TW201626814A - Compensator system for frequency response of loudspeaker - Google Patents

Abstract

A compensating system for frequency response of loudspeaker, comprising: a model unit for producing desired sound pressure based upon input voltage. Without sensors, a sound pressure estimation unit for estimating the cone velocity of the loudspeaker according to the voltage and current of a voice coil of the loudspeaker and calculating a sound pressure of a sound field around the loudspeaker according to the voice coil; with sensors, the sensors can sense the sound pressure of the sound field, a comparing unit for comparing the desired sound pressure and the sound pressure of the sound field to get the difference, and a compensation unit for adjusting the frequency response of the loudspeaker based on the difference.

Description

Speaker frequency response compensation system

The present invention relates to a frequency response compensation technique for a speaker, and more particularly to a frequency response compensation system for a speaker that measures or estimates the diaphragm speed to compensate for the speaker and the ambient sound field.

Due to the miniaturization of consumer electronic devices, the development trend of speakers is small and light, which not only limits the size of the speakers, but also causes distortion in the speakers. In this case, many studies have proposed speaker compensation methods. Thereby, nonlinear distortion is reduced. For example, research has proposed adaptive control to compensate for the nonlinear characteristics of the speaker.

Regarding the improvement of the room response, in general, because the speaker diaphragm cannot produce a sufficient volume velocity lower than the mechanical response frequency, the moving coil speaker exhibits a poor response in the low frequency range, based on The enhancement of the frequency response is very important for the reproduction of the sound signal. It is very important to improve the frequency response. One way to improve the low frequency response is to increase the radius of the speaker. However, the increase in efficiency cannot be as expected because of the speaker. The mass also increases with increasing radius. Another way to improve speaker response is to make With quantitative feedback technique (QFT), in this method, the model matching method is prone to open-loop problems. In addition, other traditional methods include electronic compensation. These audio systems are equipped with an equalizer to enhance the bass output. In this case, only the amplitude of the low frequency response is increased, rather than all frequency responses. It can be seen from the above that the improvement of the frequency response, especially through the means of electronic compensation, will contribute to the reproduction of the audio signal.

Therefore, how to find a frequency response compensation technology for a speaker can provide a pre-compensation of the speaker for a fixed coefficient or an adaptive algorithm, in particular, a different situation with a sensor or a sensorless It has become the goal pursued by those skilled in the art.

In view of the above disadvantages of the prior art, it is an object of the present invention to provide a compensation system that provides an estimate or a adaptive algorithm to provide a speaker or ambient sound with or without a sensor. Different compensation of the field to increase the low frequency gain and flatten the frequency response.

To achieve the foregoing and other objects, the present invention provides a frequency response compensation system for a speaker, including a model unit, a sound pressure estimation unit, a comparison unit, and a compensation unit, wherein the model unit generates an ideal sound pressure according to an input voltage. The sound pressure estimating unit includes a diaphragm speed measuring device that estimates a diaphragm speed of the speaker according to a voltage and a current of the voice coil in the speaker, and calculates a sound field of the speaker according to the diaphragm speed. a sound pressure differentiator, the comparison unit calculates the ideal sound pressure and the sound pressure of the sound field to obtain a difference, and the compensation unit adjusts the frequency of the speaker by the difference response.

In one embodiment, the coefficient of the speaker is fixed, and the compensation unit adjusts the input voltage according to the difference.

In another embodiment, the compensation unit further includes an adaptive algorithm for adjusting the coefficients of the speaker in a convolution manner.

In still another embodiment, adjusting the frequency response of the speaker using the adaptive algorithm includes offline operation and online operation.

The invention also provides a frequency response compensation system for a speaker, which comprises a model unit, a sensing unit, a comparison unit and a compensation unit, wherein the model unit generates an ideal sound pressure according to an input voltage, and the sensing unit is used for sensing The sound pressure of the sound field of the speaker is measured, the comparison unit calculates the ideal sound pressure and the sound pressure of the sound field to obtain a difference, and the compensation unit adjusts the frequency response of the speaker with the difference.

In one embodiment, the coefficient of the speaker is fixed, and the compensation unit adjusts the input voltage according to the difference.

In another embodiment, the compensation unit further includes an adaptive algorithm for adjusting the coefficients of the speaker in a convolution manner.

In still another embodiment, adjusting the frequency response of the speaker using the adaptive algorithm includes offline operation and online operation.

Compared with the prior art, the frequency response compensation system of the speaker provided by the present invention provides different execution schemes of the sensing unit and the non-sensing unit. Under the sensing unit, the sound pressure sensed by the sensing unit can be detected. As the basis for the compensation judgment, if there is no sensing unit, the voltage and current of the voice coil of the speaker are estimated to calculate the sound pressure of the sound field of the speaker, and then Regardless of whether there is a sensor or not, the two schemes can be matched with the fixed or variable coefficient of the speaker, that is, only the compensation of the speaker can be considered, and the coefficient of the speaker can be fixed, if the speaker is also considered In the sound field environment, an adaptive algorithm can be selected to adjust the coefficients of the speaker. It can be seen from the above that there will be four different situations to perform the compensation of the frequency response of the speaker, and through the above different compensation schemes, the better frequency response compensation effect of the speaker will be provided.

1, 2‧‧‧Speaker frequency response compensation system

11, 21‧‧‧ model unit

12‧‧‧Sound Pressure Estimation Unit

121‧‧‧diaphragm speed measuring device

122‧‧‧Differentiator

13, 23‧‧‧ comparison unit

14, 24‧‧‧Compensation unit

22‧‧‧Sensor unit

1 is a system architecture diagram showing an embodiment of a frequency response compensation system for a speaker of the present invention; and FIG. 2 is a system architecture diagram showing another embodiment of a frequency response compensation system for a speaker of the present invention; A schematic diagram of an adaptive filter according to the present invention; 4A-4C is a schematic diagram showing a feedforward compensation system in different states according to the present invention; and FIG. 5 is a diagram showing a plant including a speaker and a user microphone. 6A and 6B are schematic views showing an analog circuit of a moving coil type speaker according to the present invention; and Fig. 7 is a schematic view showing speaker compensation without a sensing unit.

The embodiments of the present invention are described below by way of specific embodiments, and those skilled in the art can easily understand the present disclosure. Other features and effects of the invention. The invention may also be embodied or applied by other different embodiments.

In order to increase the low frequency gain and flatten the frequency response, the present invention proposes two feedforward compensation methods, which are adaptive speaker compensation with sensors and adaptive speaker compensation without sensors, each. The methods include both online and offline operations, with a sensor-compensated microphone for receiving the estimated signal and an adaptive filtering system for updating the compensator to produce the best signal.

It is necessary to estimate the cone velocity of the speaker when the speaker is not compensated by the sensor and the room response. The speaker diaphragm speed has always been considered as an important parameter of the speaker compensation, and the direct access needs of the diaphragm speed. A detector, such as an accelerometer or a laser vibrometer, may cause an undesired mass burden on the speaker. The speaker protection can be achieved by limiting the output voltage by estimating the diaphragm speed. Another application is dome-shaped (dome) For compensation, the Klippe acoustic analysis system measures the input current to evaluate the maximum voice coil peak displacement. In the present invention, the sensorless compensation is to measure the voltage of the speaker to estimate the diaphragm speed, and the sound pressure obtained from the diaphragm speed, and finally, an adaptive algorithm is used to compensate the sound pressure.

The compensation method for each operation is designed to use model matching and implemented in a digital signal processor (DSP). The digital signal processor is implemented in a fixed-point algorithm. The fixed-point algorithm is well known to those skilled in the art, and the present invention omits the description. In addition, the present invention also proposes to use an adaptive algorithm based algorithm. Model matching program, Its characteristics are simple in principle and easy to apply.

Referring to Fig. 1, there is shown a system architecture diagram of an embodiment of a frequency response compensation system for a loudspeaker of the present invention. As shown in the figure, the frequency response compensation system 1 of the speaker includes: a model unit 11, a sound pressure estimation unit 12, a comparison unit 13, and a compensation unit 14, which can measure the voltage and current of the voice coil to obtain a basis for how to compensate. .

The model unit 11 generates an ideal sound pressure according to the input voltage. The ideal sound pressure described herein refers to the expected sound pressure, and if this standard is reached, a good frequency response is exhibited.

The sound pressure estimation unit 12 includes a diaphragm speed measurer 121 and a differentiator 122, and the sound pressure estimation unit 12 receives the measurement data to estimate the sound pressure. In detail, the diaphragm speed measuring device 121 is configured to estimate the diaphragm speed of the speaker according to the voltage and current of the voice coil in the speaker (the voltage is divided by the current), and the voltage can be directly measured. It is determined that the current can be connected to the voice coil by a resistor of a known resistance value, and the voltage is divided by the voltage across the resistor to obtain a current. The diaphragm speed measuring device 121 can estimate the diaphragm speed based on the voltage and current of the voice coil. In a specific embodiment, the sound pressure estimation unit 12 can be a sound pressure estimator or a sound pressure estimation component.

After the diaphragm speed is obtained, the sound pressure of the sound field of the speaker can be calculated by the differentiator 122 according to the diaphragm speed. This is the sound pressure estimated by the diaphragm speed estimation method for subsequent comparison of the sound pressure and the ideal sound. The difference in pressure.

The comparing unit 13 calculates the difference between the ideal sound pressure of the model unit 11 and the sound pressure of the sound field obtained by the sound pressure estimating unit 12, in order to compare the estimated sound pressure with the ideal sound pressure. The difference, so, can be used as Reference for subsequent compensation. In a particular embodiment, comparison unit 13 can be a comparator.

The compensation unit 14 adjusts the frequency response of the speaker with the above difference, that is, the difference can be transmitted back to the compensation unit 14, thereby determining which compensation adjustment is required. In a particular embodiment, the compensation unit 14 can be a compensator.

In the specific implementation of the embodiment, the model unit 11, the sound pressure estimation unit 12 or the compensation unit 14 can be implemented by a filter.

The above embodiment refers to the compensation of the frequency response of the speaker without the sensor. In this case, the coefficient of the speaker is fixed, and the compensation unit 14 can adjust the input voltage according to the difference.

In addition, if the sound field of the speaker is considered, an adaptive compensation adjustment can be provided, that is, the compensation unit 14 has an adaptive algorithm, and the adaptive algorithm can be used to adjust the speaker in a convolution manner. The coefficients are used to compensate both the speaker and the sound field environment.

Referring to Figure 2, there is shown a system architecture diagram of another embodiment of a frequency response compensation system for a loudspeaker of the present invention. As shown, the frequency response compensation system 2 of the speaker includes a model unit 21, a sensing unit 22, a comparison unit 23, and a compensation unit 24, which can sense the sound pressure through the sensor to determine the basis for compensation.

The model unit 21 generates an ideal sound pressure according to the input voltage. The ideal sound pressure described herein refers to the expected sound pressure, and if this standard is reached, a good frequency response is exhibited.

The sensing unit 22 is configured to sense the sound pressure of the sound field where the speaker is located. Different from the embodiment shown in FIG. 1, this embodiment uses a sensor to sense sound. Pressure, so there is no need to measure the voltage and current of the voice coil. In a specific embodiment, the sensing unit 22 can be a microphone.

The comparing unit 23 calculates the difference between the ideal sound pressure of the model unit 21 and the sound pressure of the sound field obtained by the sensing unit 22, by comparing the difference between the estimated sound pressure and the ideal sound pressure, Can be used as a reference for subsequent compensation.

The compensation unit 24 adjusts the frequency response of the speaker with the above difference, that is, the difference can be transmitted back to the compensation unit 24, thereby determining what compensation adjustment is required.

In the embodiment, the model unit 21, the sensing unit 22 or the compensation unit 24 can also be implemented by a filter. As described above, the present embodiment refers to the compensation of the frequency response of the speaker under the sensor. In this case, the coefficient of the speaker is fixed, and the compensation unit 24 can adjust the input voltage according to the difference.

In addition, if the sound field of the speaker is considered, an adaptive compensation adjustment can be provided, that is, the compensation unit 24 has an adaptive algorithm, and the adaptive algorithm can be used to adjust the speaker in a convolution manner. The coefficients are used to compensate both the speaker and the sound field environment.

In addition, when the frequency response of the speaker is adjusted using an adaptive algorithm, two scenarios of offline operation and online operation can be provided. The details are as shown in Figs. 4B and 4C of the present invention.

In the following, a more detailed implementation of the circuit diagram will be used to illustrate various mechanisms for adjusting the speaker coefficient or compensating for the room response under different situations or configurations.

First, Figure 3 shows an adaptive filter. Some signal propagation systems are time-varying and unknown, and adaptive filters are useful for these applications. Adaptive systems automatically modify their characteristics to accomplish certain aims. As shown, d(n) is the desired signal, y(n) is the output signal of the digital filter driven by the input signal x(n) , and the error signal e(n) is the desired signal d(n) and the output signal. The difference between y(n) , the purpose of the adaptive algorithm is to adjust the coefficients of the digital filter to minimize the mean square value of the error signal e(n) , and the digital filter can have a finite impulse response. (FIR) filter to achieve.

4A-4C shows a feedforward compensation system in a different state, wherein FIG. 4A illustrates the speaker compensation with the sensor, and FIG. 4B illustrates the speaker compensation under the offline operation if the system is linear non-time varying. Figure 4C illustrates the speaker compensation for operation on the line.

In order to find the impulse response of the compensator c(n) under offline operation, an adaptive filtering system can be employed. h(n) is the impulse response of the plant. The device consists of a speaker and a user microphone. As shown in Figure 5, the model m(n) is designed according to the device h(n) , the first resonant frequency of the model. Below h(n) , therefore, the amplitude at the low frequency is increased, and the expected signal d(n) is derived from the input signal x(n) and the model m(n) as shown in the following equation: d ( n )= m ( n ) * x ( n ) (1)

Where "*" denotes a convolution operation, assuming that the system is linear time invariant (LTI), the position of the device h(n) and the compensator c(n) can be changed, as shown in Fig. 4B, the output signal y(n) is derived from the following estimation pathways: x' ( n )= h ( n )* x ( n ) (2)

y ( n )= c ( n )* x' ( n ) (3)

Where x'(n) is the output signal from device h(n) , y(n) is the output signal of compensator c(n) , and error signal e(n) can be defined as: e ( n )= d ( n )- y ( n ) (4)

In order to minimize the error signal, the adaptive algorithm updates the coefficients of the compensator c(n) , and the update equation is defined as: c ( n +1)= c ( n )+ μe ( n ) x' ( n ) (5 )

Where μ is the step size, so the compensator c(n) under offline operation can be obtained.

The compensator c(n) obtained in the offline operation is fixed. In fact, the device is time-varying, and therefore, an online operation is also proposed, as shown in Fig. 4C. In the online operation, the signal received by the microphone y(n) is traced to the desired signal d(n) .

d ( n )= m ( n )* x ( n ) (6)

The model m(n) produces the expected signal d(n) .

x' ( n )= c ( n )* x ( n ) (7)

The output signal x'(n) from the compensator is transmitted to the speaker, and the signal is transmitted from the speaker to the microphone and received by the microphone. After the signal y(n) from the microphone is received, the error signal e(n) can be defined as : e ( n )= d ( n )- y ( n ) (8)

In order to minimize the error signal, the proposed adaptive algorithm: c ( n +1)= c ( n )+ μe ( n ) x ( n ) (9)

The coefficients of the compensator c(n) are automatically updated.

The electrical, mechanical, and acoustic (EMA) of the speaker will be further explained below. model.

Figures 6A and 6B show an analog circuit of a moving coil type speaker, wherein Fig. 6A illustrates an equivalent circuit in the field of electrical and mechanical, and Fig. 6B illustrates an equivalent circuit for mapping an electric appliance having an impedance from the mechanical field.

As shown in Fig. 6A, the acoustic end is mapped to the equivalent impedance of the mechanical end and the mechanical end combination, wherein Z _MA is the acoustic end mapped to the equivalent impedance of the mechanical end and the mechanical end combination, and the left side of the figure represents the speaker. The electric terminal, the right side represents the mechanical end of the speaker, the mechanical end is equivalent to the voltage in the circuit, the speaker diaphragm speed is equivalent to the current, and the middle part is the Gyrator, the relationship is as follows:

Where Z _E (s) represents the impedance in the electrical domain, where s is the Laplace operator, i(s) is the current, and Z _VC (s) is the signal source e _g (s) The equivalent impedance observed, the component between the electrical and mechanical fields is a gyrator. Is a force factor, which is defined as = B1 . f(s) is the Lorentz factor caused by the current i(s ).

Z _MA (s) is the equivalent impedance tolerance of the reflected impedance from the acoustic end to the impedance of the mechanical field. therefore,

Among them, Z _mot (s) is the dynamic resistance, and u(s) is the speaker diaphragm speed. In Figure 6B, Z _VC (s) can be written as: Z _VC ( s ) = Z _E ( s ) + Z _mot ( s ) (12)

The relationship between the diaphragm speed u(s) and the voltage signal e _g (s) plays an important role in establishing the speed estimator. From Fig. 6B, the current i(s) can be written as:

In Figure 6A, the Lorentz factor can be written as f ( s ) = Z _mot ( s ) u ( s ), so replacing equation (10) with equation (13) yields:

Therefore, the ratio of u(s) to e _g (s) can be obtained, and H _eu (s) can be defined as:

Through the above formula, the diaphragm speed u(s) can be estimated by measuring the terminal voltage of the speaker, that is , the conversion function of H _eu (s) from e _g (s) to u(s) , which is circular by plane The speaker is pressed on the axis mapped by the infinite baffle. Therefore, the relationship between the diaphragm speed u(s) and the sound pressure is as follows:

Where ρ ₀ is the density of air, r is the distance from the speaker to the observation point, and k = ω / c , where c is the speed of the sound. In order to define the transfer function, it is assumed here that r = 1 m, and e - ^{s (1/ c )} in the equation (16) represents the phase delay caused by the propagation time delay from the piston to the observation point, since e ^{-s (1/ c ) It} has a uniform vibration, which is independent of its size value, so it is omitted when defining the sound pressure conversion function.

p ( s )= Ksu ( s ) (17)

Where K = ρ ₀ /2 π and the sound pressure can be calculated from the diaphragm speed.

The speed H _eu ( s ) and the differentiator Ks can be combined to the filter ,filter Can be discretized and converted to FIR filter Figure 7 shows such a speaker.

d ( n )= m ( n )* x ( n ) (18)

Where d(n) is the desired signal, x(n) is the input signal, and m(n) is the model.

x' ( n )= c ( n )* x ( n ) (19)

Where c(n) is the compensator, x'(n) is the terminal voltage of the speaker and is transmitted to the FIR filter .

Where y(n) is the estimated signal, and the error signal can be defined as: e ( n )= d ( n )- y ( n ) (21)

In order to minimize the error signal, the proposed adaptive algorithm has an update equation of c ( n +1)= c ( n )+ μe ( n ) x ( n ) (22)

Where μ is the step size.

The invention can be implemented on the C6416 development board by an adaptive algorithm and a digital signal processing method, and the music can be sent from the computer to the C6416 development board, and after the program operation, the diaphragm speed of the speaker can be obtained. The diaphragm speed can estimate the sound pressure of the sound field, and can be played back by using an adaptive algorithm to obtain the sound pressure of the ideal sound field. In the specific implementation, three filters can be used: a model unit, a sound pressure estimation unit and a compensation unit, and the model unit calculates a signal for generating an ideal sound pressure, and the sound pressure estimation unit calculates and generates The diaphragm velocity value is used to calculate the sound pressure of the sound field, and the signal of the ideal sound pressure is compared with the sound pressure of the sound field to perform an adaptive algorithm, so that a coefficient of the compensator can be obtained FIR filter, this compensator is the compensator of the speaker.

In summary, the frequency response compensation system of the speaker provided by the present invention provides different implementation schemes of the sensor and the sensorless sensor, and the sound pressure sensed by the sensor can be used as a basis for compensation judgment. If there is no sensor, the sound pressure of the sound field of the speaker is estimated by estimating the voltage and current of the voice coil of the speaker. With or without the sensor, the coefficients of the two programs can be fixed or variable with the speaker. The situation is matched, that is to say, the coefficient of the speaker is fixed, and only the compensation consideration of the speaker, if the sound field environment of the speaker is also considered, an adaptive algorithm can be selected to adjust the coefficient of the speaker. It can be seen from the above that the present invention proposes four different situations to perform the compensation of the frequency response of the speaker, and through the above different compensation schemes, the better frequency response compensation effect of the speaker will be provided.

The above-described embodiments are merely illustrative of the principles of the invention and its effects, and are not intended to limit the invention. Modifications and variations of the above-described embodiments can be made by those skilled in the art without departing from the spirit and scope of the invention. Therefore, the scope of protection of the present invention should be as set forth in the scope of the claims described below.

1‧‧‧Speaker frequency response compensation system

11‧‧‧Model unit

12‧‧‧Sound Pressure Estimation Unit

121‧‧‧diaphragm speed measuring device

122‧‧‧Differentiator

13‧‧‧Comparative unit

14‧‧‧Compensation unit

Claims (10)

A frequency response compensation system for a speaker, comprising: a model unit for generating an ideal sound pressure according to an input voltage; and a sound pressure estimating unit comprising: a diaphragm speed measuring device according to a voltage of a voice coil in the speaker Current is used to estimate the diaphragm speed of the speaker; and the differentiator calculates the sound pressure of the sound field of the speaker according to the diaphragm speed; the comparing unit calculates the ideal sound pressure and the sound pressure of the sound field to Obtaining a difference; and a compensation unit that adjusts the frequency response of the speaker with the difference. The frequency response compensation system of the speaker according to claim 1, wherein the coefficient of the speaker is fixed, and the compensation unit adjusts the input voltage according to the difference. The frequency response compensation system for a speaker according to claim 1, wherein the compensation unit further comprises an adaptive algorithm for adjusting a coefficient of the speaker in a convolution manner. The frequency response compensation system for a speaker according to claim 3, wherein the adaptive algorithm is used to adjust the frequency response of the speaker to be performed under online operation or offline operation. The frequency response compensation system for a speaker according to claim 1, wherein the model unit, the sound pressure estimation unit or the compensation unit is a filter. A frequency response compensation system for a speaker, comprising: The model unit generates an ideal sound pressure according to the input voltage; the sensing unit is configured to sense the sound pressure of the sound field of the speaker; and the comparing unit calculates the ideal sound pressure and the sound pressure of the sound field to obtain a difference And a compensation unit that adjusts the frequency response of the speaker with the difference. The frequency response compensation system for a speaker according to claim 6, wherein the sensing unit is a microphone. The frequency response compensation system for a speaker according to claim 6, wherein the coefficient of the speaker is fixed, and the compensation unit adjusts the input voltage according to the difference. The frequency response compensation system for a speaker according to claim 6, wherein the compensation unit further comprises an adaptive algorithm for adjusting a coefficient of the speaker in a convolution manner. The frequency response compensation system for a speaker according to claim 9, wherein the adaptive algorithm is used to adjust the frequency response of the speaker to be performed under online operation or offline operation.