected in the above order. Further, the reflection system circuit includes an adder ADR in the right coupling section.
AD21, attenuator DMP to which a signal is input from
1, a delay circuit DELAYB1, a coefficient multiplier A21, and an adder ADB1 serving as a signal output terminal, and are connected in the above order. The coefficients input to the coefficient multipliers A11 and A21 are the coupling coefficients of the waveguide WG1. The coupling coefficient refers to this waveguide system (FIG. 2).
Is a parameter indicating the degree of contribution to the entire circuit).
Further, the output signal of the LPF1 of the traveling system circuit is negatively fed back to the AD21 of the reflection system circuit, and the DELAYB1 of the reflection system circuit is output.
Is negatively fed back to AD11 of the traveling system circuit. This negative feedback causes the signal to circulate (attenuate) within the waveguide. Among these functional elements, DELAYF1, DELAYB1, LPF1 and D
The characteristics of MP1 can be set by the reverberation characteristic setting unit 13. With such a configuration of the waveguide, it is possible to simulate the transmission characteristics of a tone signal in a tube and the acoustic reflection characteristics in a specific direction in a room. The structure of such a waveguide is described in detail in JP-A-63-40199.

In FIG. 2, waveguides WG2 to WG2
G5 has the same configuration as WG1. Note that a numerical value indicating the number of the waveguide is added to the end of the symbol attached to the functional element of each waveguide. A signal input from the outside (A / D converter 16) is input to the adder ADL and the adder ADR via the coefficient multipliers DINL and DINR. Further, signals of the traveling system circuits of the five systems (waveguides WG1 to WG5) are summed by the ADF1 and input to the adder ADR of the right coupling unit. further,
The signals of the five reflection circuits are summed by ADB1 and input to the adder ADL of the left coupling unit. As described above, AD
The addition output of R is input to the reflection circuit of each system, and ADL
Are input to the progressive circuits of each system, the signals in this circuit are AD1i, AD2i (i = 1
5), the signal circulates in each system, and the signal added for each of the traveling system (ADR) and the reflection system (ADL) circulates in all the systems (WG1 to WG5).

Then, the five systems (W
G1 to WG5) are output from the attenuator D at the output stage.
The signals output from the right coupling unit output terminal OUT2 via MPOUT and the signals of the five reflection-system circuits added by ADB1 are output from the left coupling unit output terminal OUT1 via the low-pass filter LPFOUT at the output stage. A low-pass filter LPFOUT is inserted into the output stage of OUT1 and OUT2
The reason why the attenuator DMPOUT is inserted in the output stage is that the low pass filters LPF1 to LPF5 are inserted in the traveling system circuit, but the low pass filter is not inserted in the reflection system circuit. Since the attenuators DMP1 to DMP5 are inserted in the traveling system circuit, no attenuator is inserted in the traveling system circuit. Therefore, by inserting elements included in the system on the opposite side into the output stage, OUT1 and OUT2 are output. This is to balance the output signal waveforms. This LPFOU
T and DMPOUT are not so significant for signals circulated many times in the waveguide, but are particularly large for components output at an early stage after a signal is input to this circuit. Function.

As described above, in this circuit, five systems of waveguides WG1 to WG5 are connected in parallel so that each signal circulates in each system and the summed signal circulates in all systems. In addition, since the characteristics of each functional element can be set independently, for example, the operation characteristics of each waveguide are set as acoustic reflection characteristics in different directions from specific points in a room. Thereby, the reverberation characteristics in the room can be simulated .

As the low-pass filter, a first-order DCF type low-pass filter as shown in FIG. 3 is used. In this low-pass filter, a coefficient multiplier COEF
When the coefficient input to is increased, the cutoff frequency increases, and when the coefficient is decreased, the cutoff frequency decreases. Note that the low-pass filter used in the circuit of FIG. 2 is not limited to this, and may have any configuration as appropriate.

As described above, the parameters for determining the characteristics of the circuit are as shown in FIG.
34 parameters consisting of 24 coefficients and 10 delay steps are required. Note that the coefficient multipliers A1i,
The symbols indicating the coefficients of A2i and the attenuator DMPi (i = 1 to 5) are represented by the same symbols as those of the respective elements. Different symbols are used for coefficients of other elements. In this device, the nine setting parameters shown in FIG. 9B are input from the reverberation characteristic setting unit 13, and the 34 parameters are determined by calculation based on the input.
DL1 to DL5 are parameters for setting the delay time of each of the waveguides WG1 to WG5, take a value of 0 to 127, and can designate a delay time in the range of 0.29 ms to 41.54 ms. DT is a parameter for setting the decay time of the impulse, that is, the time until the response output is attenuated by −20 dB when the impulse signal is input to the circuit of FIG.
A time in the range of ms to 4 s (4000 ms) can be specified. CF is a parameter that specifies the cutoff frequency of the low-pass filter, takes a value of 0 to 127, and can specify a cutoff frequency in the range of 31.1 Hz to 20.0 kHz. Further, DF is a parameter for giving a stereo feeling to the left coupling unit output OUT1 and the right coupling unit output OUT2 by making a difference between the delay time of the traveling system circuit and the reflection system circuit of each waveguide.
Values from 0 to 15 can be taken. This value and DL1
To the delay circuit DELAYF1
-5, the delay time (number of delay stages) wg of DELAYB1-5
f1 to 5 and wgb1 to 5 are determined. PH is a parameter for giving a sense of phase to the left coupling unit output OUT1 and the right coupling unit output OUT2, and takes a value of -64 to 63, and based on this value, the input level dinL,
This is done by changing dinR.

The procedure and method for calculating the coefficient and the delay time based on the setting data will be described below. First, the coupling coefficient of the waveguide filter is required to be A11 + A12 + A13 + A14 + A15 ≦ 2.0 A21 + A22 + A23 + A24 + A25 ≦ 2.0 due to the problem of network stability. Therefore, in this embodiment, a fixed value is set as follows: A11 + A12 + A13 + A14 + A15 = 2.0 A21 + A22 + A23 + A24 + A25 = 2.0 A11 = A12 = A13 = A14 = A15 = 0.4 A21 = A22 = A23 = A24 = A25 = 0.4 However, any value may be set for each coupling coefficient within a range satisfying stability. Next, delay times (number of delay stages) wgf1 to wgf
5, wgb1 to wgb5 are determined. First, the input D
The prime number table of FIG. 5 is searched with Li (i = 1 to 5) to determine the basic stage number αi (i = 1 to 5). This prime number table stores 128 prime numbers in the range of 7 to 1021, each corresponding to a parameter value of 0 to 127. Here, the reason for using a prime number for the number of steps is as follows.
If a common divisor exists in the delay time of a plurality of resonance systems (waveguides), resonance occurs at a harmonic frequency having a cycle of the common divisor delay time. This is because unnecessary peaks occur in the frequency spectrum of the output musical sound. Twice the basic stage number αi determined from the prime number table, that is, 2αi, is the delay stage number of the waveguide WGi (progressive circuit + reflective circuit). This αi is Diffusion
Increase / decrease using parameter DF, wgfi and wgb
Calculate i. This operation is as follows. However,
In order to balance the delay time between OUT1 and OUT2, the delay time of the traveling system circuit is increased in the waveguides WG1, WG3, and WG5, and the delay time of the reflection circuit is extended in the waveguides WG2 and WG4. That is, wgf1 ≧ wgb1 wgf2 ≦ wgb2 wgf3 ≧ wgb3 wgf4 ≦ wgb4 wgf5 ≧ wgb5. FIG. 6 shows these calculation methods. First, the value of DF is fetched from the reverberation characteristic setting unit 13. Alternatively, a previously input value is fetched from the memory. Next, the changing step number DD is calculated based on the basic step number αi read from the prime number table. This calculation formula is: DD = INT ｛(0.005 × αi + 10) × DF / 1
6｝. This formula is empirically determined,
When the number of delay stages is 16, it is effective to change the number of delay stages in the range of 0 to 10 stages.
It is based on an empirical rule that it is effective to set a value in the range of about 0 to 20 steps even when the number of steps increases to 4 steps × 2). Next, the number i of the waveguide is determined. If i = 1,3,5, the number of delay stages of the delay circuit of the traveling system circuit is increased,
The number of delay stages of the delay circuit of the reflection system circuit is reduced. The number of changing stages DD is compared with the number of basic stages αi. If DD is smaller, DD is added to αi to obtain wgfi, and DD is subtracted from αi to obtain wgbi. wgfi = αi + DD wgbi = αi−DD On the other hand, if DD is larger than αi, it is impossible to increase or decrease the number of delay stages using DD.
i is set to 1 as the minimum number of stages, and wgfi to be increased is 2α
Set to i-1. Thereby, wgfi + wgbi
= 2αi is maintained. wgfi = 2αi−1 wgbi = 1 On the other hand, if i = 2, 4, conversely, the number of delay stages of the delay circuit of the traveling system circuit is reduced and the number of delay stages of the delay circuit of the reflection system circuit is increased. . Number of change stages DD and number of basic stages α
i is compared with i. If DD is smaller, DD is subtracted from αi to obtain wgfi, and DD is added to αi to obtain wgbi. wgfi = αi−DD wgbi = αi + DD On the other hand, if DD is larger than αi, it is impossible to increase or decrease the number of delay stages using DD.
i is set to 1 as the minimum number of stages, and wgbi to be increased is 2α
Set to i-1. wgfi = 1 wgbi = 2αi−1 The above operation is performed for i = 1 to 5, and
gf5, wgb1 to wgb5 are calculated.

Next, the average value DL of the number of delay stages
Ask for A. This DLA is used to determine the dump multiplication factor. Since the number of delay stages mainly determines the frequency characteristics of the waveguide, the average value is obtained by the geometric mean as follows in accordance with the logarithmic sensitivity to the frequency of human hearing. ^{DLA = 5 √ (wgf1 + wgb1} ) (wgf2 + wgb2) (wgf3 + wgb3) (wgf4 +
wgb4) (wgf5 + wgb5) Next, the dump multiplication coefficients DMPI and DMPOUT
To determine. The dump multiplication coefficient is determined so as to realize the attenuation time specified by the dump parameter DT in each of the waveguide filters WG1 to WG5. Since the input signal circulates in the waveguides WG1 to WG5, the attenuator D is used for each delay time (wgfi + wgbi = 2αi).
Pass through MPi. Therefore, it is calculated how many times it passes before the designated decay time, and the number of times is -20 dB.
Determine the dump multiplication factor to attenuate. Also, DT
Takes a value in the range of 0 to 127, and this value designates the decay time in the range of 4 to 4000 ms divided into 127 steps. However, in order to correspond to the operator's intuition, the value of 4 to 4000 m is used.
Divide the range of s logarithmically. Therefore, the following calculation is performed. dt = EXP [{logn (4000/4)} × DT /
127] That is, the range of 4 to 4000 ms is logarithmically divided into 127 equal parts, multiplied by DT, and the value is returned to a real number.
Here, the time tc required to make a circuit through the waveguide (traveling circuit + reflection circuit) is tc = 2αi / FS FS: sampling frequency = 48 kHz, 2αi = wgfi + wgbi, and passes during the decay time dt. The number of times nc is nc = dt / tc, and attenuating by −20 dB with this number of times means that
DMPi ＾ nc = 10 ＾ -1 However, since “＾” is a power sign, DMPi = 10 ＾ (− 1 / nc), that is, DMPi = 10 ＾ (− 2αi / (FS · dt)). By performing this operation for i = 1 to 5, it is possible to calculate the dump multiplication factor DMPi (i = 1~5).

Further, the dump multiplication coefficient DMPOUT at the output OUT2 of the right coupling section is calculated as {20 dB when the average value DLA of the delay time previously obtained is set in the waveguide.
Is set to be dt. Therefore, DMP OUT = 10 ＾ (− DLA / (FS · dt)).

[0023]  Next, a low-pass filter LPFi (i
= 1 to 5) and the coefficient LPGi to be set in LPFOUT
(I = 1 to 5) and LPGOUT are obtained. This factor
Corresponds to the cutoff frequency. This calculator
The order is shown in FIG. First, the cutoff frequency parameter C
A basic cutoff frequency FCA is calculated based on F.
CF takes a value in the range of 0 to 127, and this value is 31.1.
A cut-off frequency in the range of Hz to 20 kHz is indicated.
However, this range is logarithmic in order to respond to the operator's intuition.
Divided. FCA = EXP [{logn (20000 / 31.1)} × CF / 127] That is, the range of 31.1 Hz to 20 kHz is logarithmically calculated.
Divide it by 127 and multiply it by CF, then return that value to a real
ing. This basic cutoff frequency FCA is
-The cutoff frequency of the LPF LPFi
According to the delay time of each waveguide WGi (i = 1 to 5)
The cutoff frequency F of each low-pass filter LPFi
Scale Ci. That is, the delay step length is
Set the cutoff frequency low for long ones,
About thingsIsSet a high cutoff frequency
By scaling according to the delay length
It is empirically confirmed that the tone improves.
Experiments show that the optimal scaling ratio is about 0.6.
Is required. Therefore, the previously calculated wave
Length of guide WG (number of delay stages) and average value DLA
Is read. The cutoff frequency F according to the number of delay stages
Scale Ci. The formula is as follows:
You. FCi = FCA ・ (DLA / 2αi) ＾ 0.6 where 2αi = wgfi + wgbi Furthermore, in order to realize this cutoff frequency, low
The coefficient given to the pass filter is obtained by the following equation. Up
As described above, the low-pass filter has a configuration as shown in FIG.
The coefficients are supplied to a coefficient multiplier COEF. What
This equation is analog low-pass filter (integration circuit
Type) and a time constant. LPGi = 1−exp (−2π · FCi / Fs) By performing this operation for i = 1 to 5,
The coefficient LPGi (i =
1 to 5) can be obtained. Note that the same operation
LPGOUT is also calculated by inputting the average FCA.
Will be issued. That is, LPGOUT = 1−exp (−2π · FCA / Fs).

[0024]  Input multiplication coefficient din of left and right joints
L and dinR depend on the PH (Phase) parameter.
To decide. Make sure that the signal levels entering the left and right
That is, if dinL = dinR, the sense of phase is OUT1,
Same for OUT2, but if dinL> dinR
The sound image is localized on OUT2 side, and if dinL <dinR
In this case, the sound image is localized on the OUT1 side. dinL + dinR =
The following calculation is performed in order to maintain 1. dinR = 0.5 (1 + PH / 64) dinL = 0.5 (1−PH / 64) In order to keep the volume power constant, dinL^Two + D
inR^Two DinL and dinR to maintain = 1.
In this case, dinR = √0.5 (1 + PH / 64) and dinL = √0.5 (1−PH / 64).

The above calculation is performed by the CPU 10 based on the setting contents of the reverberation characteristic setting section 13, and the calculation result is given to the signal processing RAM 15. As a result, the DSP 14 performs the arithmetic processing of the resonance filter shown in FIG. 2 with the characteristics desired by the operator.

Although the number of delay stages of the delay circuit DELAY is obtained from the prime number table, substantially the same effect can be obtained by obtaining it based on an exponential function (or a table). In the above embodiment, digital processing is performed using a DSP. However, this method can be applied to analog processing performed using an element such as a BBD.
Further, even a device that performs digital processing can be applied to a device having a hard logic configuration without using a DSP. In addition, when the parameter changes in response to the operation of the operation element during the processing, the coefficient and the number of delay steps can be changed in real time.

[0027]

According to the present invention , the first signal processing and the signal processing which are different from each other in the first signal path and the second signal path are performed .
And the second signal processing , the first signal path and the second signal path
The signal output from the signal path is
And the first signal processing, the balance between the two output signals is not lost, and the operator can freely set the coefficient. In the present invention, the input signal circulates through the first signal processing means-the second signal processing means and undergoes repetitive processing. Even if only the signal processing means is provided, even if a signal is input to either, both processes can be performed, so that the number of signal processing means can be reduced, and the number of parameters to be set can be reduced accordingly. Will be possible. According to the present invention, since the delay times of the plurality of delay circuits are made to have a prime relation, resonance does not occur at a common divisor frequency and harmonics do not occur.
It is possible to perform processing that is uniform in frequency.

[Brief description of the drawings]

FIG. 1 is a block diagram of a signal processing apparatus according to an embodiment of the present invention;

FIG. 2 is a circuit diagram showing functions set in a DSP of the block diagram.

FIG. 3 is a diagram showing an example of a low-pass filter in the DSP.

FIG. 4 is a diagram showing operation parameters used by the DSP and setting parameters set by an operator.

FIG. 5 is a view showing a prime number table for calculating the number of delay steps of the delay circuit of the DSP;

FIG. 6 is a diagram showing a procedure for calculating the number of delay steps of a plurality of DELAYs of the DSP.

FIG. 7 is a view showing a procedure for calculating cutoff frequencies of a plurality of low-pass filters of the DSP.

[Explanation of symbols]

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G10H 7/08 G10H 1/00 G10K 15/12

Claims (4)

(57) [Claims] 1. A first signal path having first signal processing means for performing signal processing of a first processing characteristic, and a second signal having second signal processing means for performing signal processing of a second processing characteristic. A plurality of waveguides each having a path, an input terminal for inputting a signal to the waveguide, and a signal output from a first signal path of each waveguide being synthesized and output to a first output terminal; First feedback means for supplying the combined signal to the second signal path of each waveguide; and combining the signals output from the second signal path of each waveguide and outputting to the second output terminal. And second feedback means for supplying the synthesized signal to a first signal path of each waveguide ; characteristics of a signal output from the first output terminal; and output from the second output terminal. wherein a characteristic of the signal first
Signal processing means for an output signal to be balanced in accordance with the processing characteristics and the second processing characteristics . 2. The signal processing device according to claim 1, wherein one of the first signal processing unit and the second signal processing unit is a low-pass filter, and the other is an attenuator. 3. The processing characteristic of the signal processing means for the output signal is determined by the first signal processing means of the plurality of waveguides .
Characteristics obtained by averaging processing characteristics, or second signal processing means
The signal processing device according to claim 1, wherein the signal processing device is at least one of characteristics obtained by averaging the processing characteristics. 4. The first signal path and the second signal path of the plurality of waveguides each have a delay unit, and the delay times of the respective delay units are set to a prime relationship. The signal processing device according to claim 2.