The present disclosure relates generally to digital sensors like microphone assemblies, electrical circuits therefor, and methods of operating the same.
Power consumption and size are paramount considerations in devices comprising a microphone as part of a voice interface, particularly in battery-operated devices but also in devices connected to a power grid. Such devices include smart speakers, hearing aids and other wearable devices, portable communication and computing devices such as smartphones, mobile phones and tablets among other devices.
Digital microphones and other sensors often comprise a multibit current-mode delta-sigma analog-to-digital converter (ADC) for which a digital-to-analog converter (DAC) is an essential building-block. The input current accommodated by a current-mode ADC is limited by an amount of feedback current supplied to the ADC input by the DAC. High dynamic range generally requires that the DAC provide a high feedback current to the ADC input. Thus the DAC can be a significant source of power consumption.
The objects, features and advantages of the present disclosure will become more fully apparent to those of ordinary skill in the art upon careful consideration of the following Detailed Description and the appended claims in conjunction with the accompanying drawings described below.
Those of ordinary skill in the art will appreciate that elements in the drawings are illustrated for simplicity and clarity and may thus be disproportionate or omit elements unnecessary for conveying how to make and use the subject matter disclosed, that acts or steps described or depicted in a particular order may be performed in a different order or sequence absent a contrary indication, and that the terms and expressions herein have the meaning ordinarily accorded to such terms and expressions with respect to their corresponding areas of inquiry and study except where other meanings have otherwise been defined herein.
The present disclosure relates to digital sensor assemblies, electrical circuits therefor, and methods of operating the same. The digital sensor assembly generally comprises a sensor (i.e., transduction element) electrically coupled to a sensor signal processing circuit. The sensor can be a capacitive, piezo or optical electro-acoustic transduction element among other known and future sensors configured to convert a sensed environmental condition into a corresponding electrical signal. Such sensors can be fabricated from microelectromechanical systems (MEMS) or some other known or future technology. In one implementation, the sensor is an acoustic sensor and the electrical circuit is a microphone signal processing circuit. In other implementations, the sensor is a non-acoustic sensor and the electrical circuit is a non-acoustic sensor signal processing circuit. Such other sensors include pressure, temperature, gas, and humidity sensors, among others.
In some implementations, the digital sensor assembly comprises a housing including an external-device interface for integration with a host device. The interface can be configured for surface-mounting or for through-hole mounting, among other known or future mounting techniques. The sensor and the electrical circuit can be disposed in the housing and the electrical circuit can be electrically coupled to contacts on the external-device interface. In microphone and other sensor assemblies, the housing can comprise a port by or through which the sensor can detect environmental conditions external to the housing. In a more particular implementation, the housing comprises a cover portion (e.g., a lid fabricated from metal or other material having shielding properties suitable for the particular use case) mounted on a base portion including the external-device interface. Any required port can be located on the base, sidewall or cover portion and the sensor can be mounted over or proximate to the sound port to sense sound or other conditions of interest.
In
As suggested, the electrical circuit described herein can be implemented as a microphone signal processing circuit for use with an acoustic sensor or as some other sensor signal processing circuit. The electrical circuit generally comprises a multibit delta-sigma analog-to-digital converter (ADC) including a digital-to-analog converter (DAC) having a plurality of N selectably enabled current elements in a feedback path of the ADC. Each current element comprises a current source, a current sink or both current source and current sink. The ADC is configured to generate a digital signal based on the electrical signal received from the sensor. A control circuit coupled to the plurality of current elements is configured to adaptively activate one or more segments of current elements for a sampling period, wherein each active segment comprises less than N current elements, and each current element in the active segment is enabled. Some, none or all of the enabled current elements in an active segment contribute to the DAC feedback signal during the sampling period. Current elements not in an active segment are disabled. Power consumption can thus be reduced without compromising the dynamic range of the electrical circuit.
In
In
In
The number of segments active during a particular sampling period can be based on a magnitude or other characteristics of a digital signal input to the DAC in a prior sampling period. Correlation between the magnitude of the digital signal and the number of enabled current elements required for a particular sampling period. Each active segment is allocated a fixed number of enabled current elements, less than the total number of current elements N. Thus the characteristic of the digital signal can be predictive of the number of active segments required for a particular sampling period. Current elements not allocated to an active segment are disabled. Each enabled current element in an active segment either contributes to the feedback signal or does not contribute to the feedback signal of a particular sampling period depending on the characteristic of the digital signal input to the DAC. In one implementation, for each sampling period, the number of enabled current elements is a multiple of the number of current elements in each active segment and the number of active segments. For example, 10 current elements would be enabled for 2 active segments each having 5 current elements.
In
In
An overflow event occurs when a number of enabled current elements are insufficient to provide the feedback current required during a sampling period. An overflow event may result from a possible, but unlikely, prediction error or other causes. In situations where an overflow event occurs, the control circuit is configured to increase a number of segments activated for a subsequent sampling period in response to the overflow event and to modify the DAC feedback signal of the subsequent sampling period by a deviation value associated with the overflow event. In
In
In
In the non-overlapped sliding allocation scheme described above, the current elements are allocated to active segments sequentially and rotationally. In other allocation schemes however the current elements are allocated rotationally, but not necessarily sequentially. For example, a non-sequential allocation scheme could allocate every other current element to the active segment. The rotational allocation can be performed algorithmically using an index [i] associated with the current elements, DAC[i]. In one implementation, a current element contributing to the feedback signal in a prior sampling period is not reallocated to another segment until all DAC current elements have been allocated to an active segment and have contributed to a feedback signal. A current element contributing to the feedback signal in a prior sampling period is not reallocated to a segment active until after all or substantially all N current elements have been allocated to an active segment and contributed to the feedback signal. As suggested, the allocation and enablement of current elements to active segments can be implemented by hardware, firmware or software integrated in or with the sensor signal processing circuit.
In implementations where latency is an issue, the control circuit can be configured to generate and buffer, during a current sampling period, an enable signal that activates one or more segments for a subsequent sampling period. In this case, signals for enabling current elements allocated to one or more segments of the current sampling period would have been generated and buffered during a prior sampling period temporally adjacent to the current sampling period. In
In implementations where mismatch shaping is desired, for example to compensate for DAC error caused by variation among the current elements, the circuit further comprises an executable mismatch shaping algorithm, wherein execution of the mismatch shaping algorithm reduces signal distortion. In one implementation, the mismatch shaping algorithm is a data weighted averaging algorithm. In other implementations other algorithms can be used for this purpose. The efficacy of the mismatch shaping algorithm is largely unaffected by implementation of the current allocation algorithms described herein. As noted herein, overflow events can be compensated with little or no adverse effect on audio quality by oversampling and appropriate noise shaping. Mismatch shaping can be implemented by hardware, firmware or software integrated in or with the sensor signal processing circuit.
In a first implementation, a sensor signal processing circuit comprises a delta-sigma analog-to-digital converter (ADC) configured to generate a digital output signal based on an electrical signal received from a sensor, and a control circuit controllably coupled to N current elements in a digital-to-analog (DAC) feedback path of the ADC, the control circuit configured to adaptively activate one or more segments of current elements for a current sampling period based on a digital signal input to the DAC, wherein each active segment comprises less than N current elements, each current element in the one or more active segments is enabled and either contributes to a feedback signal of the DAC or does not contribute to the feedback signal, and current elements not in the active segment are disabled.
In a variation of the first implementation, current elements contributing to the feedback signal in a prior sampling period are not reallocated to a segment active during a subsequent sampling period until after all or substantially all N current elements have contributed to the feedback signal.
The sensor signal processing circuit of the first implementation and its variations can also comprise a buffer, wherein the control circuit is configured to generate and store in the buffer, during a prior sampling period, an enable signal for enabling current elements allocated to the one or more segments active during the current sampling period.
The control circuit of the first implementation and its variations can be configured to increase a number of segments active during the current sampling period in response to an overflow condition occurring in a prior sampling period and to modify the feedback signal of the current sampling period by a deviation value associated with the overflow condition.
The control circuit of the first implementation at its variations can be configured to allocate current elements to active segments on a rotational basis, wherein current elements contributing to the feedback signal in a prior sampling period are not reallocated to a segment active during the current sampling period until after all N current elements have been allocated to an active segment and contributed to the feedback signal.
The control circuit of the first implementation and its variations can be configured to reduce distortion attributable to the DAC by performing mismatch shaping.
The sensor signal processing circuit of the first implementation and its variations can be combined with a sensor, both of which are disposed in a housing comprising an external-device interface, wherein the ADC is electrically coupled to the sensor and to contacts of the external device interface.
The sensor signal processing circuit of the first implementation and its variations can be combined with an acoustic sensor configured to convert sound into an electrical signal and disposed in a housing, wherein the sensor is acoustically coupled to a sound port of the housing and the ADC is electrically coupled to the acoustic sensor and to contacts of an external device interface of the housing.
The sensor signal processing circuit of the first implementation and its variations and combinations, wherein the control circuit is configured to allocate current elements to one or more segments active during corresponding sampling periods on a non-overlapped sliding basis, wherein current elements contributing to the feedback signal in the prior sampling period are not reallocated to the one or more segments active during the current sampling period until after all N current elements have contributed to the feedback signal.
The sensor signal processing circuit of the first implementation and its variations and combinations, wherein the acoustic sensor is a microelectromechanical systems (MEMS) sensor, and the external-device interface is a surface-mount interface disposed on a base portion of the housing.
While the disclosure and what is presently considered to be the best mode thereof has been described in a manner that establishes possession by the inventor and that enables those of ordinary skill in the art to make and use the same, it will be understood and appreciated that there are many equivalents to the embodiments disclosed herein and that myriad modifications and variations may be made thereto without departing from the scope and spirit of the invention, which are to be limited not by the exemplary embodiments but by the appended claims and their equivalents.
Number | Name | Date | Kind |
---|---|---|---|
7190038 | Dehe et al. | Mar 2007 | B2 |
7473572 | Dehe et al. | Jan 2009 | B2 |
7781249 | Laming et al. | Aug 2010 | B2 |
7795695 | Weigold et al. | Sep 2010 | B2 |
7825484 | Martin et al. | Nov 2010 | B2 |
7829961 | Hsiao | Nov 2010 | B2 |
7856804 | Laming et al. | Dec 2010 | B2 |
7903831 | Song | Mar 2011 | B2 |
20050207605 | Dehe et al. | Sep 2005 | A1 |
20070278501 | Macpherson et al. | Dec 2007 | A1 |
20080175425 | Roberts et al. | Jul 2008 | A1 |
20080267431 | Leidl et al. | Oct 2008 | A1 |
20080279407 | Pahl | Nov 2008 | A1 |
20080283942 | Huang et al. | Nov 2008 | A1 |
20090001553 | Pahl et al. | Jan 2009 | A1 |
20090180655 | Tien et al. | Jul 2009 | A1 |
20100046780 | Song | Feb 2010 | A1 |
20100052082 | Lee et al. | Mar 2010 | A1 |
20100128914 | Khenkin | May 2010 | A1 |
20100183181 | Wang | Jul 2010 | A1 |
20100246877 | Wang et al. | Sep 2010 | A1 |
20100290644 | Wu et al. | Nov 2010 | A1 |
20100322443 | Wu et al. | Dec 2010 | A1 |
20100322451 | Wu et al. | Dec 2010 | A1 |
20110013787 | Chang | Jan 2011 | A1 |
20110075875 | Wu et al. | Mar 2011 | A1 |
20190387326 | Hansen et al. | Dec 2019 | A1 |
20200010315 | Tingleff et al. | Jan 2020 | A1 |
Entry |
---|
U.S. Appl. No. 16/874,503, filed May 14, 2010, Pederson, Michael. |