Generally, active noise cancellation (ANC) can refer to the process of producing a sound from a speaker to attenuate noise (e.g., an unwanted sound) present in an area. To attenuate the noise, the speaker is configured to produce a sound having a similar amplitude, but opposite phase to the noise. Thus, the sound produced by the speaker will combine with and, due to the superposition of waves, reduce the amplitude of the opposite phase noise.
Generally, there are two methods of accomplishing ANC; feedback (typically analog) and feedforward (typically analog or digital). Feedback solutions include an error (e.g., near-field) microphone located near the speaker that senses the sound after the sound produced by the speaker has combined with the noise. The audio information from the error microphone is sent to a controller which then adjusts the sound produced by the speaker based thereon. In a feedforward solution, a reference (e.g., far-field) microphone senses noise before it combines with the sound produced by the speaker. The audio information from the reference microphone is sent to a controller which causes the speaker to produce a sound having a similar amplitude, but opposite phase to the noise sensed by the reference microphone. Feedforward solutions can be either fixed or adaptive, with adaptive solutions being generally more robust than fixed solutions. Various feedback or feedforward ANC solutions have been used in stereo ANC headsets.
This document discusses, among other things, systems and methods for active noise cancellation. One example system includes a digital ANC circuit configured to receive first audio information from a first microphone and to produce an a digital anti-noise signal configured to attenuate noise sensed by the first microphone, an analog ANC circuit configured to receive second audio information from a second microphone and to produce an analog anti-noise signal configured to attenuate noise sensed by the second microphone, and wherein the system is configured to receive an intended audio signal and to provide an output signal for a speaker using the intended audio signal, the analog anti-noise signal, and the digital anti-noise signal.
Example 1 includes a system for providing active noise cancellation (ANC) including a digital ANC circuit configured to receive first audio information from a first microphone and to produce an a digital anti-noise signal configured to attenuate noise sensed by the first microphone, an analog ANC circuit configured to receive second audio information from a second microphone and to produce an analog anti-noise signal configured to attenuate noise sensed by the second microphone, and wherein the system is configured to receive an intended audio signal and to provide an output signal for a speaker using the intended audio signal, the analog anti-noise signal, and the digital anti-noise signal.
In Example 2, the first microphone of Example 1 is optionally configured to sense ambient noise such that the digital ANC circuit includes a feedforward ANC circuit, and the second microphone of Example 1 is optionally configured to sense output from the speaker such that the analog ANC includes a feedback ANC circuit.
In Example 3, the digital ANC circuit of any one or more of Examples 1-2 is optionally implemented on a first integrated circuit (IC) and the analog ANC is implemented on a second IC.
In Example 4, the first IC of any one or more of Examples 1-3 is optionally configured to be coupled to an analog to digital convert (ADC) to convert the first audio information to a digital signal for the digital ANC circuit.
In Example 5, the digital ANC circuit of any one or more of Examples 1-4 is optionally configured to receive the intended audio signal and to provide a composite audio signal using the digital anti-noise signal and the intended audio signal, wherein the system includes a digital to analog converter (DAC) configured to convert the composite audio signal into an analog signal for the analog ANC circuit.
In Example 6, the digital ANC circuit and the DAC of any one or more of Examples 1-5 are optionally implemented with a field programmable gate array (FPGA).
In Example 7, the digital ANC circuit of any one or more of Examples 1-6 is optionally configured to receive first audio information from a plurality of microphones, and wherein the digital ANC circuit includes a first filter coupled to a first subset of the plurality of microphones and a second filter coupled to a second subset of the plurality of microphones.
In Example 8, each of the plurality of microphones of any one or more of Examples 1-7 optionally has a separate filter associated therewith.
In Example 9, at least one of the first filter or the second filter of any one or more of Examples 1-8 optionally includes an adaptive filter.
In Example 10, the digital ANC of any one or more of Examples 1-9 optionally use the second audio information to update a response of the first and second filters.
In Example 11, the digital ANC circuit of any one or more of Examples 1-10 is optionally configured to adjust the first filter and the second filter to provide dynamic beamsteering for a sensing pattern of the first and second microphone.
In Example 12, the speaker of any one or more of Examples 1-11 optionally includes a piezoelectric speaker.
In Example 13, the speaker of any one or more of Examples 1-12 optionally includes a first piezoelectric speaker and a second dynamic speaker.
Example 14 includes a method for providing active noise cancellation (ANC) including receiving first audio information from a first microphone and providing first ANC information using an analog ANC circuit, receiving second audio information from a second microphone and providing second ANC information using a digital ANC circuit, and providing a combined ANC signal using the first and second ANC information.
In Example 15, the subject matter of any one or more of Examples 1-14 optionally includes combining an intended audio signal with the first ANC information and the second ANC information to produce an output signal for a speaker.
In Example 16, the combining of any one or more of Examples 1-15 optionally includes combining the intended audio signal with the second ANC information to form a composite audio signal, and combining the composite audio signal with the first ANC information to produce the output signal.
In Example 17, the first audio information of any one or more of Examples 1-16 is optionally sensed by a first microphone configured to sense output from the speaker, and the second audio information of any one or more of Examples 1-16 is optionally sensed by a second microphone configured to sense ambient noise.
In Example 18, the subject matter of any one or more of Examples 1-17 optionally includes adaptively filtering the second audio information.
In Example 19, the adaptively filtering of any one or more of Examples 1-18 optionally includes updating a filter response based on the first audio information.
In Example 20, the receiving second audio information of any one or more of Examples 1-19 optionally includes receiving second audio information from a plurality of microphones wherein the method includes: filtering a first signal from a third microphone of the plurality of microphones using a first filter, and filtering a second signal from a fourth microphone of the plurality of microphones using a second filter.
In Example 21, the subject matter of any one or more of Examples 1-20 optionally includes adjusting the first filter and the second filter to provide dynamic beamsteering for a sensing pattern of the third and fourth microphone.
Example 22 includes a system for providing active noise cancellation (ANC) including a digital ANC circuit configured to receive first audio information from a first microphone and to produce an anti-noise signal configured to attenuate noise in the first audio information, a summation circuit configured to combine the anti-noise signal from the digital ANC circuit with an intended audio signal to form a composite audio signal, and an analog ANC circuit configured to receive second audio information from a second microphone and the composite audio signal, the analog ANC circuit configured to attenuate noise in the second audio information and to produce an output signal for a speaker based on the second audio information and the composite audio signal.
In Example 23, the digital ANC circuit of any one or more of Examples 1-22 is optionally configured to receive third audio information from a third microphone, the first and third microphones configured to sense ambient noise, to adaptively filter the first audio information with a first filter, and to adaptively filter the third audio information with a second filter, wherein the anti-noise signal is configured to attenuate noise in the third audio information.
In Example 24, the digital ANC circuit of any one or more of Examples 1-23 is optionally configured to dynamically adjust the first filter and the second filter to provide dynamic beamsteering for a sensing pattern of the first and second microphones.
In Example 25, a system or method can include, or can optionally be combined with any portion or combination of any portions of any one or more of Examples 1-24 to include, means for performing any one or more of the functions of Examples 1-24, or a machine-readable medium including instructions that, when performed by a machine, cause the machine to perform any one or more of the functions of Examples 1-24.
This overview is intended to provide an overview of subject matter of the present patent application. It is not intended to provide an exclusive or exhaustive explanation of the invention. The detailed description is included to provide further information about the present patent application.
The present inventors have recognized, among other things, that both feedback and feedforward active noise cancellation (ANC) can be combined in a single solution. In an example, the combined feedback and feedforward solution can be specifically designed for or used in a mobile phone application.
The first mobile phone 102 can also include a plurality of microphones 110, 112 for sensing sounds and producing audio information regarding the sensed sounds. The microphones 110, 112 can include directional microphones or omni-directional microphones. In an example, the first mobile phone 102 includes one or more near-field microphones 110 configured to sense output from the short range speaker 106. Accordingly, in an example, the near-field microphone 110 is located near the short range speaker 106. Typically, during use of the short range speaker 106, the first mobile phone 102 will be pressed up against the user or very close to the user such that the short range speaker 106 produces sound in a semi-enclosed area. In an example, the near-field microphone 110 is located within the semi-enclosed area in order to sense the combination of the sound produced by the speaker and noise as heard by the ear of the user. The first mobile phone 102 can also include one or more far-field microphones 112 configured to sense ambient noise (e.g., far-field sounds). Accordingly, in an example, the far-field microphones 112 can be located away from the speakers 106, 108 in order to sense the ambient noise while reducing the amount of sound sensed from the speakers 106, 108.
The ANC system combines the output from the digital ANC circuit 202, the output from the analog ANC circuit 204, and an intended audio signal to produce an output for the speaker 106. The intended audio signal includes a signal for which the user is intended to hear (e.g., audio information received from the second mobile phone 104). The intended audio signal is included with an anti-noise (e.g., intended to attenuate noise) signal from the digital ANC circuit 202 and an anti-noise signal from the analog ANC circuit 204 to produce the output for the speaker 106.
The digital ANC circuit 202 receives audio information from the far-field speakers 112 to perform feedforward ANC. In an example, the audio information from the speakers 112 is converted from analog to digital with an analog to digital converter (ADC) 206. The digital audio information from the ADC 206 is filtered with a filter 208.
In an example, the digital anti-noise signal is combined with the intended audio signal using a summation circuit 210. In an example, the intended audio signal comprises a digital signal and, as such, is digitally combined with the digital anti-noise signal in the summation circuit 210. In an example, the digital anti-noise signal comprises an accurate representation of the noise (e.g., all sound minus the sound from speakers 106, 108) sensed by the far-field microphones 112 and, as such, the digital anti-noise signal is subtracted from (e.g., inverted and combined with) the intended audio signal. The combination of the digital anti-noise signal and the intended audio signal is referred to herein as a composite audio signal.
In an example, the composite audio signal is converted to an analog form with a digital to analog converted (DAC) 212. The analog composite audio signal is sent to the analog ANC circuit 204. Along with the analog composite audio signal, the analog ANC circuit 204 receives audio information from the near-field speaker 110. The analog ANC circuit 204 forms an analog anti-noise signal using a feedback loop with one or more amplifiers 214. The analog anti-noise signal is configured to produce a sound from the speaker 106 to attenuate the noise sensed by the near-field microphone 112. The analog ANC circuit 204 combines the analog anti-noise signal with the composite audio signal received from the digital ANC circuit 202 to produce an output signal for the speaker 106. Accordingly, the output signal configures the speaker 106 to produce sound corresponding to the intended audio signal as well as sound intended to attenuate noise based on a feedforward method (the digital ANC circuit 202) and a feedback method (the analog ANC circuit 204) of ANC.
In an example, the analog ANC circuit 202 can be less expensive to implement than the combination of the digital ANC circuit 202 and the analog ANC circuit 204, but the combination of the digital ANC circuit 202 and the analog ANC circuit 204 can produce better overall noise cancellation. Accordingly, in an example, the ANC system shown in
Additionally, in an example, the one or more ADCs 206 for converting the audio information from the far-field speakers 112 into digital form are separate from the second IC which includes the digital ANC circuit 202. For example, when the microphones 112 comprise digital microphones, the ADCs 206 can be integrated into a separate IC from the second IC. Accordingly, the ADCs 206 can be physically located away from the amplifier 214 for the analog ANC circuit 204. In another example, the ADCs 206 are integrated on the second IC when the microphones 112 comprise analog microphones.
In an example, the ADCs 206 can comprise Sigma-Delta ADCs. Using Sigma-Delta ADCs can reduce system latency and simplify the adaptive filters 208. For example, when using Sigma-Delta ADCs, the adaptive filters 208 can comprise 1-bit multipliers (PDM output) instead of 24 bit. Use of 1-bit multipliers, however, can increase the number of taps for the filters 208. In examples where speaker 106 comprises a dynamic speaker, the latency of speaker 106 can dominate overall system latency. Accordingly, in an example, the speaker 106 can include a piezoelectric speaker to reduce the latency thereof. In yet another example, the speaker 106 can include a piezoelectric speaker and a dynamic speaker (hybrid solution) to reduce latency and provide a good sound response.
In an example, additional filtering for each microphone 112 individually or in combination can be provided after the adaptive filters 208 (e.g., to remove out of band (OOB) noise).
In an example, the digital ANC circuit 202 is implemented with a field programmable gate array (FPGA). Finally, as mentioned above, the digital ANC circuit 202 can use the audio information from the near-field microphone 110 combined with a copy of the intended audio signal to update a response of the filters 208 using a filter response controller 216. More detail regarding the filter response controller 216 is provided below.
In an example, the filters 208 for the digital ANC circuit 202 include adaptive filters that adjust to the noise over time.
As discussed above, the ANC system of
Combining the digital ANC circuit 202 and the analog ANC circuit 204, however, can negate the weaknesses of both approaches. For example, the analog ANC circuit 204 implementing a feedback method of ANC can provide decent noise attenuation regardless of the direction of the noise, but the noise attenuation is not as good as the noise attenuation provided by the digital ANC circuit 202 in the very good direction. Accordingly, the combination of the digital ANC circuit 202 and the analog ANC circuit 204 can provide good noise attenuation in most directions and fantastic noise attenuation in directions aligning with a far-field microphone 112 as discussed above. Additionally, in certain examples, the digital ANC circuit 202 can adapt out the interference of the analog ANC circuit 204, and the analog ANC circuit 204 can alleviate performance requirements from digital ANC circuit 202, allowing for more digital for the FPGA.
In an example, the combined feedback and feedforward ANC solution can be used with speakerphone using an extra “cone of silence” microphone (e.g., 3 usable position locations). Further, the beamsteering three microphones can produce smaller lobes to better select a speaker. In an example, the combined feedback and feedforward ANC solution can require a threshold detection to avoid dialing into undesired sources during speaking silence, or can suppress farfield noise using a notch response when a lobe is not pointing to a source.
In an example, the audio information from the near-field microphone 110 is provided to the filter response controller 216 to update the response of the filters 208. In an example, the response of the filters 208 is updated using a least mean squared (LMS) method that updates a sign of the response.
In certain examples, without using a decimation filter, the OOB noise from the microphone ADCs can be monitored. Because the noise is OOB, it is not audible, but it can increase power dissipation in the output amp and speaker. A good target is 1% of full scale for this residual energy. In an example, a traditional decimation filter can be overkill, as the adaptive filter can remove a good deal of this energy, in certain examples, virtually all of it right after a good initialization. But as the filter adapts, it can become less effective in this area due to the e(n) filter making it ignore the OOB energy. In an example, a simple filter can be added after the summing of W1(n) and W2(n), in certain examples, with less delay than a decimator. Further, the DAC DSM and DAC filtering can add significant filtering, but the DAC can re-add residual OOB noise, and in certain examples, the DAC DSM can be overloaded by OOB energy and increase the requirements on the DAC DSM filter.
In an example, an LMS algorithm can accommodate speaker latency using a parameter (est_speaker_delay) to accommodate a large range of speaker latencies. In certain examples, an e(n) filter can equalize out the speaker group delay variation, or a simple time delay can be used.
In an example, the LMS algorithm can include the sign-error-LMS algorithm, using the sign of the error signal instead the full error signal (e.g., to simplify computation). In an example, if the data is in a 1-bit format, the sign-data can be redundant. In certain examples, variations on the sign-LMS algorithm can increase residual adaptation energy, which can be compensated by shrinking mu and increasing adaptation time (e.g., saving die area and power).
In an example, the mu parameter can be updated either by I2C or by looking at the e(n) residual energy. In other examples, both (plus a hybrid version) can be supported.
In certain examples, a field programmable gate array (FPGA) can be configured to use a feedback ADC or work without one because of the simple sign-error algorithm. In an example, the intended audio signal can be subtracted on the board to work without the feedback ADC. Further, in an example, an ANC solution can use AGC circuitry instead of trim pots.
In other examples, additional adaptive filters (e.g., beyond 2) can be added in the code (e.g., by replicating the code from one to two filters). The LMS algorithm can provide the beam-steering work by adapting to provide a minimum mean square error (MMSE). In an example, a high end solution can use that algorithm.
Further, in certain examples, different W1(n) and W2(n) initializations can be used. There are sufficient taps to produce a very sharp filter, although the sharp filter can have a large natural latency. If the taps were utilized in place of decimation filters, the very sharp filter can loose response during times of silence.
In certain examples, adaptive algorithms can be turned off during times of silence (e.g., by monitoring the energy on the outputs of W1(n) and W2(n) and to zero out mu when there is low signal energy) to ensure that the LMS algorithm does not drift or adapt to undesired parameters during the times of silence.
Although the ANC systems and methods are described above with respect to a mobile phone, in other examples, the ANC systems and methods can be used with other electronic devices. For example, the ANC systems and methods can be used with headphones, car speakers, home speakers, non-mobile phones, speakerphones, and others. Additionally, the ANC systems and methods described above can be used in conjunction with other ANC systems and methods such as echo cancellation and others.
The above detailed description includes references to the accompanying drawings, which form a part of the detailed description. The drawings show, by way of illustration, specific embodiments in which the invention can be practiced. These embodiments are also referred to herein as “examples.” All publications, patents, and patent documents referred to in this document are incorporated by reference herein in their entirety, as though individually incorporated by reference. In the event of inconsistent usages between this document and those documents so incorporated by reference, the usage in the incorporated reference(s) should be considered supplementary to that of this document; for irreconcilable inconsistencies, the usage in this document controls.
In this document, the terms “a” or “an” are used, as is common in patent documents, to include one or more than one, independent of any other instances or usages of “at least one” or “one or more.” In this document, the term “or” is used to refer to a nonexclusive or, such that “A or B” includes “A but not B,” “B but not A,” and “A and B,” unless otherwise indicated. In the appended claims, the terms “including” and “in which” are used as the plain-English equivalents of the respective terms “comprising” and “wherein.” Also, in the following claims, the terms “including” and “comprising” are open-ended, that is, a system, device, article, or process that includes elements in addition to those listed after such a term in a claim are still deemed to fall within the scope of that claim. Moreover, in the following claims, the terms “first,” “second,” and “third,” etc. are used merely as labels, and are not intended to impose numerical requirements on their objects.
The above description is intended to be illustrative, and not restrictive. In other examples, the above-described examples (or one or more aspects thereof) may be used in combination with each other. Other embodiments can be used, such as by one of ordinary skill in the art upon reviewing the above description. The Abstract is provided to comply with 37 C.F.R. §1.72(b), to allow the reader to quickly ascertain the nature of the technical disclosure. It is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims. Also, in the above Detailed Description, various features may be grouped together to streamline the disclosure. This should not be interpreted as intending that an unclaimed disclosed feature is essential to any claim. Rather, inventive subject matter may lie in less than all features of a particular disclosed embodiment. Thus, the following claims are hereby incorporated into the Detailed Description, with each claim standing on its own as a separate embodiment. The scope of the invention should be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.
This patent application claims the benefit of priority, under 35 U.S.C. Section 119(e), to Delano, Cary, U.S. Provisional Patent Application Ser. No. 61/255,535 entitled “ACTIVE NOISE CANCELLATION” filed on Oct. 28, 2009 (Attorney Docket No. 2921.034PRV), which is hereby incorporated by reference herein in its entirety.
Number | Date | Country | |
---|---|---|---|
61255535 | Oct 2009 | US |