WO2021087728A1

WO2021087728A1 - Differential directional sensor system

Info

Publication number: WO2021087728A1
Application number: PCT/CN2019/115596
Authority: WO
Inventors: Weilong HUANG; Jinwei Feng
Original assignee: Alibaba Group Holding Limited
Priority date: 2019-11-05
Filing date: 2019-11-05
Publication date: 2021-05-14
Also published as: CN114586097A

Abstract

Input signals received by a plurality of directional sensors that are arranged in a particular configuration are obtained. The input signals may then be transformed from a time domain to a frequency domain to obtain respective frequency bands. In implementations, a set of predetermined weights may be applied to the respective frequency bands in the frequency domain to obtain a combined frequency band, which may then be transformed from the frequency domain to the time domain to obtain an output signal. The particular configuration may be two-dimensional or three-dimensional, and may be formed by a number of circles having a common center, or a number of groups of circles having different centers with each group of circles having a same center.

Description

Differential Directional Sensor System

BACKGROUND

With the advent of signal sensing and processing technologies, omni-directional sensors and omni-directional sensor arrays (i.e., arrays of omni-directional sensors) have been developed and applied in various technological fields. An omni-directional sensor or an omni-directional sensor array is configured to receive and detect signals from all directions, and thus does not need to be mechanically rotated to point to different directions on a regular basis. Examples of an omni-directional sensor may include an omni-directional microphone, an omni-directional camera, an omni-directional radio antenna, etc.

However, due to such omni-directional property, existing omni-directional sensors and omni-directional sensor arrays suffer quite a number of intrinsic problems. An omni-directional microphone/omni-directional microphone array may be used as an example. Since an omni-directional sensor/omni-directional sensor array can pick up signals from all directions, unwanted signals may also be received by the omni-directional sensor/omni-directional sensor array. Although algorithms have been proposed to filter these unwanted signals, existing omni-directional sensor arrays still suffer certain problems. For example, existing omni-directional microphone arrays suffer a low white noise gain (WNG) at low frequencies, and a decreasing directivity index (a measure of directional characteristic) at high frequencies. Moreover, array performance of existing omni- directional sensor arrays is not sufficient for some specific metric, such as an array gain.

SUMMARY

This summary introduces simplified concepts of a differential directional sensor system, which will be further described below in the Detailed Description. This summary is not intended to identify essential features of the claimed subject matter, nor is it intended for use in limiting the scope of the claimed subject matter.

This application describes example implementations of a differential directional sensor system. In implementations, input signals received by a plurality of directional sensors that are arranged in a particular configuration are obtained. The input signals may then be transformed from a time domain to a frequency domain to obtain respective frequency bands. In implementations, a set of predetermined weights may be applied to the respective frequency bands in the frequency domain to obtain a combined frequency band, which may then be transformed from the frequency domain to the time domain to obtain an output signal.

BRIEF DESCRIPTION OF THE DRAWINGS

The detailed description is set forth with reference to the accompanying figures. In the figures, the left-most digit (s) of a reference number identifies the figure in which the reference number first appears. The use of the same reference numbers in different figures indicates similar or identical items.

FIG. 1 illustrates an example environment in which a differential directional sensor system may be used.

FIG. 2 illustrates an example differential directional sensor system in more detail.

FIG. 3 illustrates an example method of signal processing using the example differential directional sensor system.

FIGS. 4A-4F illustrate example geometric configurations of directional sensors.

FIG. 5 illustrates an example circular configuration of directional microphones.

FIG. 6A-6F illustrate comparisons of beampatterns for two different designs between CDDMA and circular differential microphone array (CDMA) that uses omni-directional microphones.

FIGS. 7A and 7B illustrate comparisons between CDDMA and CDMA in terms of WNG and DI.

DETAILED DESCRIPTION

Overview

As noted above, existing omni-directional sensors and omni-directional sensor arrays, though allowing signal detection at all directions, suffer a number of intrinsic problems that could severely affect the usefulness of the omni-directional sensors and omni-directional sensor arrays, and thus limit practical applications of these omni-directional sensors and omni-directional sensor arrays.

This disclosure describes an example differential directional sensor system. The differential directional sensor system may include a plurality of directional (e.g., unidirectional) sensors arranged in a particular geometric configuration. The particular geometric configuration may be two-dimensional or three-dimensional, depending on a corresponding application or use, for example, of the differential directional sensor system. In implementations, the geometric configuration may be formed by, for example, a number of circles having a common center, or a number of groups of circles having different centers with each group of circles having a same center.

In implementations, the differential directional sensor system may further include one or more processors that is configured to control, coordinate, and process signals received by the plurality of directional sensors to produce an output signal.

In implementations, the differential directional sensor system may obtain input signals received by the plurality of directional sensors, and transform the input signals from a time domain to a frequency domain to obtain respective frequency bands. In implementations, the differential directional sensor system may apply a set of predetermined weights to the respective frequency bands in the frequency domain to obtain a combined frequency band, and transform the combined frequency band from the frequency domain to the time domain to obtain an output signal.

In implementations, the differential directional sensor system may store a collection of sets of predetermined weights, or may have an access to the collection of sets of predetermined weights which is stored in a peripheral or remote device. Each set of predetermined weights may cause the differential directional sensor system to preferentially weigh or emphasize signals received by the plurality of directional sensors from a certain preferred or desired direction with respect to a particular point (e.g., a common center) in the particular geometric configuration. In this case, the differential directional sensor system may select the set of predetermined weights from the collection of sets of predetermined weights, and apply this set of predetermined weights to the respective frequency bands to obtain the combined frequency band. In other words, by selecting a different set of predetermined weights (e.g., a certain set of predetermined weights for a certain preferred or desired direction) , the differential directional sensor system may electronically adjust the plurality of directional sensors to receive signal from the preferred or desired direction without the need of mechanically moving any physical parts of the plurality of directional sensors and/or the differential directional sensor system.

In implementations, the differential directional sensor system may estimate or determine a direction from which desired or certain signals originate or come. By way of example and not limitation, the differential directional sensor system may determine respective signal strengths of the desired or certain signals received by the plurality of directional sensors. Based on directional properties of the directional sensors and the geometric configuration of the directional sensors, the differential directional sensor system may estimate or determine a direction from which the desired or certain signals come or originate by, for example, weighted averaging directions of such one or more directional sensors based on the detected signal strengths of the desired or certain signals.

Furthermore, given the use of the directional sensors, the differential directional sensor system can easily filter unwanted signals from directions other than signals coming from a desired direction using the directional properties of the directional sensors.

In implementations, functions described herein to be performed by the differential directional sensor system may be performed by multiple separate units or services. For example, an acquisition service may obtain input signals received by the plurality of directional sensors, while a transformation service may the input signals from a time domain to a frequency domain to obtain respective frequency bands. A combination service may apply a set of predetermined weights to the respective frequency bands in the frequency domain to obtain a combined frequency band, and the transformation service may transform the combined frequency band from the frequency domain to the time domain to obtain an output signal. In implementations, a selection service may select the set of predetermined weights from the collection of sets of predetermined weights, and provide this set of predetermined weights to the combination service for applying to the respective frequency bands to obtain the combined frequency band.

Moreover, although in the examples described herein, the differential directional sensor system may be implemented as a combination of software and hardware installed in a single device, in other examples, the differential directional sensor system may be implemented and distributed in multiple devices or as services provided in one or more computing devices over a network and/or in a cloud computing architecture.

The application describes multiple and varied embodiments and implementations. The following section describes an example framework that is suitable for practicing various implementations. Next, the application describes example systems, devices, and processes for implementing a differential directional sensor system.

Example Environment

FIG. 1 illustrates an example environment 100 usable to implement a differential directional sensor system. The environment 100 may include a differential directional sensor system 102. In this example, the differential directional sensor system 102 is described to exist as an individual entity or device. In some instances, the differential directional sensor system 102 may be included in a computing device, such as a client device 104. In other instances, the differential directional sensor system 102 may be included in one or more servers, such as one or more servers 106 in a cloud. For example, some or all of the functions of the differential directional sensor system 102 may be included in or provided by the client device 104, and/or the one or more servers 106, which are connected and communicated via a network 108.

In implementations, the environment 100 may further include one or more groups of directional sensors 110, with corresponding directional sensors in each group being arranged in a respective geometric configuration. In implementations, some or all of the one or more groups of directional sensors 110 may be peripheral to the differential directional sensor system 102, and communicate data with the differential directional sensor system 102 via a local connection and/or a short-range communication, such as a cable/wire, Bluetooth, Infrared, WiFi, etc. In implementations, some or all of the one or more groups of directional sensors 110 may be remote to the differential directional sensor system 102, and communicate data with the differential directional sensor system 102 via a network, such as the network 108.

In implementations, the client device 104 may be implemented as any of a variety of computing devices including, but not limited to, a desktop computer, a notebook or portable computer, a handheld device, a netbook, an Internet appliance, a tablet or slate computer, a mobile device (e.g., a mobile phone, a personal digital assistant, a smart phone, etc. ) , a server computer, etc., or a combination thereof.

The network 108 may be a wireless or a wired network, or a combination thereof. The network 108 may be a collection of individual networks interconnected with each other and functioning as a single large network (e.g., the Internet or an intranet) . Examples of such individual networks include, but are not limited to, telephone networks, cable networks, Local Area Networks (LANs) , Wide Area Networks (WANs) , and Metropolitan Area Networks (MANs) . Further, the individual networks may be wireless or wired networks, or a combination thereof. Wired networks may include an electrical carrier connection (such a communication cable, etc. ) and/or an optical carrier or connection (such as an optical fiber connection, etc. ) . Wireless networks may include, for example, a WiFi network, other radio frequency networks (e.g.,

Zigbee, etc. ) , etc.

In implementations, the differential directional sensor system 102 may obtain input signals from a certain group of directional sensors in the one or more groups of directional sensors 110, and transform the input signals from a time domain to a frequency domain to obtain respective frequency bands. The differential directional sensor system 102 may then apply a set of predetermined weights to the respective frequency bands in the frequency domain to obtain a combined frequency band, and transform the combined frequency band from the frequency domain to the time domain to obtain an output signal.

Example Differential Directional Sensor system

FIG. 2 illustrates the differential directional sensor system 102 in more detail. In implementations, the differential directional sensor system 102 may include, but is not limited to, one or more processors 202, a memory 204, and program data 206. In implementations, the differential directional sensor system 102 may further include an input/output (I/O) interface 208, and/or a network interface 210. In implementations, some of the functions of the differential directional sensor system 102 may be implemented using hardware, for example, an ASIC (i.e., Application-Specific Integrated Circuit) , a FPGA (i.e., Field-Programmable Gate Array) , and/or other hardware.

In implementations, the differential directional sensor system 102 may further include a plurality of directional (e.g., unidirectional) sensors 212 that are arranged in a particular geometric configuration. Although in this example the differential directional sensor system 102 is described to include a plurality of directional sensors, in some instances, the plurality of directional (e.g., unidirectional) sensors 212 may not be included in the differential directional sensor system 102, and may be accessible to the differential directional sensor system 102 via a network (such as the network 108) , or provided to the differential directional sensor system 102 as auxiliary or peripheral devices or components. In some instances, the differential directional sensor system 102 may include or be associated with (e.g., peripheral to or remotely connected to) a plurality of groups of directional sensors, with corresponding directional sensors in each group being arranged in a respective geometric configuration. Types of the plurality of groups of directional sensors may or may not be the same. In implementations, depending on a type of signal to be detected or sensed, the directional sensors may include, but are not limited to, directional microphones, directional light sensors, directional antennas (such as directional radio antennas) , directional satellite dishes or antennas, etc.

In implementations, the one or more processors 202 may be configured to execute instructions that are stored in the memory 204, and/or received from the input/output interface 208, and/or the network interface 210. In implementations, the one or more processors 202 may be implemented as one or more hardware processors including, for example, a microprocessor, an application-specific instruction-set processor, a physics processing unit (PPU) , a central processing unit (CPU) , a graphics processing unit, a digital signal processor, a tensor processing unit, etc. Additionally or alternatively, the functionality described herein can be performed, at least in part, by one or more hardware logic components. For example, and without limitation, illustrative types of hardware logic components that can be used include field-programmable gate arrays (FPGAs) , application-specific integrated circuits (ASICs) , application-specific standard products (ASSPs) , system-on-a-chip systems (SOCs) , complex programmable logic devices (CPLDs) , etc.

The memory 204 may include processor-readable media in a form of volatile memory, such as Random Access Memory (RAM) and/or non-volatile memory, such as read only memory (ROM) or flash RAM. The memory 204 is an example of processor-readable media.

The processor-readable media may include a volatile or non-volatile type, a removable or non-removable media, which may achieve storage of information using any method or technology. The information may include a processor-readable instruction, a data structure, a program module or other data. Examples of processor-readable media include, but not limited to, phase-change memory (PRAM) , static random access memory (SRAM) , dynamic random access memory (DRAM) , other types of random-access memory (RAM) , read-only memory (ROM) , electronically erasable programmable read-only memory (EEPROM) , quick flash memory or other internal storage technology, compact disk read-only memory (CD-ROM) , digital versatile disc (DVD) or other optical storage, magnetic cassette tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission media, which may be used to store information that may be accessed by a computing device. As defined herein, the processor-readable media does not include any transitory media, such as modulated data signals and carrier waves.

Although in this example, only hardware components are described in the differential directional sensor system 102, in other instances, the differential directional sensor system 102 may further include other hardware components and/or other software components such as program units to execute instructions stored in the memory 204 for performing various operations such as processing, determination, allocation, storage, etc. In implementations, the differential directional sensor system 102 may further include a weight database 214 that is configured to store a collection of sets of predetermined weights. Each set of predetermined weights may cause the differential directional sensor system 102 to preferentially weigh or emphasize signals received by the plurality of directional sensors 208 from a certain preferred or desired direction with respect to a particular point (e.g., a common center) in the particular geometric configuration.

Example Methods

FIG. 3 shows a schematic diagram depicting an example method of signal processing using the example differential directional sensor system. The method of FIG. 3 may, but need not, be implemented in the environment of FIG. 1 and using the system of FIG. 2. For ease of explanation, method 300 is described with reference to FIGS. 1 and 2. However, the method 300 may alternatively be implemented in other environments and/or using other systems.

The method 300 is described in the general context of computer-executable instructions. Generally, computer-executable instructions can include routines, programs, objects, components, data structures, procedures, modules, functions, and the like that perform particular functions or implement particular abstract data types. Furthermore, each of the example methods are illustrated as a collection of blocks in a logical flow graph representing a sequence of operations that can be implemented in hardware, software, firmware, or a combination thereof. The order in which the method is described is not intended to be construed as a limitation, and any number of the described method blocks can be combined in any order to implement the method, or alternate methods. Additionally, individual blocks may be omitted from the method without departing from the spirit and scope of the subject matter described herein. In the context of software, the blocks represent computer instructions that, when executed by one or more processors, perform the recited operations. In the context of hardware, some or all of the blocks may represent application specific integrated circuits (ASICs) or other physical components that perform the recited operations.

Referring back to FIG. 3, at block 302, the differential directional sensor system 102 may obtain or collect input signals received by a plurality of directional sensors that are arranged in a particular geometric configuration.

In implementations, the plurality of directional sensors may detect and receive input signals, and transmit the input signals to the differential directional sensor system 102 via a wireless or wired means, depending on whether the plurality of directional sensors are included in or auxiliary to the differential directional sensor system 102. By way of example and not limitation, the plurality of directional sensors may be included in the differential directional sensor system 102, or may be auxiliary or peripheral to the differential directional sensor system 102. In this case, the plurality of directional sensors may transmit the input signals to the differential directional sensor system 102 via a cable or wire, for example.

In implementations, the particular geometric configuration may be two-dimensional or three-dimensional. Additionally, the geometric configuration may be formed by, for example, a number of circles having a common center (e.g., a two-dimensional geometric configuration) , or a number of groups of circles having different centers with each group of circles having a same center (e.g., a three-dimensional geometric configuration) . By way of example and not limitation, FIGS. 4A-4F show schematic diagrams depicting example geometric configurations of the plurality of directional sensors. The dark spots in FIGS. 4A-4F represent directional sensors and corresponding positions in respective geometric configurations. Respective numbers of dark spots (i.e., respective numbers of directional sensors) in the figures are merely examples and not limitation. The number of directional sensors in a particular geometric configuration may vary and depend on corresponding application specifications of the plurality of directional sensors and/or the differential directional sensor system 102. FIG. 4A shows an example configuration with a plurality of directional sensors being distributed on a circumstance of a circle. FIGS. 4B-4D show different example configurations with each example configuration having a plurality of concentric circles (two concentric circles are shown as the figure as examples and not limitations) and a plurality of directional sensors being distributed on circumstances of the plurality of concentric circles. As can be seen in FIGS. 4B-4D, depending on a distribution of the plurality of directional sensors, different geometric configurations or shapes are formed by the plurality of directional sensors. The different example configurations of directional sensors as shown in FIGS. 4A-4D correspond to two-dimensional geometric configurations. In implementations,

FIGS. 4E and 4F show different example geometric configurations of directional sensors in three-dimensional space. FIG. 4E shows an example configuration with a plurality of directional sensors being distributed on a cylindrical surface of a cylinder. FIG. 4F shows an example configuration with a plurality of directional sensors being distributed on a spherical surface of a sphere. Apparently, other geometric configurations, such as a configuration with a plurality of directional sensors being distributed on a plurality of co-axial cylindrical surfaces, a geometric configuration with a plurality of directional sensors being distributed on a plurality of concentric spherical surfaces, etc., may also be possible, depending on the use and application of the plurality of directional sensors, which are not limited by this disclosure.

In implementations, the plurality of directional sensors in the geometric configuration (such as the geometric configurations shown in FIGS. 4A-4F) may or may not be equally spaced with each other. In implementations, the plurality of directional sensors (or respective signal sensing or detection components of the plurality of directional sensors) may face outward from specific point (s) of the particular geometric configuration. For example, the plurality of directional sensors may face outward from a center of the particular geometric configuration (such as the geometric configurations as shown in FIGS. 4A-4D) , or respective one or more centers of the particular geometric configuration (such as the geometric configurations as shown in FIGS. 4E and 4F) . Furthermore, although FIGS. 4A-4F show certain number of directional sensors, the number of directional sensors in the particular configuration is not limited thereto, and can be any positive integer, such as 3, 4, 5, 8, 10, 20, 50, etc., depending on the use and application of the plurality of directional sensors or the differential directional sensor system 102, the desired quality and/or sensitivity of the plurality of directional sensors or the differential directional sensor system 102, the type of input signals detected by the plurality of directional sensors or the differential directional sensor system 102, etc.

In implementations, the input signals may include signals coming from a particular direction under a predefined coordinate system associated with the particular geometric configuration of the plurality of directional sensors. By way of example and not limitation, the predefined coordinate system may have an origin located at the center of the particular geometric configuration.

In implementations, a type of the plurality of directional sensors that is used by the differential directional sensor system 102 depends on a type of the input signals. For example, if the input signals are sound or acoustic signals, the type of the plurality of directional sensors are acoustic sensors such as directional microphones. If the input signals are radio signals, the type of the plurality of directional sensors are radio sensors such as directional antennas. If the input signals are visual signals, the type of the plurality of directional sensors are visual sensors such as directional cameras.

In implementations, a directional sensor of the plurality of directional sensors may be unidirectional, or substantially unidirectional having a limited signal detection range (e.g., an angular range of -2° to +2°, -5° to +5°, or -10° to +10°, etc. ) and/or a signal detection capability or gain being decreased as an angle is deviated from a center axis of a signal sensing or detection component of the directional sensor.

At block 304, the differential directional sensor system 102 may perform a first predefined transformation on the input signals to obtain respective transformed signals.

In implementations, after obtaining or collecting the input signals from the plurality of directional sensors, the differential directional sensor system 102 may perform a first predefined transformation of the input signals to convert the input signals from one domain into another domain. By way of example and not limitation, the differential directional sensor system 102 may apply a first predefined transformation on the input signals to convert the input signals from a time domain to a frequency domain and obtain respective frequency bands. In implementations, the first predefined transformation may include, but is not limited to, a short time Fourier transform, a predetermined set of filter banks, or any transformation that is capable of transforming a signal from the time domain to the frequency domain, etc.

Additionally or alternatively, the differential directional sensor system 102 may apply the first predefined transformation or decomposition on the input signals to convert the input signals from one representation to another representation. Depending on the type of the input signals (such as acoustic signals, radio signals, or visual signals, etc. ) , examples of the first predefined transformation or decomposition may include, but are not limited to, a Wavelet transform, etc.

At block 306, the differential directional sensor system 102 may apply a set of predetermined weights to the respective transformed signals to obtain a combined signal.

In implementations, after obtaining the respective transformed signals from the input signals, the differential directional sensor system 102 may apply a set of predetermined weights to the respective transformed signals. In implementations, the number of predetermined weights in the set is equal to the number of directional sensors in the geometric configuration of the plurality of directional sensors. In implementations, the predetermined weights of the set have a one-to-one correspondence relationship with the plurality of directional sensors in the geometric configuration.

In implementations, the differential directional sensor system 102 may select the set of predetermined weights from among a collection of predetermined weights based on the particular direction. The collection of predetermined weights may include respective sets of predetermined weights for different directions under the predefined coordinate system associated with the particular geometric configuration. In implementations, the set of predetermined weights selected by the differential directional sensor system 102 may correspond to a particular direction under the predefined coordinate system associated with the particular geometric configuration of the plurality of directional sensors.

In implementations, the differential direction sensor system 102 may store the collection of sets of predetermined weights in the memory 204, e.g., the weight database 214. Additionally or alternatively, the collection of sets of predetermined weights may be stored in a device that is remote or peripheral to the differential directional sensor system 102, such as the one or more servers 106 or a peripheral device of the differential directional sensor system 102, and the differential directional sensor system 102 is allowed to access the collection of sets of predetermined weights from that remote or peripheral device.

In implementations, each set of predetermined weights may cause the differential directional sensor system 102 to preferentially weigh or emphasize signals received by the plurality of directional sensors from a certain preferred or desired direction with respect to a particular point (e.g., a common center) in the particular geometric configuration or an origin of a coordinate system associated with the particular geometric configuration with the origin being the common center, for example. In this case, the differential directional sensor system 102 may select the set of predetermined weights from the collection of sets of predetermined weights, and apply this set of predetermined weights to the respective transformed signals to obtain the combined signal. In other words, by selecting a different set of predetermined weights (e.g., a certain set of predetermined weights for a certain preferred or desired direction) , the differential directional sensor system 102 may electronically adjust the plurality of directional sensors to receive signal from the preferred or desired direction without the need of mechanically moving any physical parts of the plurality of directional sensors and/or the differential directional sensor system 102.

In implementations, the collection of sets of predetermined weights may be determined or calculated in advance at least in part on one or more predefined constraints, beam patterns of the plurality of directional sensors, a number of the plurality of directional sensors, and relative positions of the plurality of directional sensors under a coordinate system associated with the particular geometric configuration. In implementations, the predetermined weights may be frequency dependent, i.e., each predetermined weight being a function depending on frequencies. An example method of determining or calculating a collection of sets of predetermined weights associated with a particular geometric configuration of a plurality of directional sensors for different incoming signal directions or angles will be described in a later section.

At block 308, the differential directional sensor system 102 may perform a second predefined transformation on the combined signal to obtain an output signal.

In implementations, after applying the set of predetermined weights to the respective transformed signals to obtain the combined signal, the differential directional sensor system 102 may perform a second predefined transformation on the combined signal to obtain an output signal. In implementations, the second predefined transformation is opposite or inverse to the first predefined transformation. For example, if the first predefined transformation is used to transform signals from a first representation or domain (such as a time domain) to a second representation or domain (e.g., a frequency domain) , the second predefined transformation is used to transform signals from the second representation or domain to the first representation or domain.

Although the above method blocks are described to be executed in a particular order, in some implementations, some or all of the method blocks can be executed in other orders, or in parallel. For example, the differential directional sensor system 102 may obtain new input signals from the plurality of directional sensors, while processing (such as transforming, weighting, inversely transforming, etc. ) input signals that are previously obtained from the plurality of directional sensors or input signals obtained from another set or group of directional sensors that the differential directional sensor system 102 is connected to and responsible for, etc.

Additionally, although in this example the differential directional sensor system 102 is described to include or be associated with a certain set of directional sensors (i.e., the plurality of directional sensors arranged in the particular geometric configuration as described above) , the differential directional sensor system 102 may include or be associated with multiple sets of directional sensors, and process respective signals received from each set of directional sensors as described above. The multiple sets of directional sensors may or may not be of the same type (i.e., an acoustic type, a radio type, a visual type, etc. ) , and may or may not have the same geometric configuration. In this case, the differential directional sensor system 102 may store or have an access to various collections of sets of predetermined weights suitable for these multiple sets of directional sensors.

Additionally, the differential directional sensor system 102 may estimate or determine a direction from which desired or certain signals originate or come. By way of example and not limitation, the differential directional sensor system 102 may determine respective signal strengths of the desired or certain signals received by the plurality of directional sensors. The differential directional sensor system 102 may then estimate or determine a direction from which the desired or certain signals come or originate by, for example, weighted averaging directions of such one or more directional sensors based on the detected signal strengths of the desired or certain signals.

Example Directional Sensor Configuration

By way of example and not limitation, directional microphones are used herein as an example of the directional sensors as described above. It should be noted that the present disclosure is not limited to this example of directional sensors, and other types of directional sensors can also be used and applicable to the present disclosure.

In implementations, a beam pattern of a directional microphone may be expressed as: p+ (1-p) cosα, where α is an off-axis angle, and p defines a property of the directional microphone. For example, a cardioid beam pattern is resulted when p=0.5, and a dipole when p=0. In implementations, the directional microphone may be implemented using two approaches, namely, a first approach which employs a dedicated directional microphone with a single microphone cartridge having two sound inlets, or a second approach which employs a two-omni-directional-element system with appropriate digital signal processing. The first approach yields a better directional microphone in term of signal-to-noise ratio (SNR) as compared to the second approach because signal processing that creates directivity is done acoustically with front and rear sound inlets of the dedicated directional microphone in the first approach. In implementations, the dedicated directional microphone may be implemented in a form of ECM (Electret Condenser Microphone) or MEMS (micro-Electro-Mechanical System) .

For the sake of description, a circular configuration having directional sensors (e.g., directional microphones) being distributed on a circumstance of a circle is used as an example. Nevertheless, the following description can also be applicable to other more complicated geometric configurations of directional sensors (e.g., directional microphones) because these complicated geometric configurations can be formed by or decomposed into a plurality of concentric circles, or a number of groups of circles having different centers with each group of circles having a same center, with the directional sensors being correspondingly distributed on circumstances of the plurality of concentric circles, or circumstances of circles in the number of groups of circles, as exemplified and shown in FIGS. 4A-4F. Results for these complicated geometric configurations of directional sensors may then be obtained through superposition of respective results of individual circles of directional sensors.

In implementations, such circular configuration of directional microphones can be called as a circular differential directional microphone array (CDDMA) . In this example, a single desired sound source in the far-field is considered, and appears as a plane wave impinging on a circular configuration of M directional microphones with a radius of r. As shown in FIG. 5, the directional microphones are distributed on the circumstance of a circle, and pointed outward. The directional microphones may or may not be evenly distributed on the circumstance. For the sake of simplicity and without loss of generality, the directional microphones are described to be evenly distributed on the circumstance in this example. An azimuth angle θ represents a direction of arrival of sound from the sound source, and c represents the speed of sound.

In this scenario, a steering vector is defined as:

d (ω, θ) = [d ₁, d ₂, …, d _m, …, d _M] ^T (1)

where the superscript ^T is a transpose operator, and each element in Equation (1) for the uniform CDDMA can be obtained as:

where

is an imaginary unit, ω=2πf is an angular frequency, f is a temporal frequency, and

is an angular position of the m ^th element in this example uniform CDDMA.

Using Equation (2) , the steering vector, d (ω, θ) , can be rewritten as:

d (ω, θ) =U (p, θ) a (ω, θ) (3)

where

a (ω, θ) = [a ₁, a ₂, …, a _m, …, a _M] ^T (4)

where

and

U (p, θ) =diag (u ₁, u ₂, …, u _M, …, u _M) (5)

U (p, θ) is called a microphone response matrix, where

In implementations, a differential beamforming with the microphone array (or a differential beamformer or simply called beamformer hereinafter) can be used for estimating a target signal arriving from a desired direction in the presence of noise and interference. In this example, the differential beamforming can be interpreted as a spatial filter to estimate a signal from a desired direction and suppress signals from other undesired directions by applying a complex weight vector:

h (ω) = [H ₁ (ω) H ₂ (ω) …H _M (ω) ] ^T (6)

Given a signal model, the beamformer may exhibit a distortion-less response in the desired direction θ _desired, while having a certain distortion in the response for other undesired directions, i.e.,

where the superscript ^H is a conjugate-transpose operator.

In implementations, three performance measures, namely, a white-noise-gain (WNG) , a directivity factor (DF) , and a (frequency-invariant) beampattern, are defined herein. In implementations, WNG can show the ability of a beamformer to suppress spatially uncorrelated noise, and can be used for evaluating the sensitivity of the beamformer to certain imperfections such as sensor noise, position errors, etc. In implementations, WNG may be defined as:

In implementations, the beampattern may illustrate the directional sensitivity of a beamformer to a plane wave impinging on the array from an incident angle θ (as shown in FIG. 1) :

In implementations, the power pattern, i.e.,

may be used for representing the performance. It is noted that the frequency-invariant beampattern is usually desirable or preferred. In implementations, DF may be defined as a ratio between the signal power (in the array output) in a desired steering direction and the signal power averaged over all directions:

where θ is an azimuth angle, φ is an elevation angle,

is a beampattern in a spherical coordinate system. In implementations, a directivity index (DI) may be defined as

In implementations, for designing a CDDMA beamformer, the problem may be formulated as a linear system of equations as follows:

R (ω, θ) h (ω) =c _θ (11)

where h (ω) is CDDMA beamforming weights (i.e., a set of predetermined weights as described above) that are to be obtained. c _θ is a constraint vector, which is

where θ= [θ ₁, θ ₂, …, θ _N] is an angle vector, and N is the number of constraints with N≤M. R (ω, θ) is a constraint matrix of N×M, which is given by:

In implementations, based on Equation (11) , properties of the beamformer may be determined by the constraint vector c _θ and the angle parameter vector θ that need to be specified in the design. By way of example and not limitation, if

in Equation (11) , the steering vector at θ ₁ follows d ^H (ω, θ ₁) h (ω) =1, which means that θ ₁ becomes the desired direction of the CDDMA beamformer that produces the distortion-less output for the sound coming from that direction. The remaining

for i=2, 3, …, N may be set to zero, i.e., these θ _i determine or decide the nulls of the beampattern.

In implementations, a minimum-norm solution may be used to solve the linear system equation such as Equation (11) , and the CDDMA beamformer may be obtained by:

h (ω) =R ^H (ω, θ) [R (ω, θ) R ^H (ω, θ) ] ^-1c _θ (13)

For example, if the desired direction, θ _desired, is 0 degree, i.e., c ₀=1 in Equation (11) , and p=0.5 in Equation (2) , i.e., taking the cardioid element as an example of a directional microphone to form a CDDMA beamformer. FIG. 6A-6F show comparisons of beampatterns for two different designs between CDDMA and circular differential microphone array (CDMA) that uses omni-directional microphones at frequencies of 1 kHz, 3 kHz, and 6 kHz, where r = 1.5 cm, and M = 8 are used. The first design is to construct a 1st-order cardioid (c _π=0) while the second design is to build a second-order cardioid

As can be seen from FIGS. 6A, 6C, and 6E, the beampatterns of CDDMA and CDMA are very close to the desired 1st-order cardioid at 1 kHz and 3 kHz. The CDMA beampattern deviates significantly at 6 kHz, whereas the CDDMA beampattern still holds for the desired design. Therefore, CDDMA is more frequency-invariant than CDMA for the design of 1st-order cardioid. Both CDMA and CDDMA for the 2nd-order cardioid are quite frequency-invariant, as depicted in FIGS. 6B, 6D, and 6F.

FIGS. 7A and 7B show comparisons between CDDMA and CDMA in terms of WNG and DI for the same designs described above. As can be seen from FIG. 7A, CDDMA for both designs exhibits a lot higher WNG as compared to CDMA at the low frequencies where WNG is usually concerned. Meanwhile, FIG. 7B indicates that CDDMA has an improvement in DI at the high frequencies as compared to CDMA. Lastly, for both CDMA and CDDMA, higher-order beamformers lead to a higher DI and a lower WNG at the low frequencies. Accordingly, the CDDMA is more frequency-invariant, and WNG of the CDDMA beamformer at low frequencies is significantly improved by employing the mini-norm solution, as compared to CDMA using omni-directional microphones, while DI of the CDDMA beamformer exhibits an improvement at high frequencies as compared to CDMA.

Conclusion

Although implementations have been described in language specific to structural features and/or methodological acts, it is to be understood that the claims are not necessarily limited to the specific features or acts described. Rather, the specific features and acts are disclosed as exemplary forms of implementing the claimed subject matter. Additionally or alternatively, some or all of the operations may be implemented by one or more ASICS, FPGAs, or other hardware.

The present disclosure can be further understood using the following clauses.

Clause 1: A method implemented by a computing device, the method comprising: obtaining input signals received by a plurality of directional sensors that are arranged in a particular configuration; transforming the input signals from a time domain to a frequency domain to obtain respective frequency bands; applying a set of predetermined weights to the respective frequency bands in the frequency domain to obtain a combined frequency band; and transforming the combined frequency band from the frequency domain to the time domain to obtain an output signal.

Clause 2: The method of Clause 1, wherein the particular configuration comprises a circular configuration with the directional sensors being located on a circumstance of a circle having a predetermined radius, and the directional sensors face outward from a center of the circle.

Clause 3: The method of Clause 1, wherein the particular configuration comprises a configuration with the directional sensors being located on respective one or more circumstances of one or more concentric circles having different radii, and the directional sensors face outward from a common center of the one or more concentric circles.

Clause 4: The method of Clause 1, wherein transforming the input signals from the time domain to the frequency domain to obtain the respective frequency bands comprising applying a short time Fourier transform or a predetermined set of filter banks to the input signals.

Clause 5: The method of Clause 1, wherein the input signals comprise signals coming from a particular direction under a predefined coordinate system associated with the plurality of directional sensors.

Clause 6: The method of Clause 5, further comprising determining the set of predetermined weights from among a collection of predetermined weights based on the particular direction, the collection of predetermined weights comprising respective sets of predetermined weights for different directions under the predefined coordinate system associated with the center of the plurality of directional sensors.

Clause 7: The method of Clause 1, wherein the predetermined weights are calculated based at least in part on one or more predefined constraints, beam patterns of the directional sensors, a number of the directional sensors, and relative positions of the directional sensors under a coordinate system associated with the plurality of directional sensors.

Clause 8: The method of Clause 1, wherein the predetermined weights are frequency dependent.

Clause 9: The method of Clause 1, wherein the plurality of directional sensors comprises a plurality of directional microphones.

Clause 10: A system comprising: a plurality of directional sensors arranged in a particular configuration, the plurality of directional sensors being configured to receive input signals; a memory configured to store a collection of sets of predetermined weights; one or more processors configured to: obtain input signals received by a plurality of directional sensors that are arranged in a particular configuration; transform the input signals from a time domain to a frequency domain to obtain respective frequency bands; apply a set of predetermined weights to the respective frequency bands in the frequency domain to obtain a combined frequency band; and transform the combined frequency band from the frequency domain to the time domain to obtain an output signal.

Clause 11: The system of Clause 10, wherein the particular configuration comprises a circular configuration with the directional sensors being located on a circumstance of a circle having a predetermined radius, and the directional sensors face outward from a center of the circle.

Clause 12: The system of Clause 10, wherein the particular configuration comprises a configuration with the directional sensors being located on respective one or more circumstances of one or more concentric circles having different radii, and the directional sensors face outward from a common center of the one or more concentric circles.

Clause 13: The system of Clause 10, wherein transforming the input signals from the time domain to the frequency domain to obtain the respective frequency bands comprising applying a short time Fourier transform or a predetermined set of filter banks to the input signals.

Clause 14: The system of Clause 10, wherein the input signals comprise signals coming from a particular direction under a predefined coordinate system associated with the plurality of directional sensors.

Clause 15: The system of Clause 14, wherein the one or more processors is further configured to determine the set of predetermined weights from among the collection of predetermined weights based on the particular direction, the collection of predetermined weights comprising respective sets of predetermined weights for different directions under the predefined coordinate system associated with the center of the plurality of directional sensors.

Clause 16: The system of Clause 10, wherein the predetermined weights are calculated based at least in part on one or more predefined constraints, beam patterns of the directional sensors, a number of the directional sensors, and relative positions of the directional sensors under a coordinate system associated with the plurality of directional sensors.

Clause 17: The system of Clause 10, wherein the predetermined weights are frequency dependent.

Clause 18: The system of Clause 10, wherein the plurality of directional sensors comprises a plurality of directional microphones.

Clause 19: One or more computer readable media storing executable instructions that, when executed by one or more processors, cause the one or more processors to perform acts comprising: obtaining input signals received by a plurality of directional sensors that are arranged in a particular configuration; transforming the input signals from a time domain to a frequency domain to obtain respective frequency bands; applying a set of predetermined weights to the respective frequency bands in the frequency domain to obtain a combined frequency band; and transforming the combined frequency band from the frequency domain to the time domain to obtain an output signal.

Clause 20: The one or more computer readable media of Clause 19, wherein the input signals comprise signals coming from a particular direction under a predefined coordinate system associated with the plurality of directional sensors, and the acts further comprise determining the set of predetermined weights from among a collection of predetermined weights based on the particular direction, the collection of predetermined weights comprising respective sets of predetermined weights for different directions under the predefined coordinate system associated with the center of the plurality of directional sensors.

Claims

A method implemented by a computing device, the method comprising:

obtaining input signals received by a plurality of directional sensors that are arranged in a particular configuration;

transforming the input signals from a time domain to a frequency domain to obtain respective frequency bands;

applying a set of predetermined weights to the respective frequency bands in the frequency domain to obtain a combined frequency band; and

transforming the combined frequency band from the frequency domain to the time domain to obtain an output signal.
The method of claim 1, wherein the particular configuration comprises a circular configuration with the directional sensors being located on a circumstance of a circle having a predetermined radius, and the directional sensors face outward from a center of the circle.
The method of claim 1, wherein the particular configuration comprises a configuration with the directional sensors being located on respective one or more circumstances of one or more concentric circles having different radii, and the directional sensors face outward from a common center of the one or more concentric circles.
The method of claim 1, wherein transforming the input signals from the time domain to the frequency domain to obtain the respective frequency bands comprising applying a short time Fourier transform or a predetermined set of filter banks to the input signals.
The method of claim 1, wherein the input signals comprise signals coming from a particular direction under a predefined coordinate system associated with the plurality of directional sensors.
The method of claim 5, further comprising determining the set of predetermined weights from among a collection of predetermined weights based on the particular direction, the collection of predetermined weights comprising respective sets of predetermined weights for different directions under the predefined coordinate system associated with the center of the plurality of directional sensors.
The method of claim 1, wherein the predetermined weights are calculated based at least in part on one or more predefined constraints, beam patterns of the directional sensors, a number of the directional sensors, and relative positions of the directional sensors under a coordinate system associated with the plurality of directional sensors.
The method of claim 1, wherein the predetermined weights are frequency dependent.
The method of claim 1, wherein the plurality of directional sensors comprises a plurality of directional microphones.
A system comprising:

a plurality of directional sensors arranged in a particular configuration, the plurality of directional sensors being configured to receive input signals;

a memory configured to store a collection of sets of predetermined weights;

one or more processors configured to:

obtain input signals received by a plurality of directional sensors that are arranged in a particular configuration;

transform the input signals from a time domain to a frequency domain to obtain respective frequency bands;

apply a set of predetermined weights to the respective frequency bands in the frequency domain to obtain a combined frequency band; and

transform the combined frequency band from the frequency domain to the time domain to obtain an output signal.
The system of claim 10, wherein the particular configuration comprises a circular configuration with the directional sensors being located on a circumstance of a circle having a predetermined radius, and the directional sensors face outward from a center of the circle.
The system of claim 10, wherein the particular configuration comprises a configuration with the directional sensors being located on respective one or more circumstances of one or more concentric circles having different radii, and the directional sensors face outward from a common center of the one or more concentric circles.
The system of claim 10, wherein transforming the input signals from the time domain to the frequency domain to obtain the respective frequency bands comprising applying a short time Fourier transform or a predetermined set of filter banks to the input signals.
The system of claim 10, wherein the input signals comprise signals coming from a particular direction under a predefined coordinate system associated with the plurality of directional sensors.
The system of claim 14, wherein the one or more processors is further configured to determine the set of predetermined weights from among the collection of predetermined weights based on the particular direction, the collection of predetermined weights comprising respective sets of predetermined weights for different directions under the predefined coordinate system associated with the center of the plurality of directional sensors.
The system of claim 10, wherein the predetermined weights are calculated based at least in part on one or more predefined constraints, beam patterns of the directional sensors, a number of the directional sensors, and relative positions of the directional sensors under a coordinate system associated with the plurality of directional sensors.
The system of claim 10, wherein the predetermined weights are frequency dependent.
The system of claim 10, wherein the plurality of directional sensors comprises a plurality of directional microphones.
One or more computer readable media storing executable instructions that, when executed by one or more processors, cause the one or more processors to perform acts comprising:

obtaining input signals received by a plurality of directional sensors that are arranged in a particular configuration;

transforming the input signals from a time domain to a frequency domain to obtain respective frequency bands;

applying a set of predetermined weights to the respective frequency bands in the frequency domain to obtain a combined frequency band; and

transforming the combined frequency band from the frequency domain to the time domain to obtain an output signal.
The one or more computer readable media of claim 19, wherein the input signals comprise signals coming from a particular direction under a predefined coordinate system associated with the plurality of directional sensors, and the acts further comprise determining the set of predetermined weights from among a collection of predetermined weights based on the particular direction, the collection of predetermined weights comprising respective sets of predetermined weights for different directions under the predefined coordinate system associated with the center of the plurality of directional sensors.