Protecting speakers from excessive excursion

Abstract

A system and method for protecting a loudspeaker from excessive excursion includes an audio source, an adaptive excursion protection filter, an audio limiter, an inverse excursion protection filter, an amplifier, and a loudspeaker. The system performs operations including receiving an audio signal, applying an excursion protection filter that is adapted in real time to one or more loudspeaker conditions, limiting the audio signal, applying an inverse excursion protection filter, and amplifying the audio signal using an amplification circuit for output to a loudspeaker.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of and priority to U.S. Patent Application No. 16 / 672,251, filed on November 1, 2019, entitled “PROTECTION OF SPEAKER FROMEXCESS EXCURSION,” which is incorporated herein by reference in its entirety. Technical Field

[0003] Various embodiments of the present disclosure generally relate to reduction of speaker distortion and / or protection of loudspeakers during operation. More particularly, for example, the present disclosure relates to systems and methods for protecting loudspeakers from excessive excursion. Background Art

[0004] Modern consumer electronic devices such as mobile phones and tablet computers are typically designed to meet various small form factor, low power and low cost goals. The components used in these devices (including loudspeakers) can be low-cost components designed to fit into small spaces. The disadvantage of these smaller loudspeakers is that the acoustic output that can be delivered from the device to the user is reduced. In order to meet the user's expectations for more loudness or volume delivered by the loudspeaker, the loudspeaker can be driven close to its mechanical limit, which may potentially lead to catastrophic loudspeaker failure if the limit is exceeded. In view of the foregoing, there is a continuing need for systems and methods that improve the acoustic performance of (one or more) loudspeakers in consumer electronic devices while protecting (one or more) loudspeakers from damage. Summary of the invention

[0005] Disclosed herein are systems and methods for protecting a loudspeaker and / or reducing distortion due to excessive excursion. In various embodiments, the system and method for protecting a loudspeaker from excessive excursion includes an audio source, an adaptive excursion protection filter, an audio limiter, an inverse excursion protection filter, an amplifier, and a loudspeaker. The system performs operations including receiving an audio signal, applying an excursion protection filter that is adapted to one or more speaker conditions in real time, limiting the audio signal, applying an inverse excursion protection filter, and amplifying the audio signal using an amplification circuit for output to a speaker.

[0006] In some embodiments, the system includes a logic device that is further configured to perform operations including estimating filter coefficients for an offset protection filter and an inverse offset protection filter based at least in part on the amplified audio signal. The operation of estimating the filter coefficients may also include receiving current and voltage data from the amplifier. The system may also include a cone measurement sensor configured to measure a speaker excursion of the output audio signal, wherein the one or more speaker conditions include a speaker excursion measurement. The logic device may also be configured to perform operations including estimating a threshold and applying the threshold when limiting the audio signal.

[0007] In some embodiments, the operation of the estimated excursion filter further includes estimating the complex impedance of the speaker and estimating the DC resistance. The operation of the estimated excursion filter may also include calculating a motion sensitivity function representing the motion of the cone as a function of the applied voltage, and estimating two or more peaks of the calculated motion sensitivity function. The operation of the estimated excursion filter may also include applying a slew rate limit to the peak frequency and amplitude, and creating a model infinite impulse response filter based at least in part on the estimated peaks.

[0008] In some embodiments, the logic device is further configured to perform operations including testing parameter values ​​of a current frame of the audio signal, the parameter values ​​including one or more parameters used in an estimation of an excursion filter, and if a parameter outside a predetermined range is detected, identifying the frame as invalid. The logic device may also be configured to perform operations including tuning the excursion protection filter by a process including measuring a scale factor, determining a clipping level threshold, and verifying the clipping level threshold.

[0009] The scope of the present disclosure is defined by the claims, which are incorporated into this section by reference. A more complete understanding of the embodiments of the present disclosure, as well as the realization of additional advantages thereof, will be provided to those skilled in the art by consideration of the following detailed description of one or more embodiments. Reference will be made to the attached sheets in the drawings which will first be briefly described. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is a flow chart illustrating real-time speaker protection according to one or more embodiments.

[0011] Figure 2 is a flow chart illustrating model estimation for real-time loudspeaker protection in accordance with one or more embodiments.

[0012] Figure 3 A real-time adaptive algorithm for updating a model estimation filter in accordance with one or more embodiments is shown.

[0013] Figure 4An offline tuning algorithm for tuning a model estimation filter in accordance with one or more embodiments is shown.

[0014] Figure 5 An example processing system configured to protect a speaker from over-excursion in accordance with one or more embodiments is shown.

[0015] The embodiments of the present disclosure and their advantages may be best understood by referring to the detailed description below. It should be understood that the same reference numerals are used to identify the same elements shown in one or more of the accompanying drawings, wherein the illustration in the accompanying drawings is for the purpose of illustrating the embodiments of the present disclosure, not for the purpose of limiting the embodiments of the present disclosure. DETAILED DESCRIPTION

[0016] The present disclosure provides systems and methods for protecting a loudspeaker from excessive excursion during operation. In various embodiments, louder sound is obtained from a loudspeaker by driving the loudspeaker at a higher voltage / power than the standard specification limit of the loudspeaker. The systems and methods are configured to protect the loudspeaker from excessive excursion of the cone of the loudspeaker, which may result in distorted sound and / or damage to the loudspeaker.

[0017] Reference Figure 1 Describe the embodiments of the present disclosure, Figure 1 An example real-time algorithm data flow for speaker protection is shown. Algorithm 100 includes a feed-forward path 102 that receives audio data from an audio source 104 and processes the audio data for output through a loudspeaker 114. Feed-forward path 102 may operate at a sampling rate suitable for high-quality audio playback, such as 48 kHz. Other paths in algorithm 100 (e.g., paths 120 to 108, paths 112 to 130, 130 to 106, and 130 to 110) are configured to operate at a lower rate, such as a frame rate of 100 Hz. In operation, the excursion of speaker 114 is limited by the value of a clipping threshold (thr) 120 used in a look-ahead clipper 108. Look-ahead clipper 108 is configured to avoid high-frequency distortion that would be produced by using a simple instantaneous threshold.

[0018] The algorithm 100 creates a model that predicts the cone excursion of the loudspeaker given a drive voltage waveform and uses the model to limit the distortion to safe values. The algorithm 100 operates in real time to apply this limiting operation and optionally dynamically updates the parameters of the model based on the real time current and voltage applied to the loudspeaker. The offline tuning algorithm measures the behavior of the loudspeaker 114 and generates an appropriate initial model and other parameters for the real time algorithm 100.

[0019] The approximation model is configured to operate in real time to predict the excursion of the microphone 114 from the applied voltage waveform. The approximation model may be implemented as an infinite impulse response (IIR) filter S106 specified by SOS coefficients (e.g., coefficients in the second order portion of an IIR filter). The filter S applied in block 106 has a stable inverse filter S applied in block 110. -1 By applying an approximation model filter 106 (eg, an IIR filter S), followed by a look-ahead limiter 108 or other limiter, followed by an inverse filter 110 (eg, an inverse filter S -1 ), limiting the loudspeaker excursion to a safe value. The excursion model S is continuously updated in real time to adapt to changes in the loudspeaker characteristics due to factors such as aging and temperature.

[0020] An initial excursion model 130 is estimated for the loudspeaker 114. For example, for a given loudspeaker, the initial excursion model 130 is estimated based on a method that combines electromechanical loudspeaker theory and heuristics, so that an approximate model can be calculated with low computational power. The process may use current (I) and voltage (V) data from an amplifier 112 (e.g., a class D amplifier) ​​driving the loudspeaker 114, and may optionally use a sensor 132 to measure instantaneous cone excursion, such as using a laser displacement measurement device, for example.

[0021] Compared to conventional algorithms, algorithm 100 is configured to track more than one peak in the offset frequency response.Most small speakers have more than one peak, for example, because of the common use of ports, passive radiators, or multiple motors to increase the low frequency response.

[0022] refer to Figure 2 , an embodiment for estimating the model will now be described. In the illustrated embodiment, the model S is estimated at 230 based on the time domain current I and voltage V (eg, IV 202) at the loudspeaker terminals. This data can be obtained, for example, by comparing Figure 1 The input data IV is provided to the subband analysis block 210, which is configured to convert the time domain input signal into a frequency domain subband, for example, by short-term Fourier transform. The adaptive filter 220 is configured to estimate the loudspeaker complex impedance in the frequency domain. In various embodiments, the absolute value of the microphone impedance A prominent peak will be shown at the peak of the mechanical resonance of the speaker cone and its elastic confinement (e.g., a combination of the spider effect, surround, and air reaction). This is due to the back electromagnetic field (EMF) caused by the movement of the speaker / voice coil in its magnetic field. When the speed of movement is high, such as at resonance, the back EMF opposes the drive voltage, and the resulting smaller effective drive voltage applied to the DC resistance of the coil results in smaller current.

[0023] The voltage in the frequency domain (V(f)) is related to the current (I(f)), the velocity of the speaker coil and cone (velocity(f)), the DC resistance of the voice coil (Rdc), and the force constant (BL) (which is the product of the (average) magnetic field band of the length L of the voice coil winding) as follows:

[0024]

[0025] refer to Figure 3 , an embodiment of a real-time adaptive algorithm for updating filter S will now be described. In step 310, algorithm 300 receives input data representing the impedance of a loudspeaker, such as Frames of Data. In some embodiments, frames may occur, for example, at 8 millisecond intervals, thereby representing many samples of actual audio time-domain data.

[0026] In step 312, the algorithm estimates the complex impedance of the speaker using a least mean square (LMS) or other algorithm , to obtain the complex impedance , so that the norm Minimize. The norm function may be the squared magnitude summed over frequency.

[0027] In step 314, the algorithm estimates the DC resistance Rdc as Due to phase cancellation, the function can have a value less than the true value of Rdc. In various embodiments, the calculation A histogram of the values ​​of , and from this histogram, select For example, you can use the 10th percentile value, which is the value below which all relevant frequencies are lower than 90% of the value of .

[0028] In step 316, the algorithm calculates the effective motion sensitivity function , which expresses the cone motion as a function of the applied voltage: In the embodiment shown, the algorithm divides by f because motion is the integral of velocity over time, and the duration of each cycle of the signal component at f is inversely proportional to f. Therefore, motion tends to be greater at lower frequencies.

[0029] In step 318, the algorithm estimates In some embodiments, the algorithm picks peaks in a coarse grid of frequencies to be fused in the frequency domain representation, and then The algorithm refines the estimate of peak frequency and amplitude by quadratic interpolation of the values ​​of . For many loudspeakers, such as the microspeakers used in PCs, phones, and smart devices, the algorithm will pick two peaks: (i) a peak corresponding to the resonant frequency of the speaker; and (ii) a lower frequency peak that is a combination of the 1 / f factor in Gmot together with possible low-frequency acoustic resonances created by the port in the speaker enclosure.

[0030] In step 320, the algorithm imposes slew rate limits on the peak frequency and amplitude to prevent the created filters (S and S -1 ) changes too quickly, which can cause audible artifacts. In other words, we do not use the observed peak and amplitude immediately - we maintain operational values ​​for the peak and amplitude that move toward the observed peak and amplitude at no more than some constant rate.

[0031] In step 322, the algorithm creates a model IIR filter S that matches the operational peak. The filter can be generated using a filter design algorithm with two (or more) pairs of poles and zeros, and can be represented by SOS coefficients.

[0032] In step 324, determine the stable inverse S -1 Filters (e.g., as in Figure 1 The inverse is obtained by simply swapping the numerator and denominator terms in the SOS representation, followed by optionally normalizing the zero numerator coefficients to 1.0.

[0033] In step 326, the algorithm verifies the validity of the input data frame by running tests to verify that the various parameters from steps 312-324 are within normal ranges. If any of these tests fail, the frame is marked as "invalid" and no modifications are made to the operational values ​​of peak and amplitude.

[0034] refer to Figure 4, an offline "tuning" algorithm 400 will now be described according to one or more embodiments. In step 410, the algorithm estimates an initial model filter S. In the illustrated embodiment, a broadband signal with a sampling rate of 48 KHz is sent to the loudspeaker. The broadband signal contains energy in the frequency range below Nyquist (e.g., all frequencies below Nyquist) and has enough energy to create an offset on the loudspeaker to a sufficient degree to enable accurate measurement of the electrical impedance of the loudspeaker voice coil, but not so much energy that significant heating or other damage to the voice coil occurs. In one embodiment, the signal includes a series of short bursts of pink noise at approximately -12 dB relative to the maximum RMS voltage drive. The instantaneous current and voltage can be measured by a Class D amplifier with a voltage / current sensing circuit. During the sub-band analysis process, these time domain signals are converted to the frequency domain by an appropriate short-term Fourier transform. In one embodiment, for example, the 48 KHz data is downsampled to 16 KHz and overlapping frames of 320 16 KHz samples are converted to 64 complex frequency domain values ​​of current I(f) and voltage V(f). The calculations from Figure 3 Steps 312 to 322 are performed to estimate the initial filter S.

[0035] In step 412, the scale factor is measured. In one embodiment, a series of short sine wave bursts of increasing amplitude at the frequency of the main peak of S are measured by Figure 1 102 is then sent to the microphone, where the clipping threshold thr 120 has a nominal constant value. The excursion of the microphone 114 is measured by a laser rangefinder or other cone measurement sensor 132. The amplitude of the component of the excursion at the burst frequency is calculated by a matched filter technique (e.g., the rms value of a Hann window sine convolved with the measured excursion waveform). The excursion amplitude will asymptotically reach a maximum value at the nominal threshold thr multiplied by a proportional factor. The proportional factor can then be calculated as the asymptotic maximum value divided by the nominal threshold thr. The proportional factor is then incorporated into the value representing the SOS coefficient of the filter S.

[0036] In step 414, a clipping level threshold thr is established for online use. In order to apply the algorithm disclosed herein in real time, the clipping threshold thr that will provide the highest possible sound level without excessive distortion is determined. To determine thr, a series of sinusoidal bursts are sent to the receiver at multiple frequencies (e.g., from 100 Hz to 8000 Hz in 1 / 12 octave logarithmic intervals) and increasing amplitudes (e.g., from -12 dB FS to 0 dB FS in 3 dB steps). Figure 11. The speaker shown, where the nominal value of thr is initially set too high (e.g., for a typical micro speaker, thr is 1.0 mm, and it is known that a practical safe thr for a typical micro speaker is about 0.5 mm). The harmonic distortion of the excursion signal (measured by a laser or other sensing component 132) is calculated for each burst. The threshold thr is set so that if the excursion had been limited to this value, the harmonic distortion would not exceed a pre-specified limit, such as 10% (corresponding to a 20 dB SDR).

[0037] In step 416, the threshold thr is verified. In one embodiment, a series of full-scale sinusoidal bursts are sent to the speaker (e.g., via Figure 1 ), where the value of thr is calculated in step 414 and the scaling factor is calculated in step 412. The excursion of the speaker is measured by a laser or other sensor 132 and plotted as a function of the burst frequency. In various embodiments, the excursion is limited to a value close to thr at each frequency.

[0038] In some embodiments, it may be desirable to run the real-time algorithm non-adaptively. In this case, the filter S, threshold thr and scale factor can be determined by a tuning algorithm. S is then kept constant for real-time use. This eliminates the adaptive step, simplifies the implementation and reduces computational requirements. The non-adaptive algorithm can be used, for example, to control distortion in a system-on-chip implementation for embedded applications.

[0039] As discussed, the various techniques provided herein may be implemented by one or more systems, which in some embodiments may include one or more subsystems and their associated components. Figure 5 1 shows a block diagram of an example processing system 500 according to an embodiment of the present disclosure. The system 500 can be used to implement any desired combination of the various components, circuits, processing steps, and other operations described herein. Figure 5 Various components are shown in FIG. 1 , but in various embodiments, components may be added and / or omitted as appropriate for different types of devices.

[0040] As shown, the processing system 500 includes an audio processing system 510, which includes a memory 520, a processor 540, and an audio output circuit 550. The processor 540 may be implemented as one or more logic devices, such as a microprocessor, a microcontroller, an application specific integrated circuit (ASIC), a programmable logic device (PLD) (e.g., a field programmable gate array (FPGA), a complex programmable logic device (CPLD), a system on a chip, or other types of programmable devices.

[0041] In some embodiments, the processor 540 executes machine-readable instructions (e.g., software, firmware, or other instructions) stored in the memory 520. In this regard, the memory 520 may include logic that causes the processor 540 to perform any of the various operations, processes, and techniques described herein. In other embodiments, the processor 540 and / or the memory 520 may be replaced and / or supplemented with dedicated hardware components to perform any desired combination of the various techniques described herein.

[0042] The memory 520 may be implemented as a machine-readable medium that stores various machine-readable instructions and data. For example, in some embodiments, the memory 520 may store one or more algorithms as machine-readable instructions that may be read and executed by the processor 540 to perform various techniques described herein, including a subband analysis algorithm 522, a filter estimation algorithm 524, an offline tuning algorithm 526, and / or other executable logic. The memory 520 may also store data used by the audio processing system 510. In some embodiments, the memory 520 may be implemented as a non-volatile memory (e.g., a flash memory, a hard drive, a solid-state drive, or other non-transitory machine-readable medium), a volatile memory, or a combination thereof.

[0043] The system 500 also includes an audio source 570 and one or more speakers 560. In operation, the audio source 570 provides audio data for processing by the audio output circuit and playback by the one or more speakers 560. The system 500 can be used to perform the processes disclosed herein to protect the one or more speakers 560 from excessive excursion. In some embodiments, each of the one or more speakers 560 has a corresponding excursion protection filter disclosed herein. The system 500 can also include other speaker protection techniques, including temperature and power protection of the speakers 560.

[0044] As used herein, the terms "substantially," "about," and similar terms are used as terms of approximation rather than terms of degree, and are intended to account for the inherent deviations of measurements or calculations that will be recognized by one of ordinary skill in the art. As used herein, the terms "use," "using," and "used" may be considered synonymous with the terms "utility," "utilizing," and "utilized," respectively. Additionally, the term "exemplary" is intended to refer to an example or illustration.

[0045] Any suitable hardware, firmware (e.g., application specific integrated circuits), software, or a combination of software, firmware, and / or hardware may be utilized to implement the electronic or electrical devices and / or any other related devices or components according to the embodiments of the present disclosure described herein. For example, the various components of these devices may be formed on one integrated circuit (IC) chip or on separate IC chips. In addition, the various components of these devices may be implemented on a flexible printed circuit film, a tape carrier package (TCP), a printed circuit board (PCB), or formed on a substrate. In addition, the various components of these devices may be processes or threads running on one or more processors in one or more computing devices, which execute computer program instructions and interact with other system components to perform the various functionalities described herein. The computer program instructions are stored in a memory, which may be implemented in a computing device using a standard memory device, such as, for example, a random access memory (RAM). The computer program instructions may also be stored in other non-transitory computer-readable media, such as, for example, a CD-ROM, a flash drive, and the like. In addition, those skilled in the art should recognize that, without departing from the spirit and scope of the exemplary embodiments of the present disclosure, the functionality of various computing devices may be combined or integrated into a single computing device, or the functionality of a particular computing device may be distributed on one or more other computing devices.

[0046] The embodiments described herein are merely examples. Those skilled in the art may recognize various alternative embodiments from those specifically disclosed. Those alternative embodiments are also intended to be within the scope of this disclosure. Therefore, the embodiments are limited only by the appended claims and their equivalents.

Claims

1. A method performed by an audio processing system, comprising: receiving an audio signal; amplifying the audio signal for output via a speaker; receiving current and voltage data associated with the amplified audio signal; estimating an impedance of the speaker based on the received current and voltage data; estimating a DC resistance of a speaker coil driving a cone of the speaker based on the estimated impedance; updating an excursion model representing the excursion of the speaker based at least in part on an algorithm that utilizes movement of a cone of the speaker as a function of the estimated impedance; as well as Amplification of the audio signal is dynamically adjusted based at least in part on the updated excursion model.

2. The method according to claim 1, wherein the updating of the offset model comprises: measuring an excursion of the speaker associated with the output audio signal; as well as The excursion model is adapted to the measured excursions of the loudspeaker.

3. The method according to claim 1, wherein the updating of the offset model comprises: Filter coefficients associated with an infinite impulse response (IIR) filter are estimated based at least in part on the motion of the cone as a function of the estimated DC resistance.

4. The method of claim 3, wherein the dynamic adjustment of the amplification of the audio signal comprises: applying the IIR filter to the audio signal; limiting the output of the IIR filter based on a limiting threshold; as well as An inverse filter is applied to the clipped output of the IIR filter, the inverse filter representing the inverse of the IIR filter.

5. The method of claim 3, wherein estimating the impedance of the speaker based on the received current and voltage data comprises: Converting the current and voltage data into frequency domain sub-bands I(f) and V(f), respectively; as well as The impedance Z(f) of the speaker is estimated based on I(f) and V(f).

6. The method of claim 5, wherein estimating the filter coefficients further comprises: The motion sensitivity function Gmot(f) representing the motion of the cone of the loudspeaker as a function of frequency f is calculated, where:

7. The method of claim 6, wherein the estimating of the filter coefficients further comprises: Estimate two or more peaks of |Gmot(f)|; as well as A model of the IIR filter is created based on the estimated peak value of |Gmot(f)|.

8. The method of claim 7, wherein estimating the two or more peaks of |Gmot(f)| comprises: A slew rate limit is imposed on the peak frequency and amplitude associated with Gmot(f).

9. The method according to claim 3, further comprising: The IIR filter is tuned based on a scaling factor and a clipping level threshold associated with the excursion of the speaker.

10. An audio processing system, comprising: an amplifier configured to receive and amplify an audio signal for output via a speaker; as well as A loudspeaker protection system configured to protect the loudspeaker from excessive excursion by: receiving current and voltage data associated with the amplified audio signal; estimating an impedance of the speaker based on the received current and voltage data; estimating a DC resistance of a speaker coil driving a cone of the speaker based on the estimated impedance; updating an excursion model representing the excursion of the speaker based at least in part on an algorithm that utilizes movement of a cone of the speaker as a function of the estimated impedance; as well as Amplification of the audio signal is dynamically adjusted based at least in part on the updated excursion model.

11. The audio processing system of claim 10, wherein the speaker protection system is configured to update the excursion model by: measuring the excursion of the speaker associated with the outputted audio signal; and The excursion model is adapted to the measured excursions of the loudspeaker.

12. The audio processing system of claim 10, wherein the speaker protection system is configured to update the excursion model by: Filter coefficients associated with an infinite impulse response (IIR) filter are estimated based at least in part on the motion of the cone as a function of the estimated DC resistance.

13. The audio processing system of claim 12, wherein the speaker protection system is configured to dynamically adjust the amplification of the audio signal by: applying the IIR filter to the audio signal; Clipping the output of the IIR filter based on a clipping threshold; and An inverse filter is applied to the clipped output of the IIR filter, the inverse filter representing the inverse of the IIR filter.

14. The audio processing system of claim 12, wherein estimating the impedance of the speaker based on the received current and voltage data comprises: Converting the current and voltage data into frequency domain sub-bands I(f) and V(f), respectively; as well as The impedance Z(f) of the speaker is estimated based on I(f) and V(f).

15. The audio processing system of claim 14, wherein the speaker protection system further estimates the filter coefficients by: The motion sensitivity function Gmot(f) representing the motion of the cone of the loudspeaker as a function of frequency f is calculated, where:

16. The audio processing system of claim 15, wherein the speaker protection system is configured to estimate the filter coefficients by: estimating two or more peaks of |Gmot(f)|; and A model of the IIR filter is created based on the estimated peak value of |Gmot(f)|.

17. The audio processing system of claim 16, wherein the loudspeaker protection system is configured to estimate the two or more peak values ​​of |Gmot(f)| by: A slew rate limit is imposed on the peak frequency and amplitude associated with Gmot(f).

18. The audio processing system of claim 12, wherein the speaker protection system is further configured to: The IIR filter is tuned based on a scaling factor and a clipping level threshold associated with the excursion of the speaker.

19. A speaker protection system, comprising: one or more processors; as well as a memory storing instructions that, when executed by the one or more processors, cause the speaker protection system to: receiving an audio signal; receiving current and voltage data associated with amplification of the audio signal for output via a speaker; receiving current and voltage data associated with the amplified audio signal; estimating an impedance of the speaker based on the received current and voltage data; estimating a DC resistance of a speaker coil driving a cone of the speaker based on the estimated impedance; updating an excursion model representing the excursion of the speaker based at least in part on an algorithm that utilizes movement of a cone of the speaker as a function of the estimated impedance; as well as The amplification of the audio signal is dynamically adjusted based at least in part on an updated excursion model.

20. The loudspeaker protection system of claim 19, wherein the dynamic adjustment of the amplification of the audio signal comprises: estimating filter coefficients associated with an infinite impulse response (IIR) filter based at least in part on the received current and voltage data; applying the IIR filter to an audio signal; Clipping the output of the IIR filter based on a clipping threshold; and An inverse filter is applied to the clipped output of the IIR filter, the inverse filter representing the inverse of the IIR filter.

Images