JP5883580B2 - Filter coefficient determination device - Google Patents

Description

  The present invention relates to a filter coefficient determination device, and more particularly, to determine a filter coefficient for determining a filter coefficient for performing a filter process on an audio signal output from a speaker in order to suppress the occurrence of a dip in a vehicle interior. Relates to the device.

  Conventionally, an increasing number of users enjoy music while driving by installing a car audio system in the passenger compartment. In particular, since the vehicle interior is a closed space, it is possible to construct an optimum sound field environment for the user by performing acoustic adjustment. In order to construct such an optimal sound field environment, a method of adjusting the frequency characteristics of the sound field in the vehicle interior (listening environment) according to the vehicle interior has been proposed (see, for example, Patent Document 1).

  In this method, a microphone is installed at the listening position in the vehicle interior, and after measuring the frequency characteristics between the speaker and the microphone, the frequency, amplitude and bandwidth of the filter are set so that they fall within the allowable range of the target response curve. Is automatically optimized to correct the frequency characteristics.

JP 2000-152374 A

  However, the audio signal output from the speaker in the passenger compartment has a problem that a dip due to a standing wave occurs due to reflection from a window glass or the like. In particular, when monaural signals such as vocal sounds are output from a plurality of speakers, interference between the speakers is also added, and the standing wave dip is increased.

  Although it is possible to improve to some extent by performing amplification as frequency correction as in the prior art for the occurrence of dip, noise and distortion of amplifiers and speakers increase, and in car audio systems There was a problem that power consumption increased.

  For this reason, it is required to reduce the interference between the speakers by performing a filtering process on the audio signal output from the speakers.

  The present invention has been made in view of the above problems, and determines a filter coefficient for performing a filtering process on an audio signal output from a speaker in order to reduce interference of a standing wave at a listening position. It is an object to provide a filter coefficient determination device.

  In order to solve the above-described problems, a filter coefficient determination device according to the present invention includes a front seat speaker installed on the front seat side of a vehicle interior in which seats are arranged in the front and rear, and a rear seat speaker installed on the rear seat side. A front seat measurement microphone installed at the front seat position in the vehicle interior, a rear seat measurement microphone installed at the rear seat position, and a signal output from the front seat speaker as the front seat measurement microphone and the rear seat Impulse corresponding to each combination from each speaker to each measurement microphone by measuring with the seat measurement microphone and measuring the signal output from the rear seat speaker with the front seat measurement microphone and the rear seat measurement microphone Impulse response measurement means for measuring response and measurement signal reproduction capable of outputting different M-sequence code signals orthogonal to each other And one orthogonal M-sequence code signal output from the measurement signal reproducing means using the impulse response from the front seat speaker to the front seat measurement microphone measured by the impulse response measurement means. And using the impulse response from the rear seat speaker to the front seat measurement microphone measured by the impulse response measuring means, the other orthogonal signals output from the measurement signal reproducing means are used. The convolution integration of the M-sequence code signal is performed, and the orthogonal M-sequence code signals that have been subjected to the convolution integration are synthesized, so that the influence of interference between the front seat speaker and the rear seat speaker is not included. A first signal synthesizing means for generating one signal and the front seat speaker measured by the impulse response measuring means; The impulse response up to the front seat measurement microphone is used to perform convolution integration of the one M-sequence code signal orthogonal to each other output from the measurement signal reproduction means, and the impulse response measurement means measures the measurement result. Using the impulse response from the rear seat speaker to the front seat measurement microphone, convolution integration of the one M-sequence code signal output from the measurement signal reproducing means is performed, and each M sequence subjected to the convolution integration is performed. A second signal synthesizing unit for synthesizing a sign signal to generate a second signal including an influence of interference between the front seat speaker and the rear seat speaker; and the first signal from the second signal. In the frequency characteristic of the subtracted signal, a frequency indicating the minimum value of the signal level is obtained, and a frequency indicating the minimum value is included. A front seat side standing wave frequency detecting means for obtaining a frequency range equal to or lower than a signal level as a front seat side standing wave frequency, and the front seat speaker measured by the impulse response measuring means to the rear seat measurement microphone The impulse response is used to perform convolution integration of the one orthogonal M-sequence code signal output from the measurement signal reproduction means, and the rear seat measurement is performed from the rear seat speaker measured by the impulse response measurement means. Using the impulse response to the microphone, convolution integration of the other orthogonal M-sequence code signals output from the measurement signal reproducing means is performed, and each orthogonal M-sequence code signal subjected to the convolution integration is obtained. A third signal is generated by combining and not including the influence of interference between the front seat speaker and the rear seat speaker. Using the impulse response from the front seat speaker to the rear seat measurement microphone measured by the impulse response measurement means and the orthogonal signal output from the measurement signal reproduction means. The M-sequence code signal is convolved and integrated, and the impulse response from the rear seat speaker to the rear seat measurement microphone measured by the impulse response measurement means is used to output the measurement signal reproduction means. A convolution integral of one M-sequence code signal is performed, and each of the M-sequence code signals subjected to the convolution integration is synthesized to include the influence of interference between the front seat speaker and the rear seat speaker. A fourth signal combining means for generating four signals, and a frequency characteristic of a signal obtained by subtracting the third signal from the fourth signal; Rear-seat side standing wave frequency detection that obtains the frequency at which the bell indicates the minimum value and calculates the frequency range that includes the frequency that indicates the minimum value and that is equal to or lower than the predetermined signal level as the rear-seat side standing wave frequency Means, an upper limit frequency and a lower limit frequency in the frequency range of the front seat side standing wave frequency, and a phase difference value of the front seat side standing wave frequency with respect to the rear seat side standing wave frequency. Front seat side filter coefficients, and upper seat side filter coefficients determined based on the upper limit frequency and lower limit frequency in the frequency range of the rear seat side standing wave frequency and the value of the phase difference, A filter coefficient table including a filter coefficient table calculated in advance according to acoustic characteristics in the vehicle interior based on combinations of an upper limit frequency, various lower limit frequencies, and various phase difference values. A front-seat-side filter processing unit that performs filtering on the one M-sequence code signal output from the measurement signal reproduction unit based on the front-seat filter coefficient, and the measurement signal reproduction unit. A rear seat filter processing means for performing a filtering process on the one M-sequence code signal to be output based on the rear seat filter coefficient, and an impulse response from the front seat speaker to the front seat measurement microphone. And performing convolution integration of the one M-sequence code signal filtered by the front seat filter processing means, and using the impulse response from the rear seat speaker to the front seat measurement microphone, The convolution integration of the one M-sequence code signal filtered by the seat side filter processing means is performed, and the convolution integration is performed. Fifth signal synthesizing means for synthesizing each of the one M-sequence code signals to generate a fifth signal that does not include the influence of interference between the front seat speaker and the rear seat speaker; The impulse response from the seat speaker to the rear seat measurement microphone is used to perform convolution integration of the one M-sequence code signal filtered by the rear seat side filter processing means, and from the rear seat speaker to the rear seat Using the impulse response to the seat measurement microphone, convolution integration of the one M-sequence code signal filtered by the rear seat side filter processing means is performed, and each of the one M-sequence subjected to convolution integration is performed. A sixth signal synthesizing unit that synthesizes a sign signal and generates a sixth signal that does not include an influence of interference between the front seat speaker and the rear seat speaker; Subtracting the second signal from the five signals to obtain the difference signal on the front seat side, subtracting the fourth signal from the sixth signal to obtain the difference signal on the rear seat side, The signal level increase / decrease change in the difference signal and the signal level increase / decrease change amount in the rear seat difference signal are detected in association with the phase difference value set in the filter coefficient table means. The value of the phase difference at which either the increase / decrease change amount of the signal level on the seat side or the increase / decrease change amount of the signal level on the rear seat side does not increase / decrease, and the increase / decrease change amount of the other increases greatly A phase difference detection means to be obtained, and the filter coefficient table means includes a phase difference value detected by the phase difference detection means, an upper limit frequency in a frequency range of the front seat side standing wave frequency, and A front-side filter coefficient obtained from the filter coefficient table based on a lower-limit frequency is determined as a filter coefficient for performing a filtering process on an audio signal output from the front-seat speaker; A rear-seat filter coefficient obtained from the filter coefficient table based on the phase difference value detected in step S, and the upper-limit frequency and the lower-limit frequency in the frequency range of the rear-seat side standing wave frequency, from the rear-seat speaker The filter coefficient for performing the filtering process on the output audio signal is determined.

  In the filter coefficient determination device described above, a predetermined signal level set for obtaining the front seat side standing wave frequency, and a predetermined signal level set for obtaining the rear seat side standing wave frequency. Is preferably −6 dB.

  In general, since an audio signal output from a speaker installed in a vehicle interior is reflected by a window glass or the like, a dip due to a standing wave may occur at any listening position. In particular, when monaural signals such as vocal sounds are output from a plurality of speakers, interference between the speakers is also added, and the standing wave dip is increased. In order to reduce such a dip due to a standing wave, it is required to reduce interference between speakers. For this reason, in the filter coefficient determination device according to the present invention, a filter coefficient that can reduce standing wave interference at the listening position by performing filter processing on the audio signal output from each speaker in advance. The purpose is to decide.

  In the filter coefficient determination device according to the present invention, first, an impulse response from each speaker to the front seat measurement microphone and an impulse response from each speaker to the rear seat measurement microphone are obtained. In the filter coefficient determination device according to the present invention, the first signal synthesis means performs convolution integration of one M-sequence code signal orthogonal to each other using an impulse response from the front seat speaker to the front seat measurement microphone. Using the impulse response from the rear seat speaker to the front seat measurement microphone, convolution integration of other orthogonal M-sequence code signals is performed, and each orthogonal M-sequence code signal subjected to the convolution integration is synthesized. A first signal is generated.

  Since the first signal is a combined signal obtained by convolution integration based on M-sequence code signals that are orthogonal to each other, there is no signal (monaural component signal) that is common between the speakers. For this reason, the signal does not include the influence of interference between the front seat speaker and the rear seat speaker.

  In addition, since the first signal is a composite signal that is convolved and integrated using an impulse response from each speaker to the front seat measurement microphone, the front seat speaker and the front seat speaker are installed at the installation position (front seat position) of the front seat measurement microphone. The signal does not include the influence of interference with the rear seat speaker.

  Further, in the filter coefficient determination device according to the present invention, the second signal synthesis means performs convolution integration of one M-sequence code signal orthogonal to each other using an impulse response from the front seat speaker to the front seat measurement microphone. Then, convolution integration of one M-sequence code signal is performed using an impulse response from the rear seat speaker to the front-seat measurement microphone, and a second signal is generated by synthesizing each M-sequence code signal subjected to the convolution integration. .

  Since the second signal is a combined signal obtained by convolution integration based on the same M-sequence code signal (both are one M-sequence code signal), there is a common signal (monaural component signal) between the speakers. It will be. For this reason, the signal includes the influence of interference between the front seat speaker and the rear seat speaker.

  In addition, since the second signal is a composite signal that is convolved and integrated using an impulse response from each speaker to the front seat measurement microphone, the front seat speaker and the front seat speaker are installed at the installation position of the front seat measurement microphone (front seat position). The signal includes the influence of interference with the rear seat speaker.

  Then, in the front seat side standing wave frequency detection means, by subtracting the first signal from the second signal, the signal including the influence of interference between the front seat speaker and the rear seat speaker is converted into the front seat speaker. A signal not including the influence of interference with the rear seat speaker is subtracted. The signal calculated by this subtraction processing is a signal that shows only the influence of the interference between the front seat speaker and the rear seat speaker, and the influence of the interference between the speakers at the installation position of the front seat measurement microphone is determined from the frequency characteristics. It can be obtained.

  The front-seat-side standing wave frequency detection means determines the minimum signal level based on the frequency characteristics of the signal that shows only the influence of interference between the front-seat speaker and the rear-seat speaker at the installation position of the front-seat measurement microphone. By obtaining the indicated frequency, the frequency at which the dip is generated can be easily obtained.

  Furthermore, the frequency range where the dip occurs can be specified by obtaining the frequency range including the frequency indicating the minimum value and being equal to or lower than the predetermined signal level as the front seat side standing wave frequency. The predetermined signal level is set to a value (a level at which a human can clearly determine that the sound changes) that the auditory effect is clearly felt by the phase control. In particular, it is preferable to set to −6 dB, which is generally halved in amplitude.

  Similarly, at the position where the rear seat measurement microphone is installed, the rear seat side standing wave frequency detection means subtracts the third signal from the fourth signal, so that the difference between the front seat speaker and the rear seat speaker is obtained. A signal not including the influence of the interference between the front seat speaker and the rear seat speaker is subtracted from the signal including the influence of the interference. The signal calculated by this subtraction processing is a signal that shows only the influence of the interference between the front seat speaker and the rear seat speaker, and the influence of the interference between the speakers at the installation position of the rear seat measurement microphone is determined from the frequency characteristics. It can be obtained.

  Even at the position where the rear-seat measurement microphone is installed, the rear-seat standing-wave frequency detection means has a minimum signal level from the frequency characteristic of the signal that shows only the influence of interference between the front-seat speaker and the rear-seat speaker. By obtaining the frequency indicating the value, it is possible to easily obtain the frequency at which the dip occurs.

  Furthermore, the frequency range where the dip occurs can be specified by obtaining the frequency range including the frequency indicating the minimum value and being equal to or lower than the predetermined signal level as the rear seat side standing wave frequency.

  Thus, the frequency and frequency range of the dip at the installation position of the front seat measurement microphone (front seat standing wave frequency) and the frequency and frequency range of the dip at the installation position of the rear seat measurement microphone (rear seat) (Side standing wave frequency) is obtained, it is possible to obtain a frequency band and a signal level necessary for reducing the dip by the filter processing.

  Subsequently, in the filter coefficient determination device according to the present invention, the fifth signal synthesizing unit includes the upper seat frequency of the front seat side standing wave frequency and the lower limit frequency and the front seat side standing wave frequency with respect to the rear seat side standing wave frequency. One M-sequence code signal filtered using the front seat side filter coefficient determined by the filter coefficient table means based on the phase difference value is used as the impulse response from the front seat speaker to the front seat measurement microphone. The convolution integration is performed, and one M-sequence code signal filtered using the front seat filter coefficient is subjected to convolution integration using the impulse response from the rear seat speaker to the front seat measurement microphone. Each one of the performed M-sequence code signals is synthesized to generate a fifth signal that does not include the influence of interference between the front seat speaker and the rear seat speaker.

  This fifth signal is a combined signal that is convolution integrated based on the same M-sequence code signal (both are one M-sequence code signal), and is a signal that is convolution-integrated by the same M-sequence code signal. There is a common signal (monaural component signal) between the speakers. However, since one M-sequence code signal in the fifth signal is a signal filtered using the front seat side filter coefficient, the influence of interference between the front seat speaker and the rear seat speaker is affected by the filter processing. It is possible to determine whether the degree is reduced.

  Further, since the fifth signal is a composite signal that is convolved and integrated using an impulse response from each speaker to the front seat measurement microphone, the front seat speaker at the installation position (front seat position) of the front seat measurement microphone is used. This is a signal that can determine the influence of interference between the speaker and the rear seat speaker by filter processing using the front seat side filter coefficient.

  On the other hand, as described above, since the second signal is a composite signal that is convolution integrated based on the same M-sequence code signal (both are one M-sequence code signal), the signal common to the speakers (monaural) Component signal). For this reason, the signal includes the influence of interference between the front seat speaker and the rear seat speaker.

  Further, since the second signal is a composite signal that is convolved and integrated using an impulse response from each speaker to the front seat measurement microphone, the front seat speaker and the rear are located at the installation position of the front seat measurement microphone (front seat position). The signal includes the influence of interference with the seat speaker.

  Therefore, the phase difference detection means subtracts the second signal from the fifth signal to obtain the front-seat differential signal, so that the difference in signal level of the differential signal is obtained when the dip is not reduced by the filter processing. When the dip is reduced by the filter processing, a difference occurs in the signal level of the differential signal.

  On the other hand, the influence of dip differs for each user listening position. For this reason, similarly, the phase difference detection means detects the front seat speaker and the rear seat from the sixth signal filtered using the rear seat filter coefficient at the installation position (rear seat position) of the rear seat measurement microphone. A difference signal on the rear seat side is obtained by subtracting the fourth signal including the influence of interference with the speaker. If there is no difference in the signal level of the difference signal, it can be determined that the dip is not reduced by the filtering process. If there is a difference in the signal level of the difference signal, the dip is reduced by the filtering process. It is possible to judge that

  In addition, since the influence of the dip becomes remarkable due to the interference between the speakers, the phase difference detecting means reduces the signal level increase / decrease change amount of the difference signal on the front seat side or the rear seat side in order to reduce the interference between the speakers. The phase difference value at which either the increase / decrease change amount of the signal level of the difference signal does not increase / decrease (that is, indicates a state where a stable value is maintained) and the other increase / decrease change amount greatly increases. Ask.

  In this way, the phase difference value with one small increase / decrease change amount and the other increase / decrease change amount is obtained, and the filter coefficient table means determines the phase difference value and the upper limit frequency and lower limit of the front-side standing wave frequency. The front seat filter coefficient obtained from the filter coefficient table based on the frequency is determined as a filter coefficient for filtering the audio signal output from the front seat speaker, and the phase difference value and the rear seat side are determined. The rear seat filter coefficient obtained from the filter coefficient table based on the upper limit frequency and the lower limit frequency of the standing wave frequency is determined as a filter coefficient for filtering the audio signal output from the rear seat speaker. By performing filter processing on the audio signal output from each speaker using the filter coefficient determined in this way, it is possible to effectively reduce the interference between the speakers and suppress the occurrence of dip. Is possible.

  According to the filter coefficient determination device of the present invention, the interference between speakers is reduced by controlling the phase of the audio signal output from the speaker based on either the front seat position or the rear seat position in the vehicle interior. And the occurrence of dip can be suppressed.

It is the block diagram which showed schematic structure of the standing wave measuring apparatus which concerns on embodiment. It is the schematic which showed the installation position of the four speakers installed in a vehicle interior, and the installation position of four measurement microphones. It is an example of the impulse response recorded in the data recording part which concerns on embodiment, Comprising: Impulse response C1 between the speakers FL in the measurement microphone FLM, Impulse response C5 between the speakers FL in the measurement microphone FRM, An impulse response C9 between the measurement microphone RLM and the speaker FL and an impulse response C13 between the measurement microphone RRM and the speaker FL are shown. It is the block diagram which showed schematic structure of the standing wave control apparatus which concerns on embodiment. The schematic structure of the standing wave detection part which concerns on embodiment is shown. It is the block diagram which showed schematic structure of the standing wave detection unit corresponding to the measurement microphone FLM installed in the left front seat (passenger seat), a measurement signal reproduction | regeneration part, and a standing wave determination part. In the standing wave control device according to the embodiment, the same M-sequence code and orthogonal M-sequence code are output as measurement signals from the measurement signal reproduction unit, and convolution integration is performed in each standing wave detection unit corresponding to each measurement microphone. It is the figure which showed the frequency characteristic of the signal of the M series code orthogonal to the signal of the same M series code | symbol equalized for every measurement microphone. It is the figure which showed the difference value of the frequency characteristic at the time of subtracting the signal of the orthogonal M series code from the signal of the same M series code shown in FIG. (A) shows the frequency characteristic after performing the averaging process on the front seat side, and (b) shows the frequency characteristic after performing the averaging process on the rear seat side. It is the block diagram which showed schematic structure of the filter coefficient calculation part which concerns on embodiment. A standing wave detection unit corresponding to the measurement microphone FLM installed in the left front seat (passenger seat), a measurement signal reproduction unit, a level determination unit, a filter coefficient table unit, a first sub correction filter unit to a fourth unit. It is the block diagram which showed schematic structure with a sub correction filter part. It is the figure which showed the filter coefficient table which concerns on embodiment. It is the block diagram which showed the internal structure of the 1st sub correction | amendment filter part which concerns on embodiment-the 4th sub correction filter part. It is the graph which showed the value of the increase / decrease amount (increase / decrease amount) of the signal level of the difference signal on the front seat side, and the increase / decrease amount (increase / decrease amount) of the signal level of the difference signal of the rear seat side. It is the figure which showed the phase characteristic of the all-pass filter when a phase difference is set to -90 degree | times, Comprising: It is the figure which showed each phase characteristic (phase value) of the front seat side and the rear seat side. It is the figure which showed the phase characteristic of the all-pass filter when a phase difference is set to -90 degree | times, Comprising: It is the figure which showed the value of the phase at the time of subtracting the phase characteristic of the front seat side from the phase characteristic of the rear seat side. . The frequency characteristics after the signal subjected to the filter processing in the first sub correction filter unit to the fourth sub correction filter unit is averaged by the logarithmic averaging processing unit, the lower limit frequency is 80 Hz, the upper limit frequency is 140 Hz, The frequency characteristics when the phase control is actually performed by the first sub correction filter unit to the fourth sub correction filter unit using the filter coefficient when the phase difference is set to −180 degrees and when the phase control is not performed are shown. FIG. 5 is a diagram illustrating each standing wave detection unit (measurement microphones FLM, FRM, RLM, RRM) corresponding to the installation positions of the measurement microphones. With respect to the diagram shown in FIG. 17, in the filter coefficient table according to the embodiment, the frequency characteristics of the signal when the filter coefficient is used when the lower limit frequency is set to 80 Hz, the upper limit frequency is set to 140 Hz, and the phase difference is set to −180 degrees. Is a diagram showing each standing wave detection unit (measurement microphones FLM, FRM, RLM, RRM) corresponding to the installation position of each measurement microphone.

  Hereinafter, a filter coefficient determination device according to the present invention will be described in detail with reference to the drawings, showing a standing wave control device as an example. First, a standing wave measuring device for measuring the impulse response (acoustic characteristics) in the vehicle interior will be described, and then a filter for suppressing the occurrence of dip in the vehicle interior using the measured impulse response. A standing wave control device for calculating a coefficient will be described.

[Standing wave measuring device]
FIG. 1 is a block diagram showing a schematic configuration of a standing wave measuring apparatus. The standing wave measuring apparatus 10 includes a measurement signal reproduction unit (measurement signal reproduction unit) 11, a speaker selection unit 12, an amplifier unit 13, and four (four-channel) speakers (speaker FL, speaker FR, speaker RL, (Speaker RR) 14, measurement microphones for four seats (measurement microphone FLM, measurement microphone FRM, measurement microphone RLM, measurement microphone RRM) 15, measurement unit (impulse response measurement means) 16, and data recording unit 17 doing.

  The measurement signal reproducing unit 11 plays a role of reproducing an M-sequence code (M-sequence code signal) as a measurement signal. The measurement signal reproducing unit 11 can output four types of M-sequence codes (M1 to M4) orthogonal to each other as measurement signals. For example, the same M-sequence code is an output signal corresponding to a case where monaural sound such as vocals is reproduced from a plurality of speakers, and the orthogonal M-sequence code is an output signal corresponding to a stereo signal, for example. Thus, the signal does not cause interference between speakers.

  The speaker selection unit 12 has a role of selecting from which speaker 14 (speaker FL, speaker FR, speaker RL, speaker RR) the measurement signal input from the measurement signal reproducing unit 11 is output.

  The amplifier unit 13 amplifies the measurement signal output to the speaker selected by the speaker selection unit 12. The speaker 14 selected by the speaker selection unit 12 outputs the measurement sound amplified by the amplifier unit 13.

  FIG. 2 shows an installation position of four speakers 14 (speaker FL, speaker FR, speaker RL, speaker RR) installed in the vehicle interior, and a measurement microphone 15 (measurement microphone FLM, measurement microphone FRM, measurement microphone RLM, measurement microphone). It is the schematic which showed the installation position of RRM).

  As shown in FIG. 2, the speaker FL is installed, for example, in a door or A pillar portion of the left front seat (passenger seat side) of the passenger compartment, and the speaker FR is, for example, a door of the right front seat (driver's seat side). The speaker RL is installed at a left rear seat, for example, a door portion, and the speaker RR is installed at a right rear seat, for example, a door portion. Note that the speaker FL and the speaker FR correspond to the front seat speaker according to the present invention, and the speaker RL and the speaker RR correspond to the rear seat speaker according to the present invention.

  The measurement microphone 15 (measurement microphone FLM, measurement microphone FRM, measurement microphone RLM, measurement microphone RRM) has a role of measuring the measurement sound output from the speaker 14. As shown in FIG. 2, in the standing wave measuring apparatus 10 according to the present embodiment, the measurement microphone FLM is installed at the left front seat position (passenger seat position), and the measurement microphone FRM is at the right front seat position (driver seat position). ), The measurement microphone RLM is installed at the left rear seat position, and the measurement microphone RRM is installed at the right rear seat position. The measurement microphone FLM and the measurement microphone FRM correspond to the front seat measurement microphone according to the present invention, and the measurement microphone RLM and the measurement microphone RRM correspond to the rear seat measurement microphone according to the present invention.

  Since the measurement microphone 15 is installed at each seat position, the measurement sound is measured by each measurement microphone 15, so that the installation position of each measurement microphone 15 (the measurement microphone FLM is the passenger seat, the measurement microphone FRM is the driver seat, the measurement It is possible to measure the measurement sound having the acoustic characteristics of each seat when the user is seated on the left rear seat of the microphone RLM and the position of the right rear seat of the measurement microphone RRM.

  The measurement unit 16 acquires the measurement signal output from the speaker 14 through each measurement microphone 15 and performs correlation calculation processing using the measurement signal output from the measurement signal reproduction unit 11 as a reference signal. Then, the measurement unit 16 calculates an impulse response between each measurement microphone 15 and each speaker 14 by this correlation calculation process, and records it in the data recording unit 17 as impulse response data.

  The data recording unit 17 has a role of recording an impulse response between each measurement microphone 15 and each speaker 14 calculated by the measurement unit 16. The data recorded in the data recording unit 17 is composed of 16 combinations of impulse responses according to the number of installed speakers 14 (4) × the number of installed measurement microphones 15 (4).

  Specifically, the standing wave measuring apparatus 10 according to the present embodiment includes an impulse response C1 between the speaker FL and the measurement microphone FLM, an impulse response C2 between the speaker FR and the measurement microphone FLM, and a speaker. An impulse response C3 between the RL and the measurement microphone FLM and an impulse response C4 between the speaker RR and the measurement microphone FLM are recorded.

  The standing wave measuring apparatus 10 includes an impulse response C5 between the speaker FL and the measurement microphone FRM, an impulse response C6 between the speaker FR and the measurement microphone FRM, and between the speaker RL and the measurement microphone FRM. And the impulse response C8 between the speaker RR and the measurement microphone FRM are recorded.

  Further, the standing wave measuring apparatus 10 includes an impulse response C9 between the speaker FL and the measurement microphone RLM, an impulse response C10 between the speaker FR and the measurement microphone RLM, and between the speaker RL and the measurement microphone RLM. And the impulse response C12 between the speaker RR and the measurement microphone RLM are recorded.

  The standing wave measuring apparatus 10 includes an impulse response C13 between the speaker FL and the measurement microphone RRM, an impulse response C14 between the speaker FR and the measurement microphone RRM, and between the speaker RL and the measurement microphone RRM. And the impulse response C16 between the speaker RR and the measurement microphone RRM are recorded.

  FIG. 3 shows an example of an impulse response (impulse response data) recorded in the data recording unit 17, and is an impulse response C1 between the speaker FL and the measurement microphone FLM, and between the speaker FL and the measurement microphone FRM. The impulse response C5, the impulse response C9 between the speaker FL and the measurement microphone RLM, and the impulse response C13 between the speaker FL and the measurement microphone RRM are shown.

[Standing wave control device]
Next, the standing wave control device will be described. The standing wave control device has a role of reducing dip that may occur for each seat by using the impulse responses C1 to C16 measured by the standing wave measuring device 10.

  FIG. 4 is a block diagram showing a schematic configuration of the standing wave control device. The standing wave control device 20 includes a standing wave detection unit 21, a filter coefficient calculation unit 22, a main correction filter unit 23, an amplifier unit 13, and a speaker 14. In the standing wave control device 20, since the amplifier unit 13 and the speaker 14 have already been described in the standing wave measuring device 10, detailed description thereof is omitted here.

  Further, the data recording unit 17 and the measurement signal reproduction unit 11 are connected to the standing wave detection unit 21. The data recording unit 17 records the impulse responses C1 to C16 measured by the standing wave measuring device 10 as described above. The standing wave detection unit 21 receives impulse responses C1 to C1 from the data recording unit 17. C16 can be read and used.

  In FIG. 4, the measurement signal reproduction unit 11, the standing wave detection unit 21, the filter coefficient calculation unit 22, and the data recording unit 17 are a first sub correction filter unit 61 a to a fourth sub correction filter provided in the main correction filter unit 23. It has a role of determining the filter coefficient of the part 61d. In this respect, the measurement signal reproduction unit 11, the standing wave detection unit 21, and the filter coefficient calculation unit 22 correspond to the filter coefficient determination device according to the present invention. The main correction filter unit 23 performs a filter process using the filter coefficient determined by the filter coefficient calculation unit 22 on the audio signal output from the sound source (not shown), and then the signal level is changed by the amplifier unit 13. Amplification is performed and output from the speaker 14.

  FIG. 5 shows a schematic configuration of the standing wave detection unit 21. The standing wave detection unit 21 includes four standing wave detection units 31 a to 31 d and one standing wave determination unit 32. The four standing wave detection units 31a to 31d are provided so as to correspond to the four measurement microphones 15 (measurement microphone FLM, measurement microphone FRM, measurement microphone RLM, and measurement microphone RRM), respectively. In FIG. 5, the standing wave detection unit 31a corresponding to the measurement microphone FLM installed in the left front seat (passenger seat) and the standing wave corresponding to the measurement microphone FRM installed in the right front seat (driver's seat). There are four constants: a detection unit 31b, a standing wave detection unit 31c corresponding to the measurement microphone RLM installed in the left rear seat, and a standing wave detection unit 31d corresponding to the measurement microphone RRM installed in the right rear seat. A standing wave detection unit 31 is shown.

  Four types of measurement signals output from the measurement signal reproducing unit 11 are input to the standing wave detection units 31a to 31d. Specifically, M-sequence codes M1 to M4 that are orthogonal to each other are input as measurement signals. Further, the signals (difference signals) output from the respective standing wave detection units 31 a to 31 d are configured to be input to the standing wave determination unit 32.

  FIG. 6 shows a schematic configuration of the standing wave detection unit 31a corresponding to the measurement microphone FLM installed in the left front seat (passenger seat), the measurement signal reproduction unit 11, and the standing wave determination unit 32. The block diagram which abbreviate | omitted description of the standing wave detection units 31b-31d is shown. In FIG. 6, only the block diagram of the standing wave detection unit 31a is shown as a representative example, but the standing wave detection unit 31b corresponding to the measurement microphone FRM installed in the right front seat (driver's seat) and the left side are shown. The standing wave detection unit 31c corresponding to the measurement microphone RLM installed in the rear seat and the standing wave detection unit 31d corresponding to the measurement microphone RRM installed in the right rear seat are configured by similar functional blocks. Hereinafter, for convenience of description, the standing wave detection unit 31a shown in FIG. 5 will be described.

  As shown in FIG. 6, the standing wave detection unit 31a includes two sets of four impulse response units corresponding to the impulse response between the measurement microphone FLM and each speaker 14 (speakers FL, FR, RL, RR) ( A total of 8 impulse response units consisting of a set of impulse response units 41 to 44 and impulse response units 45 to 48), two addition units 51a and 51b, two frequency conversion units 52a and 52b, and two logarithms Average processing sections 53a and 53b and a difference calculation section 54 are provided.

  The impulse responses between the measurement microphone FLM and the four speakers 14 have already been detected by the standing wave measuring apparatus 10, and two sets of impulse response units (a total of eight response units installed in the standing wave detection unit 31a). The impulse response unit) has a role of convolving and integrating the impulse response between the corresponding measurement microphone 15 and the four speakers 14 into the measurement signal (M-sequence code).

  Specifically, the standing wave detection unit 31a includes impulse response units 41 and 45 that perform convolution integration using an impulse response C1 between the speaker FL and the measurement microphone FLM, and between the speaker FR and the measurement microphone FLM. Impulse response units 42 and 46 that perform convolution integration using the impulse response C2 of the above, impulse response units 43 and 47 that perform convolution integration using the impulse response C3 between the speaker RL and the measurement microphone FLM, and a speaker RR Impulse response units 44 and 48 that perform convolution integration using an impulse response C4 with the measurement microphone FLM are provided.

  In the standing wave detection units 31a to 31d, one set (first set) of each of two sets of impulse response units (two sets of impulse response units 41 to 44 and impulse response units 45 to 48) installed. M-sequence codes M1 to M4 are respectively input to the impulse response sections (impulse response sections 41 to 44) of M, and the M-sequence codes are input to the impulse response sections (impulse response sections 45 to 48) of the other set (second set). A code M4 is input.

  Further, impulse response data C1 to C4 corresponding to the impulse response units 41 to 48 read from the data recording unit 17 are set as the impulse response values in the impulse response units 41 to 48, respectively.

  The impulse response units 41 to 48 perform convolution integration using the respective impulse responses on the measurement signals of the respective M-sequence codes M1 to M4, and the addition units 51a and 51b synthesize and combine the convolution integration measurement signals. The measured signals (first signal and second signal) are input to the frequency converters 52a and 52b.

  Here, since the M sequence codes M1 to M4 are orthogonal to each other, when each of the M sequence codes M1 to M4 is input to the impulse response units 41 to 44, a signal corresponding to a so-called stereo signal is input and convolved. Applicable when integration is performed. On the other hand, when the M sequence code M4 which is the same measurement signal is input to each of the impulse response units 45 to 48, since the respective signals are common, a signal corresponding to a so-called monaural signal is input. Applicable when convolution integration is performed.

  The impulse response units 41 to 44 and the addition unit 51a in the standing wave detection unit 31a and the standing wave detection unit 31b correspond to the first signal synthesis unit according to the present invention, and the standing wave detection unit 31a and the standing wave detection unit 31a. The impulse response units 45 to 48 and the addition unit 51b in the wave detection unit 31b correspond to the second signal synthesis unit according to the present invention.

  The frequency converters 52a and 52b have a role of converting the measurement signals synthesized in the adders 51a and 51b into the frequency domain by Fourier transform. The logarithmic averaging processing units 53a and 53b perform logarithmic smoothing processing in the frequency domain by increasing the averaging number of the measurement signals converted by the frequency converting units 52a and 52b as the frequency increases. , And has a role of converting a linear signal into a decibel signal.

  The difference calculation unit 54 performs a process of subtracting the orthogonal M-sequence code side from the same M-sequence code side from the same M-sequence code side in the decibel frequency domain for each signal converted into the decibel signal by the logarithmic averaging processing units 53a and 53b. Have a role to do.

  Note that the processing of the impulse response unit, the addition unit, the frequency conversion unit, the logarithmic averaging processing unit, and the difference calculation unit is not limited to the standing wave detection unit 31a shown in FIG. 5, but the measurement microphone FRM installed in the driver's seat. The standing wave detection unit 31b, the standing wave detection unit 31c of the measurement microphone RLM installed in the left rear seat, and the standing wave detection unit 31d of the measurement microphone RRM installed in the right rear seat are similarly performed. For this reason, in the standing wave detection units 31c and 31d of the measurement microphones RLM and RRM installed on the rear seat side, the impulse response units 41 to 48 receive the respective impulses for the measurement signals of the M series codes M1 to M4. Convolution integration is performed using the response, and convolution integration measurement signals are synthesized in the respective addition units 51a and 51b. The synthesized measurement signal corresponds to the third signal and the fourth signal of the present invention.

  In addition, the impulse response units 41 to 45 and the addition unit 51a of the standing wave detection unit 31c and the standing wave detection unit 31d correspond to the third signal synthesis unit according to the present invention, and the standing wave detection unit 31c and the standing wave detection unit 31c. The impulse response units 44 to 48 and the addition unit 51b of the wave detection unit 31d correspond to a fourth signal synthesis unit according to the present invention.

  In FIG. 7, an impulse response is actually obtained for each installation position of the measurement microphone 15 in the passenger compartment, and the standing wave control device 20 uses the same M-sequence code and orthogonal M-sequence code as a measurement signal as a measurement signal reproduction unit. 11 shows the frequency characteristics of the M-sequence code signal orthogonal to the same M-sequence code signal that is output from 11 and subjected to convolution integration or the like in the standing wave detection units 31a to 31d. FIG. 8 shows the difference value of the frequency characteristics shown in FIG.

  As already described, the same M-sequence code corresponds to a case where a monaural signal such as vocal is reproduced from a plurality of speakers, and the frequency characteristics of the same M-sequence code in FIG. In addition to the reflection and the like, a state where interference between the speakers occurs is shown. On the other hand, the frequency characteristic of the orthogonal M-sequence code in FIG. 7 shows a state in which no interference occurs between the speakers.

  FIG. 8 shows frequency characteristics when an orthogonal M-sequence code signal is subtracted from the same M-sequence code signal, and shows a signal in which interference between speakers is manifested. In FIG. 8, the portion where the signal level is 0 dB or less indicates the frequency at which the dip occurs due to the interference between the speakers, and the magnitude in the minus direction indicates the depth of the dip.

  A signal obtained by subtracting an orthogonal M-sequence code signal from the same M-sequence code signal by the difference calculation unit 54 (a signal corresponding to the frequency characteristic shown in FIG. 8) is input to the standing wave determination unit 32. The standing wave determination unit 32 includes not only the signal output from the difference calculation unit 54 of the standing wave detection unit 31a but also the signal output from the difference calculation unit 54 of the standing wave detection units 31b to 31d. Entered.

  The standing wave determination unit 32 has a role of detecting a frequency band in which a dip occurs in the signals input from the standing wave detection units 31a to 31d. In general, the interior of the vehicle has a symmetrical space configuration. Therefore, the frequency characteristic measured at the driver's seat and the frequency characteristic measured at the passenger seat show the same characteristic tendency, and the frequency characteristic measured at the left rear seat and the frequency characteristic measured at the right rear seat are the same. The characteristic tendency of is shown.

  Therefore, in the standing wave determination unit 32 according to the present embodiment, the frequency characteristics of the signal input from the standing wave detection unit 31a of the measurement microphone FLM installed in the passenger seat and the measurement installed in the driver seat. Detection of the frequency characteristic in which a dip occurs by setting a threshold (threshold) after combining both the frequency characteristic of the signal input from the standing wave detection unit 31b of the microphone FRM and performing an averaging process. I do.

  Similarly, in the standing wave determination unit 32, the frequency characteristics of the signal input from the standing wave detection unit 31c of the measurement microphone RLM installed in the left rear seat, and the measurement microphone installed in the right rear seat. After combining both the frequency characteristics of the signal input from the standing wave detection unit 31d of the RRM and performing an averaging process, the frequency characteristics in which the dip occurs are detected by setting a threshold.

  Specifically, the standing wave determination unit 32 detects a frequency band that is below the threshold, and detects a frequency at which the signal level is minimum. By this detection, the frequency band including the minimum value is determined as the frequency band in which the deep dip is generated in the frequency band that is equal to or lower than the threshold.

  FIG. 9A shows the frequency characteristics after performing the averaging process on the front seat side, and FIG. 9B shows the frequency characteristics after performing the averaging process on the rear seat side. 9A and 9B show a case where the threshold is set to −6 dB in the standing wave determination unit 32. FIG. The minimum frequency on the front seat side is 380 Hz as shown in FIG. 9A, and the signal level is −8.5 dB. On the other hand, the frequency that is the minimum value on the rear seat side, as shown in FIG. 9B, is 90 Hz, and the signal level thereof is −12.0 dB. By these thresholds, the frequency band where the deep dip is generated includes the signal level at which the signal level becomes the minimum value, and is determined to be the frequency band of the signal level that is equal to or lower than the threshold. The front seat side is determined as 250 Hz to 400 Hz, and the rear seat side is determined as 80 Hz to 140 Hz. The filter coefficient calculation unit 22 uses the frequency thus determined as the standing wave frequency on the front seat side (front seat side standing wave frequency) and the standing wave frequency on the rear seat side (rear seat side standing wave frequency). Output to.

  Note that the threshold value set by the standing wave determination unit 32 is set to a value (a level at which a human can clearly determine that the sound changes) by a phase control described later. In this embodiment, a value of −6 dB, which is generally halved in amplitude, is set as the threshold. Further, the difference calculation unit 54 and the standing wave determination unit 32 in the standing wave detection unit 31a and the standing wave detection unit 31b correspond to the front seat side standing wave frequency detection unit according to the present invention, and the standing wave detection. The difference calculation unit 54 and the standing wave determination unit 32 in the unit 31c and the standing wave detection unit 31d correspond to the rear seat side standing wave frequency detection unit according to the present invention.

  Next, the filter coefficient calculation unit 22 will be described. The filter coefficient calculation unit 22 is based on the standing wave frequencies on the front seat side and the rear seat side acquired from the standing wave determination unit 32 of the standing wave detection unit 21 and the filter coefficient table. Has a role to select filter coefficients.

  FIG. 10 is a block diagram showing a schematic configuration of the filter coefficient calculation unit 22. The filter coefficient calculation unit 22 includes a first sub correction filter unit 61a, a second sub correction filter unit 61b, a third sub correction filter unit 61c, a fourth sub correction filter unit 61d, and standing wave detection units 31a to 31a. 31d, a level determination unit 62, and a filter coefficient table unit 63. In addition, the measurement signal regeneration unit 11 is also connected to the filter coefficient calculation unit 22, and the measurement signal regeneration unit 11 includes the standing wave detection units 31a to 31d, the first sub correction filter unit 61a, and the second sub correction filter unit. The same M-sequence code M4 is output to 61b, the third sub correction filter unit 61c and the fourth sub correction filter unit 61d.

  The standing wave detection units 31a to 34d are the same as the standing wave detection units described in FIGS. FIG. 11 shows a standing wave detection unit 31a corresponding to the measurement microphone FLM installed in the left front seat (passenger seat), the measurement signal reproduction unit 11, the level determination unit 62, the filter coefficient table unit 63, the first A schematic configuration of the first sub correction filter unit 61a to the fourth sub correction filter unit 61d is shown, and a block diagram in which the standing wave detection units 31b to 31d are omitted is shown. In FIG. 11, only the block diagram of the standing wave detection unit 31a is shown as a representative example, but the standing wave detection unit 31b corresponding to the measurement microphone FRM installed in the right front seat (driver's seat), and the left side The standing wave detection unit 31c corresponding to the measurement microphone RLM installed in the rear seat and the standing wave detection unit 31d corresponding to the measurement microphone RRM installed in the right rear seat are configured by similar functional blocks. Become. Hereinafter, for convenience of description, the standing wave detection unit 31a shown in FIG. 11 will be described.

  The measurement signal (M-sequence code M4) output from the measurement signal reproduction unit 11 is a first sub correction filter unit 61a, a second sub correction filter unit 61b, a third sub correction filter unit 61c, and a fourth sub correction filter unit 61d. Is input. The M-sequence code M4 input to the first sub correction filter unit 61a, the second sub correction filter unit 61b, the third sub correction filter unit 61c, and the fourth sub correction filter unit 61d is the sub correction filter units 61a to 61d. After the filtering process is performed at, the signals are input to the impulse response units 41 to 44, respectively.

  The first sub correction filter unit 61a and the second sub correction filter unit 61b correspond to front seat filter processing means according to the present invention, and the third sub correction filter unit 61c and the fourth sub correction filter unit 61d are This corresponds to the rear seat filter processing means according to the present invention.

  In the impulse response units 41 to 44, the input signals are subjected to convolution integration by the impulse responses C1 to C4 in the respective impulse response units 41 to 44, and then synthesized in the addition unit 51a. Thereafter, the measurement signal synthesized in the adding unit 51a is converted into the frequency domain by Fourier transform in the frequency converting unit 52a, and logarithmic smoothing processing in the frequency domain is performed in the logarithmic averaging processing unit 53a. The signal is converted to a decibel signal.

  The impulse response units 41 to 44 and the addition unit 51a in the standing wave detection unit 31a and the standing wave detection unit 31b correspond to the fifth signal synthesis unit according to the present invention. Further, the impulse response units 41 to 44 and the addition unit 51a in the standing wave detection unit 31c and the standing wave detection unit 31d correspond to the sixth signal synthesis unit according to the present invention. Further, the measurement signals synthesized in the respective addition units 51a correspond to the fifth signal and the sixth signal according to the present invention.

  On the other hand, the measurement signal (M-sequence code) directly input from the measurement signal reproduction unit 11 to the impulse response units 45 to 48 without passing through the sub correction filter units 61a to 61d is transmitted to the respective impulse response units 45 to 45. 48, convolution integration is performed using the impulse responses C1 to C4, and then synthesis is performed in the adder 51b. Thereafter, the measurement signal synthesized in the adder 51b is converted into the frequency domain by Fourier transform in the frequency converter 52b, and logarithmic smoothing processing in the frequency domain is performed in the logarithmic averaging processor 53b. The signal is converted to a decibel signal. Since the measurement signals input to any of the impulse response units 41 to 48 are the same M-sequence code, processing for a monaural signal is performed.

  Thereafter, signals converted into decibel signals by the logarithmic averaging processing units 53a and 53b are signals that have not been subjected to correction processing from the signals that have been subjected to filter processing by the sub correction filter units 61a to 61d in the difference calculation unit 54. The process which subtracts is performed. By this subtraction process, it is possible to obtain the amount of increase or decrease in the signal level when the sub correction filter units 61a to 61d perform phase control by the filter process.

  The above-described impulse response unit, addition unit, frequency conversion unit, logarithmic averaging processing unit, and difference calculation unit of the standing wave detection unit 31 are not limited to the standing wave detection unit 31a illustrated in FIG. Standing wave detection unit 31b of measurement microphone FRM installed in the driver's seat, standing wave detection unit 31c of measurement microphone RLM installed in the left rear seat, standing wave detection of measurement microphone RRM installed in the right rear seat The unit 31d is similarly performed.

  As described above, the first sub correction filter unit 61a to the fourth sub correction filter unit 61d have a role of performing filter processing on the M-sequence code that is the measurement signal. The first sub-correction filter unit 61a to the fourth sub-correction filter unit 61d are configured by an all-pass filter using a second-order IIR filter, and have a role of controlling the phase while keeping the frequency amplitude characteristic constant. ing. Specifically, the first sub correction filter unit 61a has a role of performing phase control for the left front speaker FL, and the second sub correction filter unit 61b performs phase control for the right front speaker FR. Have a role. Further, the third sub correction filter unit 61c has a role of performing phase control for the left rear speaker RL, and the fourth sub correction filter unit 61d performs phase control for the right rear speaker RR. Have a role.

  The filter coefficient of each of the sub correction filter units 61a to 61d is set by the filter coefficient table unit 63. The internal configuration of the first sub correction filter unit 61a to the fourth sub correction filter unit 61d will be described later.

  The filter coefficient table unit 63 is obtained by the band information of the standing wave frequency on the front seat side acquired from the standing wave detection unit 21, the band information of the standing wave frequency on the rear seat side, and the level determination unit 62 described later. The filter coefficients of the first sub correction filter unit 61a to the fourth sub correction filter unit 61d are determined based on the phase difference information. The filter coefficient table unit 63 includes a filter coefficient table. Based on the upper limit frequency and the lower limit frequency in the band information of the standing wave frequency on the front seat side, the filter coefficient table unit 63 includes the first sub correction filter unit 61a. A filter coefficient for the second sub correction filter unit 61b is determined, and the third sub correction filter unit 61c and the fourth sub correction filter unit are determined based on the upper limit frequency and the lower limit frequency in the band information of the rear seat standing wave frequency. A filter coefficient of 61d is determined. The filter coefficient table unit 63 corresponds to the filter coefficient table means in the present invention.

  FIG. 12 is a diagram showing a filter coefficient table. In the filter coefficient table, each value is calculated in advance based on the acoustic characteristics in the passenger compartment. The filter coefficient table includes a value from −180 degrees to 175 degrees as a phase difference between a standing wave frequency at the front seat side and a standing wave frequency from the rear seat side, and a value from 20 Hz to 495 Hz as a lower limit frequency. As the upper limit frequency, there are front seat filter coefficient values a2, a3, b1, b2 and rear seat filter coefficient values a2, a3, b1, b2 corresponding to combinations of values from 25 Hz to 500 Hz. Is set.

  The filter coefficient set in the filter coefficient table is the filter coefficient table indicating the phase difference between the standing wave frequency on the front seat side and the standing wave frequency on the rear seat side between the lower limit frequency and the upper limit frequency in the setting items of the table. Filter coefficients for correcting (filtering) the phase difference values described in the above are set in advance. Therefore, in the filter coefficient table unit 63, based on the filter coefficient table, the phase difference information acquired from the level determination unit 62, and the standing wave frequencies of the front seat side and the rear seat side acquired from the standing wave detection unit 21 are used. By using the values of the upper limit frequency and the lower limit frequency in, the filter coefficients in the first sub correction filter unit 61a to the fourth sub correction filter unit 61d can be obtained.

  FIG. 13 is a block diagram showing the internal structure of the first sub correction filter unit 61a to the fourth sub correction filter unit 61d. Each of the sub correction filter units 61a to 61d is an all-pass filter using a secondary IIR filter having a direct II transposition configuration. In FIG. 13, K represents a gain, and gains corresponding to the respective codes b1, b2, a1, and a2 are set based on the filter coefficients obtained from the filter coefficient table. Z indicates a delay, and a delay of one sample is set in the sub correction filter units 61a to 61d according to the present embodiment. In the first sub correction filter unit 61a to the fourth sub correction filter unit 61d, the same coefficient is set for each of a2, a3, b1, and b2 in which the values of the filter coefficients obtained from the filter coefficient table are set. Yes.

  The level determination unit 62 has a role of changing the phase difference of the filter coefficient table unit in steps of 5 degrees from −180 degrees to 175 degrees and outputting the set phase difference information to the filter coefficient table unit 63. Yes. Then, the filter coefficient table unit 63 sets the filter coefficients of the first sub correction filter unit 61a to the fourth sub correction filter unit 61d based on the phase difference information acquired from the level determination unit 62. In the level determination unit 62, a difference signal based on the set filter coefficient is acquired from the difference calculation unit 54 of each of the standing wave detection unit 31a and the standing wave detection unit 31b, and the left front seat (passenger seat) is obtained. The difference signal from the standing wave detection unit 31a corresponding to the measurement microphone FLM installed and the difference signal from the standing wave detection unit 31b corresponding to the measurement microphone FRM installed in the right front seat (driver's seat). The amount of increase / decrease in the signal level is synthesized to determine the amount of increase / decrease in the signal level on the front seat side with respect to the phase difference.

  Similarly, the level determination unit 62 generates a difference signal based on the filter coefficients set by the first sub correction filter unit 61a to the fourth sub correction filter unit 61d, and the standing wave detection unit 31c and the standing wave detection. The difference signal from the standing wave detection unit 31c corresponding to the measurement microphone RLM installed in the left rear seat and the measurement microphone RRM installed in the right rear seat are acquired from the respective difference calculation units 54 of the unit 31d. The increase / decrease amount of the signal level with respect to the phase difference is obtained by synthesizing the increase / decrease amount of the signal level with the difference signal from the corresponding standing wave detection unit 31d.

  In FIG. 14, as shown in FIG. 9B, by setting the threshold to −6 dB, a deep dip is detected based on the lower limit frequency 80 Hz and the upper limit frequency 140 Hz of the standing wave frequency determined on the rear seat side. Increase / decrease amount (increase / decrease change amount) in signal level of the difference signal on the front seat side and difference signal on the rear seat side when the generated rear seat side phase difference is controlled between −180 degrees and 180 degrees. It is the graph which showed the value with the increase / decrease amount (increase / decrease change amount) of the signal level.

  As shown in the graph of FIG. 14, the increase / decrease amount on the rear seat side increases to the plus side as the phase difference decreases from 0 degree to −180 degrees with reference to the case where the phase difference is 0 degree. Moreover, the increase / decrease amount tends to increase toward the plus side as the phase difference increases from 0 degree to 180 degrees.

  On the other hand, the increase / decrease amount on the front seat side is stable with no change when the phase difference is between −90 degrees and 0 degrees, but the increase / decrease amount is negative as it decreases from −90 degrees to −180 degrees. In addition, the amount of increase / decrease tends to decrease to the minus side as the phase difference increases from 0 degrees to 180 degrees.

  Here, when the depth of the dip on the front seat side shown in FIG. 9A is compared with the depth of the dip on the rear seat side shown in FIG. 9B, the rear seat is compared with the front seat side. The depth of the dip on the side is deeper (the signal level at which the dip occurs is lower). For this reason, it is possible to reduce the dip on the rear seat side by actively controlling the phase difference on the rear seat side where the deep dip occurs, and to control the dip on the front seat side by the phase difference control. desirable.

  Therefore, in the graph shown in FIG. 14, the change in the increase / decrease amount on the front seat side is small, and the increase / decrease amount on the rear seat side is large (that is, the rear seat side can be greatly affected by the phase difference adjustment). By obtaining the phase difference, it is possible to greatly improve the signal level on the rear seat side while suppressing fluctuations in the signal level on the front seat side. For this reason, in the case shown in FIG. 14, it is possible to determine −90 degrees, which is a phase difference in which the change in the increase / decrease amount on the front seat side is small and the increase / decrease amount on the rear seat side is large, as the optimum phase difference. it can.

  Based on the amount of change in the signal level on the front seat side and the amount of change in the signal level on the rear seat side as shown in FIG. A value of −90 degrees is detected as a phase difference in which the increase / decrease amount on the rear seat side becomes large. Then, the detected phase difference value is output to the filter coefficient table unit 63. Note that the difference calculation unit 54 and the level determination unit 62 of the standing wave detection units 31a to 31d obtain a difference signal between the front seat side and the rear seat side, the change in the amount of increase / decrease on the front seat side is small, and the rear It corresponds to the phase difference detecting means according to the present invention in that a phase difference in which the increase / decrease amount on the seat side is detected is detected.

  In this embodiment, the case of controlling the phase difference on the rear seat side where a deep dip occurs will be described as an example, but phase control is limited to the case of controlling only the phase on the rear seat side. Not. For example, if a dip is detected in the measurement microphones on the front seat side and the rear seat side, the entire signal level on the front seat side and the rear seat side will be optimized. When a deep dip occurs, it is possible to use a method of controlling the phase difference on the front seat side.

  The filter coefficient table unit 63 that has acquired the value of the phase difference of −90 degrees from the level determination unit 62 obtains a filter coefficient to be set in the main correction filter unit 23 based on the phase difference of −90 degrees. Based on the filter coefficient table shown in FIG. 12, the filter coefficients of the sub correction filter unit 61a and the sub correction filter unit 61b obtained when the phase difference is −90 degrees and the front seat side upper limit frequency is 140 Hz are as follows. Become.

Filter coefficients of the front seat sub correction filter section (first sub correction filter section 61a and second sub correction filter section 61b)
a2 = -1.9810692365072391
a3 = 0.98154930488456404
b1 = 0.98154930488456404
b2 = -1.9810692365072391

  Further, based on the filter coefficient table shown in FIG. 12, the filter coefficients of the sub correction filter units 61 c and 61 d calculated when the phase difference is −90 degrees and the rear seat side lower limit frequency is 80 Hz are as follows. become that way.

Filter coefficients of the rear seat sub-correction filter units (the third sub-correction filter unit 61c and the fourth sub-correction filter unit 61d)
a2 = -1.98133469305836
a3 = 0.98154930488456404
b1 = 0.98154930488456404
b2 = -1.98133469305836

  The phase characteristics of the all-pass filter when the phase difference is −90 degrees are shown in FIGS. 15 and 16. FIG. 15 shows the phase characteristics (phase values) of the front seat side and the rear seat side, and FIG. 16 shows the phase values when the front seat phase characteristics are subtracted from the rear seat phase characteristics. Show. As shown in FIG. 15, there is a delay in the phase on the front seat side with respect to the rear seat side, and the phase difference shows a maximum value of −90 degrees near 110 Hz as shown in FIG.

  FIG. 17 shows that when the filter coefficients described above are set, the signals subjected to the filter processing in the first sub correction filter unit 61a to the fourth sub correction filter unit 61d are averaged by the logarithmic averaging processing units 53a and 53b. The frequency characteristics after the measurement is performed, and the frequency characteristics when the phase control is actually performed by the first sub-correction filter unit 61a to the fourth sub-correction filter unit 61d and when the phase control is not performed are shown for each measurement microphone. It is the figure shown for every standing wave detection unit 31a-31d (measuring microphone FLM, FRM, RLM, RRM) corresponding to 15 installation positions.

  As shown in FIG. 17, when correction processing (phase processing) based on the optimum phase difference is performed by the sub correction filter units 61a to 61d, the frequency characteristics on the front seat side hardly change, but the frequency characteristics on the rear seat side In the band around 110 Hz, the signal level is greatly increased, and it can be determined that the dip due to the interference between the speakers is reduced.

  Further, FIG. 18 shows a logarithmic averaging processing unit 53a, similarly to FIG. 17, using filter coefficients when the lower limit frequency is set to 80 Hz, the upper limit frequency is set to 140 Hz, and the phase difference is set to −180 degrees in the filter coefficient table. It is the figure which showed the frequency characteristic of the signal output from 53b for every standing wave detection unit 31a-31d (measurement microphone FLM, FRM, RLM, RRM) corresponding to the installation position of each measurement microphone 15. FIG.

  As shown in FIG. 18, when phase control is performed by the sub correction filter units 61a to 61d, the signal level on the front seat side decreases, but the signal level is further higher than that in FIG. It can be seen that the dip due to interference between the speakers is reduced. Therefore, when making adjustments, when the rear seat dip is extremely deep, etc., a phase difference that actively reduces rear seat interference while taking into account some reduction in the front seat signal level. In some cases, it is also effective to perform the correction processing by setting.

  As described above, in the filter coefficient calculation unit 22, the interference between the speakers is based on the standing wave frequencies on the front seat side and the rear seat side acquired from the standing wave detection unit 21 (standing wave determination unit 32). Select filter coefficients to reduce dip due to.

  Next, the main correction filter unit 23 will be described. As shown in FIG. 4, the main correction filter unit 23 includes four all-pass filters having the same configuration as the first sub correction filter unit 61 a to the fourth sub correction filter unit 61 d described in FIGS. 10 and 11. ing. In the main correction filter unit 23 according to the present embodiment, as shown in FIG. 4, four all-pass filters are connected to the first sub correction filter unit 61 a to the fourth sub correction filter unit 61 d described in FIGS. 10 and 11. Use the same sign.

  As shown in FIG. 4, the L channel audio signal and the R channel audio signal are respectively input to the main correction filter unit 23. The L-channel audio signal input to the main correction filter unit 23 includes a first sub correction filter unit 61a for performing phase control of the output signal on the left front seat side, and phase control of the output signal on the left rear seat side. In the third sub correction filter unit 61c for performing the filtering process based on the filter coefficient obtained by the filter coefficient calculation unit 22. The R channel audio signal input to the main correction filter unit 23 includes a second sub correction filter unit 61b for controlling the phase of the output signal on the right front seat side, and the phase of the output signal on the right rear seat side. In the fourth sub correction filter unit 61d for performing control, filter processing based on the filter coefficient obtained by the filter coefficient calculation unit 22 is performed.

  In this filter processing, the filter coefficient in the main correction filter unit 23 is the filter coefficient obtained in the filter coefficient calculation unit 22 described above, and is based on the phase difference between the audio signal on the front seat side and the audio signal on the rear seat side. The value of the filter coefficient that can most effectively suppress the occurrence of dip due to interference. Therefore, the audio signal filtered by the first sub correction filter unit 61a to the fourth sub correction filter unit 61d constituting the main correction filter unit 23 is installed in the passenger compartment after the signal level is amplified by the amplifier unit 13. When the sound is output from the speaker 14 (FL, FR, RL, RR), dip that may occur in each seat can be effectively suppressed.

  In particular, in standing wave control apparatus 20 according to the present embodiment, it is possible to effectively suppress dip that is likely to occur on the rear seat side. Therefore, the audio signal output from the speaker 14 (FL, FR, RL, RR) hardly affects the front seat side in the main correction filter unit 23 and reduces interference between the speakers on the rear seat side. Is output from the speaker 14 as a possible signal.

  As described above, the standing wave control apparatus 20 according to the present embodiment uses the same M-sequence code and orthogonal M-sequence code to determine standing wave interference based on the difference between them, so that a dip occurs. It is possible to easily detect the seat position in the passenger compartment and its frequency band.

  Further, since the standing wave control device 20 according to the present embodiment does not need to perform signal level amplification or the like as in the conventional frequency correction, noise or distortion is increased in the amplifier unit 13 or the speaker 14. It can be prevented from occurring. In addition, since the audio signal output from the speaker 14 can be attenuated in advance, the overall power consumption can be reduced.

  The filter coefficient determination device according to the present invention has been described in detail using the standing wave control device 20 as an example and with reference to the drawings. However, the filter coefficient determination device according to the present invention is described in the above embodiment. It is not limited to what was illustrated. It will be apparent to those skilled in the art that various changes and modifications can be made within the scope of the claims, and these are naturally within the technical scope of the present invention. Understood.

  For example, in the main correction filter unit 23, four all-pass filters (first sub correction filter unit 61a to fourth sub correction filter unit 61d) are used for correction filters for other frequencies of the input audio signal. When applying the above, it is possible to add another all-pass filter.

10: Standing wave measuring device 11: Measurement signal reproduction unit (measurement signal reproduction means)
12 ... Speaker selection unit 13 ... Amplifier unit 14, FL, FR, RL, RR ... Speaker (front seat speaker, rear seat speaker)
15, FLM, FRM, RLM, RRM ... Measurement microphone (front seat measurement microphone, rear seat measurement microphone)
16: Measuring unit (impulse response measuring means)
17 Data recording unit 20 Standing wave control device (filter coefficient determination device)
21 ... Standing wave detector (filter coefficient determination device)
22 ... Filter coefficient calculation unit (filter coefficient determination device)
23 ... Main correction filter units 31, 31a to 31d ... Standing wave detection unit 32 ... Standing wave determination unit (front seat side standing wave frequency detection means, rear seat side standing wave frequency detection means)
41-44 ... Impulse response section (first signal synthesis means, third signal synthesis means, fifth signal synthesis means, sixth signal synthesis means)
45 to 48... Impulse response section (second signal synthesis means, fourth signal synthesis means)
51a, 51b... Adder (first signal synthesis means, second signal synthesis means, third signal synthesis means, fourth signal synthesis means, fifth signal synthesis means, sixth signal synthesis means)
52a, 52b ... frequency conversion units 53a, 53b ... logarithmic averaging processing unit 54 ... difference calculation unit (front seat side standing wave frequency detection means, rear seat side standing wave frequency detection means, phase difference detection means)
61a ... 1st sub correction filter part (front seat side filter processing means)
61b ... 2nd sub correction filter part (front seat side filter processing means)
61c ... 3rd sub correction filter part (rear seat side filter processing means)
61d Fourth sub correction filter unit (rear seat side filter processing means)
62 ... Level determination unit (phase difference detection means)
63 ... Filter coefficient table section (filter coefficient table means)

Claims (2)

A front seat speaker installed on the front seat side of the vehicle interior where the seats are arranged in the front and rear, a rear seat speaker installed on the rear seat side,
A front seat measurement microphone installed at the front seat position in the vehicle interior, and a rear seat measurement microphone installed at the rear seat position;
The signal output from the front seat speaker is measured by the front seat measurement microphone and the rear seat measurement microphone, and the signal output from the rear seat speaker is measured by the front seat measurement microphone and the rear seat measurement microphone. An impulse response measuring means for measuring an impulse response corresponding to each combination from each speaker to each measurement microphone;
Measurement signal reproducing means capable of outputting different M-sequence code signals orthogonal to each other;
Using the impulse response from the front seat speaker to the front seat measurement microphone measured by the impulse response measurement means, convolution integration of the one M-sequence code signal orthogonal to each other output from the measurement signal reproduction means And using the impulse response from the rear seat speaker to the front seat measurement microphone measured by the impulse response measuring means, the other M-sequence codes orthogonal to each other output from the measurement signal reproducing means. The signal is subjected to convolution integration, and the respective orthogonal M-sequence code signals subjected to the convolution integration are combined to generate a first signal that does not include the influence of interference between the front seat speaker and the rear seat speaker. First signal combining means for generating;
Using the impulse response from the front seat speaker to the front seat measurement microphone measured by the impulse response measurement means, convolution integration of the one M-sequence code signal orthogonal to each other output from the measurement signal reproduction means And the convolution of the one M-sequence code signal output from the measurement signal reproduction means using the impulse response from the rear seat speaker to the front seat measurement microphone measured by the impulse response measurement means. A second signal that performs integration and combines the respective M-sequence code signals that have undergone convolution integration to generate a second signal that includes the influence of interference between the front seat speaker and the rear seat speaker. Combining means;
In the frequency characteristic of the signal obtained by subtracting the first signal from the second signal, a frequency at which the signal level indicates a minimum value is obtained, and the frequency including the frequency indicating the minimum value is equal to or lower than a predetermined signal level. Front seat side standing wave frequency detecting means for obtaining a range as a front seat side standing wave frequency;
Using the impulse response from the front seat speaker to the rear seat measurement microphone measured by the impulse response measuring means, convolution integration of the one M-sequence code signal orthogonal to each other output from the measurement signal reproducing means And using the impulse response from the rear seat speaker to the rear seat measurement microphone measured by the impulse response measuring means, the other M-sequence codes orthogonal to each other output from the measurement signal reproducing means. The signal is subjected to convolution integration, and the respective orthogonal M-sequence code signals subjected to the convolution integration are combined to generate a third signal that does not include the influence of interference between the front seat speaker and the rear seat speaker. Third signal synthesizing means for generating;
Using the impulse response from the front seat speaker to the rear seat measurement microphone measured by the impulse response measuring means, convolution integration of the one M-sequence code signal orthogonal to each other output from the measurement signal reproducing means And the convolution of the one M-sequence code signal output from the measurement signal reproduction means using the impulse response from the rear seat speaker to the rear seat measurement microphone measured by the impulse response measurement means. A fourth signal for performing integration and synthesizing the respective M-sequence code signals subjected to convolution integration to generate a fourth signal including the influence of interference between the front seat speaker and the rear seat speaker; Combining means;
In the frequency characteristic of the signal obtained by subtracting the third signal from the fourth signal, a frequency at which the signal level indicates a minimum value is obtained, and the frequency including the frequency at which the minimum value is included is equal to or lower than a predetermined signal level. A rear seat side standing wave frequency detecting means for obtaining a range as a rear seat side standing wave frequency;
The front seat determined based on the upper limit frequency and the lower limit frequency in the frequency range of the front seat side standing wave frequency and the value of the phase difference of the front seat side standing wave frequency with respect to the rear seat side standing wave frequency A side filter coefficient, an upper limit frequency and a lower limit frequency in a frequency range of the rear seat side standing wave frequency, and a rear seat filter coefficient determined based on the value of the phase difference, the various upper limit frequencies, Filter coefficient table means comprising a filter coefficient table calculated in advance according to acoustic characteristics in the vehicle interior based on various combinations of the lower limit frequency and various values of the phase difference;
Front-seat-side filter processing means for performing filter processing on the one M-sequence code signal output from the measurement signal reproducing means based on the front-seat-side filter coefficient;
Rear seat filter processing means for performing filter processing on the one M-sequence code signal output from the measurement signal reproducing means based on the rear seat filter coefficient;
The impulse response from the front seat speaker to the front seat measurement microphone is used to perform convolution integration of the one M-sequence code signal filtered by the front seat side filter processing means, and from the rear seat speaker The impulse response to the front seat measurement microphone is used to perform convolution integration of the one M-sequence code signal filtered by the front seat side filter processing means, and each one of the ones subjected to convolution integration is performed. A fifth signal synthesizing unit that synthesizes an M-sequence code signal and generates a fifth signal that does not include the influence of interference between the front seat speaker and the rear seat speaker;
The impulse response from the front seat speaker to the rear seat measurement microphone is used to perform convolution integration of the one M-sequence code signal filtered by the rear seat side filter processing means, and from the rear seat speaker Using the impulse response to the rear seat measurement microphone, convolution integration of the one M-sequence code signal filtered by the rear seat filter processing means is performed, and each of the one subjected to convolution integration is performed. A sixth signal synthesizing unit that synthesizes an M-sequence code signal to generate a sixth signal that does not include the influence of interference between the front seat speaker and the rear seat speaker;
Subtracting the second signal from the fifth signal to obtain the front-seat differential signal and subtracting the fourth signal from the sixth signal to obtain the rear-seat differential signal, By detecting an increase / decrease change of the signal level in the difference signal on the seat side and an increase / decrease change amount of the signal level in the difference signal on the rear seat side in association with the value of the phase difference set in the filter coefficient table means, Either the increase / decrease change amount of the signal level on the front seat side or the increase / decrease change amount of the signal level on the rear seat side does not increase / decrease, and the phase difference that greatly increases / decreases the other change amount Phase difference detection means for obtaining a value,
The filter coefficient table means is obtained from the filter coefficient table based on the phase difference value detected by the phase difference detecting means and the upper limit frequency and the lower limit frequency in the frequency range of the front seat side standing wave frequency. The front seat side filter coefficient is determined as a filter coefficient for performing a filtering process on the audio signal output from the front seat speaker, and the value of the phase difference detected by the phase difference detecting means and the rear seat Filtering the rear seat filter coefficient obtained from the filter coefficient table based on the upper limit frequency and the lower limit frequency in the frequency range of the side standing wave frequency on the audio signal output from the rear seat speaker A filter coefficient determination device characterized in that the filter coefficient is determined. The predetermined signal level set for obtaining the front seat side standing wave frequency and the predetermined signal level set for obtaining the rear seat side standing wave frequency are -6 dB. The filter coefficient determination apparatus according to claim 1.

Images