Device location prediction based on active sound perception

By recording and processing predefined sound samples within the ultrasonic frequency range using a microphone, and combining matched filtering and artificial intelligence models, the privacy and accuracy issues of device detection in the same space in existing technologies are solved, achieving efficient, wide-ranging, and interference-resistant device detection in the same space.

CN121693777APending Publication Date: 2026-03-17HUAWEI TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202380101376.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-08-25
Publication Date
2026-03-17

AI Technical Summary

Technical Problem

Existing solutions for detecting devices in the same space lack universality, efficiency, accuracy, coverage, robustness, and privacy. In particular, those based on cameras and electromagnetic fingerprint technology have limitations and privacy issues, and two-way ranging cannot determine whether devices are in the same space.

Method used

Sound recordings are recorded and processed using microphones to extract multipath version features between devices. An artificial intelligence model is used to determine whether devices are in the same space. Predefined sound samples within the ultrasonic frequency range are used for detection, and location classification is performed by combining matched filtering and feature extraction.

Benefits of technology

It enables efficient and accurate detection of whether devices are in the same space without infringing on privacy. It is applicable to various smart devices, has a wide coverage, and has anti-interference capabilities, meeting the requirements of universality and cost-effectiveness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121693777A_ABST
    Figure CN121693777A_ABST
Patent Text Reader

Abstract

A computer-implemented method and system for predicting device location includes recording a sound record using a microphone on a first device; processing the sound record to extract a feature corresponding to a multipath version of a sound sample played by a second device; based on the extracted features, classifying the physical location of the second device into one of (i) the second device is located in the same space as the first device or (ii) the second device is not located in the same space as the first device; an operation is performed on the first device based on the classification.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] Relevant application information This is the first application filed for this invention. Technical Field

[0002] This application generally relates to methods, systems, and computer media related to device location prediction and detection based on active sound perception. Background Technology

[0003] Electronic devices equipped with processors and capable of communicating with each other via various wireless protocols (such as Bluetooth, Zigbee, NFC, Wi-Fi, LiFi, or 5G) are commonly referred to as smart devices and are now ubiquitous. Some notable smart commercial-off-the-shelf (COTS) devices include smartphones, smart TVs, smart speakers, smart earphones, smart thermostats, smart doorbells, smart locks, smart refrigerators, phablets and tablets, smartwatches, smart bracelets, smart keychains, smart glasses, and many other devices.

[0004] Smart COTS devices typically include speakers and microphones, enabling audio notifications and voice commands. In many modern use cases, a single user can associate with multiple smart devices that can interact with each other. Furthermore, multiple smart devices can be associated with or registered to a public smart home or smart office network. Cross-device interoperability can be achieved, for example, through collaborative software installed on multiple devices. For instance, a distributed operating system enables multiple smart devices to collaborate and connect, especially when they are close to each other. Cross-device interoperability is often required when devices are located in public or shared spaces (such as a room in a residence (e.g., a living room, bedroom, study, home office, family room, kitchen, etc., or a room in a commercial or business setting such as a meeting room or office) or inside a vehicle (e.g., the interior of a car, RV, boat, bus, etc.)).

[0005] A key prerequisite for cross-device interoperability within the same space is that the corresponding devices can detect and identify other devices present in the same space, thereby enabling interaction. The desired characteristics of a cross-device detection scheme for electronic devices within the same space include: (1) universality (e.g., the scheme can be easily implemented on various COTS devices); (2) efficiency (e.g., the scheme can efficiently utilize computing / memory resources while being cost-effective); (3) accuracy (e.g., the scheme can successfully detect and identify other devices in public spaces with high precision); (4) coverage (e.g., the scheme covers all or substantially all public spaces); (5) robustness (e.g., the scheme is robust against interference); and (6) privacy (e.g., the scheme mitigates or does not introduce privacy issues).

[0006] While similar spatial detection schemes have been proposed, they typically lack at least some of the desired characteristics described above. For example, some device detection systems rely on images captured by the device's built-in camera. By processing and analyzing these images, the device can learn about its environment, including information about the presence of objects or users around the device and their distances. Computer vision-based technologies have been developed for many years and can provide high accuracy; however, camera-based schemes raise privacy concerns, as users may feel they are being monitored and surveilled. Furthermore, cameras in devices typically have limited fields of view, thus limiting their directional coverage when detecting other objects.

[0007] Another existing approach is to use electromagnetic (EM) fingerprinting to detect electronic devices. This approach collects Wi-Fi, BLE signal strength, or magnetic fingerprint data from various devices located in space through EM signal field surveying to create a local fingerprint dataset. Devices capable of detection can then collect location fingerprints and use this dataset to estimate the location of other devices in the space. This approach requires EM field surveying and may lack robustness and accuracy.

[0008] Some known methods employ two-way ranging to determine the distance between devices. In these methods, the time of flight (TOF) of ultra-wideband radio frequency signals or acoustic signals is measured and used to calculate the distance between the two devices. While two-way ranging can indicate the distance between objects, it cannot indicate whether the two objects are physically located in the same internal space.

[0009] Therefore, existing methods for detecting objects in the same space all have their corresponding drawbacks. There is a need for methods, systems, and computer media for detecting objects in the same space that can overcome the shortcomings of known methods. Summary of the Invention

[0010] According to a first exemplary aspect, a computer-implemented method for detecting devices within an environment is disclosed. The method includes: recording sound using a microphone on a first device; processing the sound recording to extract features corresponding to a multipath version of a sound sample played by a second device; classifying the physical location of the second device based on the extracted features as one of: (i) being located in the same space as the first device, or (ii) not being located in the same space as the first device; and performing an operation on the first device based on the classification.

[0011] In some examples, the sound samples are in a frequency range that is inaudible to humans.

[0012] In one or more of the foregoing aspects, processing the sound recording includes: performing bandpass filtering to obtain a sound signal within a specified frequency band; performing matched filtering on the sound signal to extract a time series corresponding to the multipath version of the sound sample played by the second device; and extracting the features from the time series.

[0013] In one or more of the foregoing aspects, prior to extracting the features, the time series is processed to identify a first indexed sound segment within the time series, wherein the first indexed sound segment corresponds to the shortest sound propagation path from the second device to the first device for the multipath version of the sound sample, and the features are extracted based on attributes of a subset of the time series, the subset being selected based on the first indexed sound segment.

[0014] In one or more of the foregoing aspects, the feature includes one or more of the following: the maximum amplitude value included in the selected subset; the average amplitude value of the selected subset; the standard deviation amplitude value of the selected subset; the kurtosis amplitude value of the selected subset; the skewness amplitude value of the selected subset; the 25th percentile amplitude value of the selected subset; the 75th percentile amplitude value of the selected subset; the root mean square amplitude value of the selected subset; the number of sampled amplitude values ​​in the selected subset that are greater than the product of a specified coefficient value and the maximum amplitude value; the selected The sum of a specified number of amplitude peaks appearing in the selected subset; the time offset between the first and last amplitude peaks in the selected subset; the average amplitude of the amplitude peaks included in the selected subset; the standard deviation of the amplitude peaks included in the selected subset; the kurtosis of the amplitude peaks included in the selected subset; the skewness of the amplitude peaks included in the selected subset; the 25th percentile of the amplitude peaks included in the selected subset; and the 75th percentile of the amplitude peaks included in the selected subset.

[0015] In one or more of the foregoing aspects, processing the time series to identify the first indexed sound segment includes: (i) identifying the maximum amplitude peak within the time series; (ii) identifying whether there are any amplitude peaks that satisfy a specified amplitude peak criterion and are located within a specified search range preceding the maximum amplitude peak; (iii) if one or more amplitude peaks are identified within the specified search range, selecting the amplitude peak immediately preceding the maximum amplitude peak to identify the first indexed sound segment; and if no amplitude peaks are identified within the specified search range, selecting the maximum amplitude peak to identify the first indexed sound segment.

[0016] In one or more of the foregoing aspects, classifying the physical location of the second device includes: applying an artificial intelligence model, wherein the artificial intelligence model has been trained to classify the physical location of the second device as one of the following: (i) located in the same space as the first device, or (ii) not located in the same space as the first device.

[0017] In one or more of the foregoing aspects, the second device is categorized as being located in the same space as the first device or physically located in the same room of the building as the first device, or the second device is categorized as not being located in the same space as the first device or physically not located in the same room of the building as the first device.

[0018] In one or more of the foregoing aspects, classifying the second device as being located in the same space as the first device corresponds to both the second device and the first device being physically located in a continuous interior space of the vehicle, and classifying the second device as not being located in the same space as the first device corresponds to both the second device and the first device not being physically located in a continuous interior space of the vehicle.

[0019] In one or more of the foregoing aspects, the frequency of the sound sample is greater than or equal to 17.5 kHz.

[0020] In one or more of the above aspects, the frequency of the sound sample is approximately between 20 kHz and 24 kHz.

[0021] In one or more of the foregoing aspects, the sound sample includes a fade-in tone portion, a constant amplitude chirp portion, and a fade-out tone portion.

[0022] In one or more of the foregoing aspects, performing the operation on the first device based on the classification includes: when the classification classifies the physical location of the second device as being in the same space as the first device, causing media content to be automatically streamed for playback through the speakers of the second device.

[0023] In one or more of the foregoing aspects, performing the operation on the first device based on the classification includes: when the classification classifies the physical location of the second device as not being in the same space as the first device, causing the first device to generate a notification output indicating that the second device does not exist.

[0024] In one or more of the foregoing aspects, performing the operation on the first device based on the classification includes: when the classification classifies the physical location of the second device as being in the same space as the first device, causing the first device to establish a connection with the second device to share media content with the second device.

[0025] In one or more of the foregoing aspects, the first device, the second device, and one or more other devices are all associated with a public wireless network, the sound recording includes a received multipath version of the sound sample played by the second device and one or more other sound samples played by the one or more other devices respectively; processing the sound recording includes: extracting other features, wherein the other features include corresponding features corresponding to the one or more other sound samples; the classification further includes: classifying the physical location of each of the one or more other devices based on the other features as one of the following: (i) the other device is located in the same space as the first device, or (ii) the other device is not located in the same space as the first device.

[0026] In one or more of the foregoing aspects, the method includes: the first device sending a request via the public wireless network to cause the second device to play the sound sample, and causing the one or more other devices to play a corresponding sound sample from the one or more other sound samples, wherein the sound sample and the one or more other sound samples have unique waveform properties that enable the sound sample and the one or more other sound samples to be uniquely identified.

[0027] In one or more of the foregoing aspects, the method includes: determining, based on the classification, the total number of the second device and the one or more other devices that are physically located in the same space as the first device, wherein performing the operation on the first device is based on the total number. In some examples, performing the operation on the first device includes: (i) when the total number is 1, automatically streaming or sharing media content for playback by the second device or the one or more other devices classified as being in the same space as the first device; (ii) when the total number is greater than 1, presenting a user-selectable device option on the first device, wherein the user-selectable device option identifies devices classified as being in the same space as the first device; (iii) when the total number is 0, generating an output on the first device, wherein the output indicates to the user that no device is classified as being in the same space as the first device.

[0028] According to another exemplary aspect, a system is disclosed, the system comprising one or more processors; and one or more memories storing machine-executable instructions, wherein, when executed by the one or more processors, the instructions cause the system to perform the method of any of the above methods.

[0029] According to another exemplary aspect, a non-transitory processor-readable medium storing machine-executable instructions is disclosed, wherein when the machine-executable instructions are executed by one or more processors, the one or more processors cause the one or more processors to perform the method of any of the above methods.

[0030] According to another exemplary aspect, a computer program is disclosed, wherein the computer program configures a computer system to perform the methods in any of the above methods. Attached Figure Description

[0031] The following figures illustrate exemplary embodiments of this application by way of example, in which: Figure 1 is a block diagram of an example interior environment, in which an exemplary embodiment of proximity detection based on active sound perception can be applied; Figure 2 This is a block diagram of a processor system according to an exemplary embodiment, wherein the processor system can be used to configure an electronic device to perform proximity detection based on active sound perception in the environment of FIG1; Figure 3 It is a flowchart illustrating a detection process that can be executed according to an exemplary implementation. Figure 4A is a plot showing the frequency versus time of a sound sample transmitted according to an exemplary implementation. Figure 4B is a plot of amplitude versus time for a sound sample transmitted according to an exemplary implementation. Figure 5 includes a first plot showing the time series of sound segments extracted from the received sound recording; Figure 6 This is a flowchart illustrating another example of a detection process that can be performed according to an exemplary implementation. Figure 7A illustrates how an electronic device displays a quick settings panel as part of a graphical user interface (GUI). Figure 7B shows the GUI of an electronic device displaying a first possible operation; Figure 7C shows the GUI of the electronic device displaying a second possible operation; Figure 7D shows the GUI of an electronic device displaying a third possible operation; Figure 8This is a flowchart illustrating another example of a detection process that can be performed according to an exemplary implementation. Figure 9 This is a flowchart illustrating another example of a detection process that can be implemented according to an exemplary model.

[0032] Similar reference numerals can be used to denote similar components in different accompanying drawings. Detailed Implementation

[0033] This invention describes methods, systems, and computer media for device detection based on active sound perception. In some examples, detection can be used to identify electronic devices located in the same space. In at least some examples, a first electronic device and a second electronic device are considered to be in the same space when they are located in a public physical space not separated by walls or other spatial partitions. For example, the same space can be: a continuous space within a building or other structure, which may be separated from other spaces within the building or structure by room partitions; the driver's cab or other continuous interior space of a vehicle; or a continuous space in an outdoor area. In an exemplary embodiment, the determination that the first and second devices are in the same space is based on features extracted from sound samples played from the speaker of the second device and received by the microphone of the first electronic device. Specifically, the extracted features are analyzed to determine whether these features meet criteria indicating that the first and second devices are in the same space. In an exemplary implementation, the sound samples are designed to be inaudible to the average person. In an exemplary implementation, standard electronic devices (e.g., COTS devices) are configured with software that enables such devices to perform active sound perception to detect and identify nearby devices, such as devices located in the same space, without additional hardware.

[0034] Figure 1 is a block diagram of an example of an interior environment 100, in which the examples described herein can be applied. In the illustrated example, environment 100 is an enclosed environment comprising multiple interior spaces or areas 130, 132, defined by respective sets of spatial separation barriers 104, which remain stationary relative to the interior areas and can at least partially reflect sound. Objects co-located in space 130 are located in the “same space.” Objects co-located in space 132 are located in the “same space.” In contrast, an object located in space 130 (e.g., electronic device 102A) is not located in the “same space” as an object located in space 132 (e.g., electronic device 102D). For example, in some scenarios, environment 100 may be an interior environment of a residence, office, or other structure, where the spatial separation barriers 104 include walls, floors, ceilings, closed windows, and closed doors, and the interior spaces 130, 132 are separate rooms (e.g., room A and room B). In the illustrated example, interior spaces 130, 132 are typically separated by barrier 104, but can be connected by unobstructed openings 124 (e.g., open doorways). In some alternative examples, environment 100 may be the interior of a vehicle, where the space-separating barrier 104 includes structural elements defining the vehicle's cockpit or interior space. Furthermore, environment 100 may include multiple objects (not shown) that serve as space-separating barriers, such as furniture, plants, and decorative items.

[0035] In the example of Figure 1, environment 100 includes multiple electronic devices 102A, 102B, 102C, and 102D (collectively referred to as electronic devices 102) that interact with each other via a local wireless network 108. For example, electronic devices 102 may all be pre-registered to a smart network (e.g., a smart home network) associated with the local wireless network 108. Identifications and other device data of the smart network member electronic devices 102 may be, for example, stored in a distributed register that can be accessed by the member electronic devices 102 when they are active in the smart network. Electronic devices 102 are all processor-enabled devices, including a corresponding processor system 110 and one or more of the following: (i) a speaker 112 for converting input audio signals into output sound waves propagating within environment 100, and (ii) a microphone 114 for capturing sound propagating within environment 100 and converting that sound into an input audio signal. Electronic devices 102 (at least some of which may be COTS devices) are configured with dedicated software instructions that configure a detection module 116 for their respective processor system 110. The detection module 116 enables electronic devices 102 to perform one or more active sound sensing functions described herein. For example, electronic devices 102 may include COTS devices such as smart TVs (e.g., device 102D), interactive smart speaker systems (e.g., device 102B), workstations (e.g., device 102C), and smartphones (e.g., device 102A), and other smart devices.

[0036] Figure 2An example of a processor system 110 architecture that can be applied to any corresponding electronic device 102 is shown. The processor system 110 includes one or more processors 202, such as a central processing unit, microprocessor, graphics processing unit (GPU), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), special-purpose logic circuit, tensor processing unit, neural processing unit, special-purpose artificial intelligence processing unit, or a combination thereof. The one or more processors 202 may be collectively referred to as a "processor device". The processor system 110 also includes one or more input / output (I / O) interfaces 204 that connect to input devices (e.g., microphone 114) and output devices (e.g., speaker 112). In some examples, other I / O devices, such as an inertial measurement unit (IMU) 210, may also be connected to provide input data to (or receive output data from) the processor system 110.

[0037] The processor system 110 may include one or more network interfaces 206, which, for example, enable the processor system 110 to communicate with one or more other devices via a wireless local area network 108 using one or more wireless protocols (such as Bluetooth, Zigbee, near field communication, Wi-Fi, LiFi, or 5G).

[0038] Processor system 110 includes one or more memories 208, which may include volatile or non-volatile memories (e.g., flash memory, random access memory (RAM), and / or read-only memory (ROM)). The one or more non-transient memories 208 may store instructions executable by one or more processors 202, for example, to perform the examples described in this invention. The one or more memories 208 may include other software instructions, for example, to implement an operating system and other applications / functions. In the illustrated example, memory 208 includes dedicated software instructions 116I for implementing detection module 116.

[0039] In some examples, processor system 110 may also include one or more electronic storage units (not shown), such as solid-state drives, hard disk drives, disk drives, and / or optical disk drives. In some examples, one or more datasets and / or modules may be provided by external memory (e.g., an external drive that communicates with processor system 110 via wired or wireless communication) or by transient or non-transient computer-readable media. Examples of non-transient computer-readable media include RAM, ROM, erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, CD-ROM, or other portable storage. Components of processor system 200 may communicate with each other via a bus or the like.

[0040] As used herein, “module” can refer to a combination of hardware processing circuitry (e.g., processor 202) and machine-readable instructions (software (e.g., detection module instructions 160I) and / or firmware) that can be executed on the hardware processing circuitry.

[0041] In an exemplary embodiment, electronic device 102, through its corresponding detection module 116, collaboratively executes an active sound perception process that enables the first electronic device (e.g., device 102A) to detect and identify which other devices 102 are located near it. For example, the active sound perception process may be used to detect which other devices 102B, 102C, and 102D are located in the same space (e.g., interior space 130) as the first electronic device 102A. In an exemplary embodiment, the first electronic device 102A performs one or more operations based on the detected devices.

[0042] Figure 3 A flowchart illustrating a basic example of a same space detection process 300 is shown, which can be performed for a first electronic device (e.g., device 102A) and a second electronic device (e.g., device 102B) in the context of environment 100 shown in FIG1.

[0043] Process 300 begins with a triggering event (operation 302). While triggering events can take many forms, in one exemplary embodiment, the triggering event is detected by the first device 102A. For example, the operating system of the first device 102A may be used to monitor one or more predefined user input events corresponding to the triggering event, and notify the detection module 116 of the first device 102A of the occurrence of the event and its type when such an event occurs. For example, the triggering event may be triggered by user input (e.g., button selection, verbal command, gesture) that requests the streaming of a song through the first device 102A to output sound via an external device (e.g., projection or sharing request).

[0044] In one example, after a trigger event is detected (operation 302), the detection module 116 of the first device 102A sends a sound sample request to one or more other electronic devices 102 present in the environment 100 (operation 303). In one example, the sound sample request may be an RF message sent via the local wireless network 108 and may include a type indication of the reason for the request (e.g., request type = looking for a device to play audio). In some examples, the sound sample request may be broadcast (e.g., via one or more access points or routers of the local wireless network 108) to all other devices 102 currently registered and active within the local wireless network 108. In some examples, the sound sample request may be sent to one or more specific other devices 102 known to the first device 102A.

[0045] In the current example, the second electronic device 102B receives a sound sample request via a local wireless network 108. The sound sample request is passed to the detection module 116 of the second electronic device 102B, and in response, the detection module 116 causes the device to play one or more predefined sound samples 400 through its speaker 112 (operation 304). The sound samples 400 can take many different forms in different applications. For example, each sound sample 400 can be a tone pulse, chirp, a combination of pulses and chirps, a Zadoff-Chu sequence, or other coded signal sequences. In at least some examples, the device 102 can be pre-associated with unique predefined sound samples during the configuration phase, enabling the receiving device to identify the device sending the sound sample. For example, each device 102 can be assigned a unique sound sample waveform. For ease of illustration, Figures 4A and 4B show exemplary waveforms of predefined sound samples 400 that the electronic device 102B can send, illustrating frequency versus time and amplitude versus time diagrams, respectively. In an exemplary implementation, a predefined sound sample 400 assigned to each electronic device 116 is configured to minimize interference with normally audible human sounds, while ensuring that the frequency of the sound sample 400 falls within a sound frequency range that can be generated by the speaker of a typical COTS electronic device and measured by the microphone of a typical COTS device. For example, if the COTS device includes a microphone supporting a 48 kHz sampling rate, the predefined frequency range for the sound sample 400 can be approximately between 20 kHz and 24 kHz. If the COTS device includes a microphone with a lower sampling rate, the predefined frequency range for the sound sample 400 can be approximately between 17.5 kHz and 20 kHz.

[0046] In this regard, in the illustrated example, the predefined sound sample 400 falls within or near the ultrasonic range, which is at or above the upper limit of human audible sound. Sound signals falling within or near the ultrasonic range tend to have relatively short bandwidths, thus attenuating very quickly in the air medium and reflecting from many different types of surfaces. These properties make such sound signals ideal for detecting electronic devices in the same space. In the illustrated non-limiting example, the predefined sound sample 400 includes a fade-in tone portion 402 within a fade-in duration (Tin), followed by a chirp portion 404 within a chirp duration (Tc), and finally a fade-out tone portion 406 within a fade-out duration (Tout). The fade-in tone portion 202 has a constant frequency (e.g., 23.2 kHz) and a linearly increasing amplitude (e.g., from zero volume to the maximum volume of the sound sample) during its duration (Tin) (e.g., Tin = 10 ms). The chirping tone portion 204 exhibits a linear frequency variation (e.g., increasing from 21.8 kHz to 22.6 kHz) while maintaining a constant amplitude (e.g., the maximum volume of the sound sample) over its duration (Tc) (e.g., Tc = 25 ms). The fade-out tone portion 206 exhibits a constant frequency (e.g., 23.2 kHz) while linearly decreasing amplitude (e.g., decreasing from the maximum volume of the sound sample to zero volume) over its duration (Tout) (e.g., Tout = 10 ms). In the illustrated example, the total sound sample duration (Tsd) is 45 ms.

[0047] In some examples, the second electronic device 102B responds to a sound sample request by playing a periodic sequence of a predefined number of sound samples 400, wherein the consecutive sound segments 402 are separated by null values ​​or interval durations (Tgap). In a particular illustrative and non-limiting example, the interval duration (Tgap) between the individual sound samples is Tgap = 35 ms, and the sequence includes 3 sound samples.

[0048] The sound sample format described above is illustrative. In some examples, different waveform configurations, frequencies, and numbers of sound samples may be used, in addition to those described above. Furthermore, as mentioned above, device-specific sound samples 400 with unique waveforms (with unique segments 402) can be assigned to different electronic devices 102 to achieve device differentiation. For example, sound samples 400 associated with the second device 102B and the third device 102C may have the following corresponding waveform attributes: Table 1: Waveform Attributes of Different Transmitting Devices

[0049] Refer again to Figure 1 and Figure 3In addition to sending a sound sample request, the first device 102A also uses its detection module 116 to begin recording the received sound record 400R via its microphone 118 for a duration equal to (or longer than) the duration (Tsd) of the transmitted sound sample 400 (operation 306). In the example where the sound sample 400 is part of a sequence of a predefined number of sound samples (e.g., a sequence of N sound samples 400), the recording duration of the received sound record 400R can be set to the expected length of the sound sample sequence, for example, N×Tsd×Tgap.

[0050] Then, the received sound recording 400R is processed by a set of processing operations 307 to determine: (i) position data indicating the relative position of the second device 102B and the first device 102A, and (ii) identification data indicating the identifier of the second device 102.

[0051] In the illustrated example, processing operation 307 may include operations 308 through 316. A bandpass filter (operation 308) is applied to the received sound recording 400R to extract an audio signal falling within a near-ultrasound / ultrasound bandwidth (e.g., approximately 17.5 kHz to approximately 22 kHz in a non-limiting example), wherein this audio signal corresponds to the transmitted audio sample 400. Then, matched filtering (operation 310) is performed on the extracted audio signal to extract a time series of audio segments matching the transmitted audio sample 400. For example, matched filtering can be implemented by associating audio segments within the received sound recording 400R with known possible waveforms of the transmitted audio sample 400. In at least some examples, when the environment includes multiple electronic devices 102, wherein each electronic device 102, in response to a sample request from a first device 102A, transmits a corresponding sequence of one or more unique audio samples 400, the received sound recording 400R may be a combination of the unique audio samples 400 from the multiple electronic devices 102. In some examples, matched filtering (operation 310) can be applied based on a pre-assigned waveform pattern to extract the corresponding time series of the received sound samples for each transmitting electronic device 102. In such cases, the following combination... Figure 6 The remaining operations (e.g., operations 312, 313, and 316) can be performed separately for each extracted sound sample time series, and the receiving first device 102A will be aware of the identifier of the corresponding transmitting device for each extracted time series. In some alternative examples, each participating sound sample transmitting device 102B, 102C, 102D can use the same waveform for its corresponding sound sample, but be allocated different time slots to transmit its corresponding sound sample.

[0052] Referring to Figure 1, taking the recording of received sound record 400R to capture sound sample 400 transmitted by second device 102B as an example, it should be noted that due to the multipath effect caused by sound reflections within environment 100, the received sound record 400R actually includes a combination of multiple received versions of the transmitted sound sample 400. For example, the direct or LOS sound propagation path 120B (as shown by the solid line) and the indirect or non-LOS path 122B (as shown by the dashed line) of sound sample 400 between second device 102B and first device 102A are shown. The time series of extracted sound sample 400 generated from the received sound record 400R by matched filtering includes multipath results. For example, Figure 5 shows an example of the time series 500 output by matched filter 310 (as extracted from the received sound record 400R). The extracted time series 500 represents the multipath version of sound sample 400 received by first device 102A (as played by second device 102B).

[0053] In an exemplary embodiment, the time series generated by the matched filtering operation 310 is processed to select a portion of the time series representing received sound samples that have traversed the shortest path (shortest path selection operation 312). In a particular illustrated example, time series 500 is processed to identify a first indexed sound segment present within the extracted time series 500. In the example illustrated in FIG5, the first indexed sound segment (represented by a local amplitude value peak 504) represents sound sample 400 received via the shortest sound propagation path (e.g., LOS propagation path 120B) between a transmitting electronic device (e.g., second device 102B) and a receiving electronic device (e.g., first device 102A). In an exemplary embodiment, the shortest path selection employs a heuristic search technique to identify the first indexed sound sample.

[0054] In one example, shortest path selection may include the following operations: (i) First, identify the maximum amplitude peak within the time series 500 of the matched filter output (in the example illustrated in Figure 5, peak 506 is identified as the maximum amplitude peak; note that in various scenarios, the maximum amplitude peak may correspond to the shortest path or the strongest reflection). (ii) Second, identify whether there exists an amplitude peak that meets a specified amplitude peak criterion and is located within a specified search range 502 preceding the maximum amplitude peak 506. In the illustrated example, the specified peak criterion is a threshold amplitude, which is the product of the maximum amplitude value (e.g., peak 506) and a predefined coefficient value (e.g., 0.4, but other values ​​may be used based on analysis of historical results). The search range 502 may be set to the duration of the multipath representation expected to include the transmitted sound samples. (iii) Third, if the search range 502 includes one or more amplitude peaks that satisfy the specified amplitude peak criterion (e.g., peak 504 in the illustrated example), then the peak immediately preceding the largest peak (e.g., peak 506) (e.g., peak 504) is selected to represent the first indexed sound segment corresponding to the shortest path. If the search range 502 does not include any amplitude peaks that satisfy the specified amplitude peak criterion, then the largest peak itself is selected to represent the first indexed sound segment corresponding to the shortest path.

[0055] As an illustration, in the example of Figure 5, the peak value 504 is identified to represent the first indexed audio segment corresponding to the shortest path (LOS path in this example).

[0056] Once the first index segment is selected, a corresponding subset 510 of the search range 502 can be extracted for further evaluation. Subset 510 can be, for example, based on a predefined period of sound sample 400 and the time position of the selected first index sound segment. In some examples, subset 510 can be selected to include the first index sound segment and the duration of the maximum peak (e.g., a time period extending from the amplitude peak 504 of the first index segment to the maximum index amplitude peak 506). In some examples, subset 510 can be equal to the search range 502.

[0057] Understandably, in at least some scenarios, identifying the subset 510 in the matched filter output that corresponds to the sound sample 400 that has passed through the shortest propagation path does not definitively determine whether the identified subset 510 corresponds to a LOS path. For example, even in the absence of a LOS path, multiple non-LOS paths may still exist, and one of these non-LOS paths may be identified and selected as the shortest propagation path.

[0058] Refer again Figure 3Once the shortest path selection (operation 312) has been performed to identify a subset 510 of the time series of time series 500, a set of features can be extracted from subset 510 (operation 314). For example, this set of features may include one or more of the following: a. Select the maximum amplitude value included in subset 510; b. Select the average amplitude value of subset 510; c. Select the standard deviation amplitude value of subset 510; d. Select the kurtosis amplitude value of subset 510; e. Select the skewness amplitude value for subset 510; f. Select the 25th percentile amplitude value of subset 510; g. Select the 75th percentile amplitude value of subset 510; h. Select the root mean square amplitude value of subset 510; i. Select the proportion of sampled amplitude values ​​within subset 510 that are greater than the product of the coefficient value (e.g., 0.1, 0.2) and the maximum value (a.); j. The sum of a specified number (e.g., 9) of local maximum amplitude peaks that appear after the first index amplitude peak within the selected subset 510; k. The time offset between the first and last amplitude peaks of the selected subset 510; l. Select the average amplitude value of the peak amplitudes included in the subset 510; m. The standard deviation of the peak amplitude values ​​included in the selected subset 510; n. The kurtosis value of the peak amplitude included in the selected subset 510; o. Select the skewness value of the amplitude peak value included in subset 510; p. The 25th percentile of the amplitude peak value included in the selected subset 510; and / or q. Select the 75th percentile of the amplitude peak values ​​included in subset 510. The extracted features are then provided as input to a classification operation (operation 316), which outputs a result indicating the relative position of the transmitting electronic device (e.g., second device 102B) and the receiving electronic device (e.g., first device 102A). In the illustrated example, the relative position is one of two possible states for the transmitting electronics (e.g., second device 102b): (a) relative to the receiving electronic device (e.g., first device 102A). lie in Within the same space; or (b) with a receiving electronic device (e.g., first device 102A). Not located Within the same space.

[0059] In some exemplary embodiments, the classification operation 316 is performed based on a set of predefined rules, which can be determined, for example, based on expert statistical analysis of features extracted from many different real and / or simulated use case scenarios. In some exemplary embodiments, the classification operation 316 can be performed using a trained artificial intelligence model trained using training datasets derived from multiple real and / or simulated use case scenarios to distinguish between "..." lie in Scenes within the same space and Not located Scenes within the same space.

[0060] The classification result ((a) within the same space or (b) not within the same space) is used to determine (decision operation 318) the operation process of the first device 102A (e.g., when the classification result is (a), operation A 320 is executed, and when the classification result is (b), operation B 320 is executed). For example, if the triggering event of process 300 is caused by user input (a request to stream a song (or other audio media content) through the first device 102A to output sound through an external device), operation A 320 may be: causing the song (or other audio media content) to be automatically streamed for playback through the speaker 112 of the second device 102B, and operation B 320 may be: causing the user interface of the first device 102A to generate an output informing the user that the external speaker is unavailable. In this respect, the operation includes: when the classification classifies the physical location of the second device 102B as not being within the same space as the first device 102A, causing the first device 102A to generate a notification output indicating that the second device 102B does not exist.

[0061] In some exemplary embodiments, the classification operation 316 is performed based on a set of predefined rules, which can be determined, for example, based on expert statistical analysis of features extracted from many different real and / or simulated use case scenarios. In some exemplary embodiments, the classification operation 316 can be performed using a trained artificial intelligence model trained using training datasets derived from multiple real and / or simulated use case scenarios to distinguish between "..." lie in Scenes within the same space and Not located A scene "within the same space".

[0062] Having provided the basic examples, the following sections will describe other configurations and use cases for device detection methods, systems, and computer media based on the basic examples provided above.

[0063] in this regard, Figure 6A flowchart of an exemplary same-space detection process 600 is shown, which is built upon process 300 and can be executed for a first electronic device (e.g., smartphone device 102A), a second electronic device (e.g., smart speaker device 102B), a third electronic device (e.g., desktop device 102C), and a fourth electronic device (e.g., smart TV device 102D) within the context of environment 100 shown in Figure 1. In the illustrated example, the same-space detection process 600 is executed in response to a trigger event (operation 302) corresponding to user input from the first device 102A requesting to share content with other smart devices or project content to other smart devices. For example, Figure 7A shows an example of the first device 102A displaying a "Quick Settings" graphical user interface (GUI) panel including "Share" and "Project" buttons. The trigger event in operation 302 can correspond to the user selecting one of the "Share" and "Project" buttons.

[0064] Refer again Figure 6 In response to a triggering event, the first device 102A initiates a sound sample request, which is sent as an RF message to other devices (devices 102B, 102C, 102D) connected to the local area network 108. In some examples, the sound sample request can be implemented through an intelligent network control module 602, which is connected to the local area network 108 and can be deployed on one or more electronic devices 102 or other devices. For example, the detection module 116 of the first device 102A can enable the sound sample request to be provided to the intelligent network control module 602 through the local area network 108, which then distributes the request to devices connected to the local area network 108 that have the technical capability to participate in content sharing / projection.

[0065] Participating devices 102B, 102C, and 102D each respond to a sound sample request (operation 304) by playing corresponding sound samples 400B, 400C, and 400D using their respective speakers 112. In some examples, as described above, devices 102B, 102C, and 102D are each assigned corresponding sound samples 400B, 400C, and 400D, which have unique waveforms, thus enabling differentiation between different sound samples 400B, 400C, and 400D. In some alternative examples, each participating device 102B, 102C, and 102D may use the same waveform for its corresponding sound samples 400B, 400C, and 400D, but is assigned different time slots to transmit its corresponding sound samples. For example, control module 602 can allocate an initial sound sample time slot for the second device 102B at time T (duration Tsd), allocate the next sound sample time slot for the third device 102C at time T+Tsd (duration Tsd), and allocate other sound sample time slots for the fourth device 102D at time T+2Tsd (duration Tsd). Control module 602 can also inform the first device 102A of the time slot order allocated to devices 102B, 102C, and 102D, thereby distinguishing the time slots between devices.

[0066] While the second device 102B, the third device 102C, and the fourth device 102D transmit sound samples 400B, 400C, and 400D respectively, the first device 102A uses its microphone 114 to record the received sound record 400R (operation 306A). Then, a bandpass filter is applied to the received sound record 400R (operation 308) to extract the sound signals corresponding to the bandwidths of the transmitted sound samples 400B, 400C, and 400D. Next, matched filtering is performed on the extracted sound signals (operation 310) to extract the corresponding time series 500B, 500C, and 500D, respectively, corresponding to the transmitted sound samples 400B, 400C, and 400D. For example, matched filtering can be achieved by associating sound segments within the received sound record 400R with known waveforms of the transmitted sound samples 400B, 400C, and 400D. In the example of distinguishing sound samples 400B, 400C, and 400D by differentiating waveforms, the received sound record 400R is a combination of received versions of the transmitted sound samples 400B, 400C, and 400D, all of which are included within a common duration including the sound sample duration Tsd. Related techniques can be used to extract each time series from the independent time series 500B, 500C, and 500D. In the example of distinguishing sound samples 400B, 400C, and 400D by differentiating time slots, the received sound record 400R includes received versions of the transmitted sound samples 400B, 400C, and 400D, all of which fall within consecutive time slots with a duration approximately equal to the sound sample duration Tsd, such that the duration of the received sound record 400R is greater than 3Tsd. Related techniques can be used to extract each time series from the corresponding time slots of the independent time series time series 500B, 500C, and 500D.

[0067] Understandably, the waveform differentiation method can shorten the recording time during audio recording operation 306 at the receiving device 102A because all unique waveform sound samples can be transmitted simultaneously. However, the matched filtering operation 310 requires more complex correlation operations to differentiate the corresponding waveforms. In contrast, the time slot differentiation method may require a longer recording time in audio recording operation 306, but it does not require as complex correlation operations in matched filtering operation 310 because waveform sound samples 400B, 400C, and 400D can be processed independently, and only a single waveform configuration needs to be matched. The choice of a suitable method can be a configuration decision, depending on the intended application, the number of devices involved, and the nature of environment 100.

[0068] like Figure 6As shown, each of the corresponding time series 500B, 500C, and 500D can be processed independently through corresponding processing channels, which respectively apply shortest path selection, feature extraction, and classification operations 312, 315, and 316 in the manner described above. Referring to Figure 1, the time series 500B extracted from the received version of sound sample 400B includes multipath results corresponding to the LOS propagation path 120B and multiple non-LOS paths 122B. Similarly, since the third device 102C is located in the same space 130 (e.g., room A) as the first device 102A, the received version of sound sample 400C includes multipath results corresponding to the LOS propagation path 120C and multiple non-LOS paths (not shown). However, the fourth device 102D is not located in the same space 130 as the first device 102A, but in a different space 132 (e.g., room B), and there is no LOS propagation between the fourth device 102D and the first device 102A. Therefore, the received version of sound sample 400D includes only multipath results corresponding to one or more non-LOS propagation paths 122D, and does not include any LOS sound segments.

[0069] Therefore, in the example of Figure 1, the classification result of the extracted time series 500B is that the second device 102B and the first device 102A are located in the same space; the classification result of the extracted time series 500C is that the third device 102C and the first device 102A are located in the same space; and the classification result of the extracted time series 500D is that the fourth device 102D and the first device 102A are not located in the same space.

[0070] like Figure 6 As shown, the classification results can be processed by decision operation 318 to select the operation to be performed by the first device 102A. For example, if only one projecting device (e.g., the third device 102C) is classified as being in the same space as the first device 102A, the first device 102A will perform operation A 320, which includes automatically sharing or projecting content to the identified “same space” device (e.g., the third device 102C) without any further user interaction. This operation A 320 is shown in Figure 7B, which shows the first device 102A displaying a GUI including a bottom panel indicating whether to connect “Device C – Desktop” for sharing or projection (based on the initially selected GUI button).

[0071] If no device is classified as being located in the same space as the first device 102A, the first device 102A will perform operation B 322, which may include, for example, as shown in FIG7C, displaying a list of all transmitting devices (e.g., the second device 102B, the third device 102C, and the fourth device 102D) and an indication that no device is located in the same space or room as the first device 102A.

[0072] If no device is classified as being located in the same space as the first device 102A, the first device 102A will perform operation B 322, which may include, for example, as shown in FIG7C, displaying a list of all transmitting devices (e.g., the second device 102B, the third device 102C, and the fourth device 102D) and an indication that no device is located in the same space or room as the first device 102A.

[0073] If only one device is categorized as residing in the same space as the first device 102A (e.g., as in the scenario of FIG. 1), the first device 102A will perform operation C 324. This may include, for example, displaying a list of all sending devices (e.g., the second device 102B, the third device 102C, and the fourth device 102D), as shown in FIG. 7D. Devices categorized as residing in the same space as the first device 102A will be identified as such (e.g., the second device 102B and the third device 102C are listed as "same room" devices in the illustrated example), while any device categorized as not residing in the same space as the first device 102A will be identified as such (e.g., the fourth device 102D is listed as "different room"). In an exemplary embodiment, the user selecting a device from the displayed list will cause the operation associated with the initially selected button (e.g., share or project) to be performed using the selected device.

[0074] The following is combined Figure 8 The same spatial detection process 800 describes another exemplary scenario. Figure 8 The same spatial detection process 800 is essentially the same as the same spatial detection process 600, except for the trigger condition operation 302 and the classification decision operation and corresponding operations. In the current scenario, the first device 102A is initially located in room B 132 of environment 100 along with the fourth device 102D, and is currently in the process of sharing or projecting content to the fourth device 102D for playback or display. The first device 102A (in the current example, a smartphone) includes an internal IMU 210 (see...). Figure 2The internal IMU 210 enables the processor system 110 of the first device 102A to estimate the amount of movement of the first device 102A. For example, the first device 102A may include a step tracking module that estimates the number of steps taken by a user carrying the first device 102A over a period of time. It is understood that the amount of movement of the first device 102A exceeding a threshold amount may be an indication that the first device 102A has left its previous location in a space or room (e.g., in the context of FIG. 1, the first device 102A may have moved from room B 132 to room A 130). Therefore, in an exemplary embodiment, the first device 102A is used to monitor a trigger event (operation 302) that corresponds to a situation where the amount of movement of the first device 102A exceeds a specified threshold when sharing or projecting content to another device 102 (e.g., a fourth device 102D). In one example, such a triggering event could be: based on data collected by IMU 210, determining that the number of steps taken by the user carrying the first device 102A has exceeded a threshold number, indicating that the first device 102A may have left the space where it began sharing or projecting with an external device. In this scenario, upon detecting a triggering event corresponding to excessive movement, a same-space detection process 600 can be triggered to perform a same-space check. This check can be used to determine whether the first device 102A is still in the same room as the device it was currently sharing or projecting to. If the first device 102A is classified as no longer in the same room as the device it was currently connected to for sharing or projecting (e.g., the fourth device 102D), other possible sharing / projection device options are identified.

[0075] When a movement-based triggering event is detected, the subsequent operations of the same spatial detection process 800 are the same as those of the same spatial detection process 800, until one or more classification operations 316 are performed.

[0076] Specifically, in process 800, the “same space” classification result is analyzed at decision box 318' to determine which of several possible operations (e.g., operation A 320'; operation B 322' or operation C 324') should be performed. In one example, the first step in deciding operation 318' is to determine whether the classification result regarding the electronic device 102 (e.g., the fourth device 102D in the current example) initially connected to the first device 102A for sharing or projection activities indicates that the first device 102A is still in the same space as that device (e.g., the first device 102A is still in room B 132 with the fourth device 102D). If so, operation A 320' is selected, which is equivalent to maintaining the status quo and continuing to use the fourth device 102D for sharing or projection. Decision operation 318' can also be used to determine whether the classification result indicates that the first device 102A is no longer in the same space as the fourth device 102D, and is not in the same space as any other suitable device (e.g., no, and there are no other alternative devices). In this case, operation B 322' is performed, which may include pausing the sharing or projection operation and displaying the GUI list of devices “not in the same space but in the same network” as shown in FIG. 7C. Decision operation 318' can also be used to determine whether the classification result indicates that the first device 102A is no longer in the same space as the fourth device 102D, but one or more other devices 102 are in the same space as the first device 102A (e.g., no, but there are other alternative devices). In this case, operation C 324' is performed, which may include pausing the sharing or projection operation and displaying the GUI list of devices “in the same space” as shown in FIG. 7D, thereby enabling the user to select an alternative device to continue the sharing or projection activity.

[0077] The following is combined Figure 9 To explain another exemplary scenario, Figure 9 Another example of the same spatial detection process 900, similar to process 300, is shown, but differences can be seen from the following description. Figure 9 The exemplary scenario can represent a conference room collaboration example. For example, a user carrying a first device 102A enters a conference room that includes other devices, such as device 102D (e.g., a smart TV). The user wishes to share content from their first device 102A using a fourth device 102D. In the illustrated example, triggering event 302 could be detecting that the user selects the "share" button on the first device 102A. In the illustrated example, the smart TV device 102D is used to periodically play sound samples 400 without any prompting. The sound segments 402 within the sound samples 400 are encoded with a universally unique identifier (UUID).

[0078] Upon detecting trigger event 302, the first device 102A begins recording the received sound record 400R for a sufficient duration to capture at least one transmission of the sound sample 400 by device 102D. The first device 102A processes the received sound record 400R through a set of processing operations 307 as described above. The final classification result is processed by decision operation 318 to select an appropriate operation: if the classification result indicates that the first device 102A and device 102D are in the same space, operation A 320'' is selected; otherwise, operation B 322'' is selected. Operation A 320'' causes the first device 102A to exchange information to establish a connection, thereby achieving the intended sharing. For example, operation B 322'' may include displaying a GUI message on the first device 102A indicating that the sharing request failed because the devices were not detected as being in the same space, and may also provide the user with an option for the first device 102A to loop back to the sound sample recording operation 306.

[0079] As can be clearly seen from the above description, the device detection method, system, and computer medium based on active sound perception disclosed in this paper can be implemented using standard COTS equipment, and have higher cost-effectiveness, lower resource consumption, and stronger robustness than known detection schemes.

[0080] Although the present invention describes methods and processes by steps performed in a certain order, one or more steps in the methods and processes may be omitted or modified as appropriate. Where appropriate, one or more steps may be performed in an order other than that described.

[0081] While the invention has been described at least partially in terms of method, those skilled in the art will understand that the invention is also directed to various components for performing at least some aspects and features of the method, whether by hardware components, software, or any combination thereof. Accordingly, the technical solutions of the invention can be embodied in the form of a software product. Suitable software products can be stored in pre-recorded storage devices or other similar non-volatile or non-transitory computer-readable media, including DVDs, CD-ROMs, USB flash drives, removable hard drives, or other storage media. The software product includes instructions tangibly stored thereon, which enable a processing device (e.g., a personal computer, server, or network device) to perform examples of the methods disclosed herein.

[0082] The invention may be embodied in other specific forms without departing from the subject matter of the claims. The exemplary embodiments described are merely illustrative in all respects and not restrictive. Features selected from one or more of the foregoing embodiments may be combined to create alternative embodiments not explicitly described, and features suitable for such combinations will be understood within the scope of the invention.

[0083] All values ​​and sub-ranges within the scope of the disclosure are also disclosed. Furthermore, while the systems, devices, and processes disclosed and shown herein may include a specific number of elements / components, the systems, devices, and components may be modified to include more or fewer such elements / components. For example, while any element / component disclosed may be referenced as a single quantity, embodiments disclosed herein may be modified to include multiple such elements / components. The subject matter described herein is intended to cover and encompass all appropriate technical changes.

[0084] The terms "substantially" and "approximately" as used in this invention may indicate that the feature, parameter, or value does not need to be precisely implemented, but deviations or variations may occur, including tolerances, measurement errors, measurement accuracy limitations, and other factors known to those skilled in the art, the amount of which will not affect the intended effect provided by the feature. For illustration, in some examples, the terms "substantially" and "approximately" may refer to a range within 5% of the feature.

[0085] The statement that the second term is "based on" the first term, as used in this document, can indicate that a property of the second term is at least partially influenced by or determined by a property of the first term. The first term can be viewed as an operation or computation, or as the input to a series of operations or computations, which produces the second term as an output, and the second term is related to the first term.

Claims

1. A computer-implemented method for predicting a device location, the method comprising: recording a sound recording using a microphone on a first device; processing the sound recording to extract a feature corresponding to a multipath version of a sound sample played by a second device; based on the extracted feature, classifying a physical location of the second device as one of: (i) within a same space as the first device, or (ii) not within the same space as the first device; based on the classification, performing an operation on the first device.

2. The method of claim 1, wherein, the sound sample is in a frequency range that is inaudible to humans.

3. The method according to claim 1 or 2, characterized in that, processing the sound recording comprises: performing bandpass filtering to obtain a sound signal within a specified frequency band; performing matched filtering on the sound signal to extract a time series corresponding to the multipath version of the sound sample played by the second device; extracting the feature from the time series.

4. The method of claim 3, wherein, comprises: prior to extracting the feature, processing the time series to identify a first indexed sound segment within the time series, wherein the first indexed sound segment corresponds to a shortest sound propagation path from the second device to the first device for the multipath version of the sound sample, the feature is extracted based on properties of a subset of the time series, the subset is selected based on the first indexed sound segment.

5. The method of claim 4, wherein, the feature comprises one or more of: a. a maximum amplitude magnitude value included within the selected subset; b. an average amplitude magnitude value of the selected subset; c. a standard deviation amplitude value of the selected subset; d. a kurtosis amplitude value of the selected subset; e. a skewness amplitude value of the selected subset; f. a 25th percentile amplitude value of the selected subset; g. a 75th percentile amplitude value of the selected subset; h. a root mean square amplitude value of the selected subset; i. a number of sample amplitude values within the selected subset that are greater than a product of a specified coefficient value and the maximum amplitude magnitude value; j. a sum value of a specified number of amplitude peak values occurring within the selected subset; k. a time offset between a first occurring amplitude peak value and a last amplitude peak value of the selected subset; l. an average magnitude value of amplitude peak values included within the selected subset; m. a standard deviation value of amplitude peak values included within the selected subset; n. a kurtosis value of the amplitude peak values included within the selected subset; o. a skewness value of the amplitude peak values included within the selected subset; p. a 25th percentile value of the amplitude peak values included within the selected subset; q. a 75th percentile value of the amplitude peak values included within the selected subset.

6. The method according to claim 4 or 5, characterized in that, Processing the time series to identify the first index sound segment includes: (i) identifying a maximum amplitude peak within the time series; (ii) identifying whether any amplitude peak meets specified amplitude peak criteria and is within a specified search range prior to the maximum amplitude peak; (iii) if one or more amplitude peaks are identified within the specified search range, selecting the amplitude peak immediately preceding the maximum amplitude peak to identify the first index sound segment, and if no amplitude peaks are identified within the specified search range, selecting the maximum amplitude peak to identify the first index sound segment.

7. The method according to any one of claims 1 to 6, characterized in that, Categorizing the physical location of the second device includes applying an artificial intelligence model, wherein the artificial intelligence model has been trained to categorize the physical location of the second device as one of: (i) within the same space as the first device, or (ii) not within the same space as the first device.

8. The method according to any one of claims 1 to 7, characterized in that, Categorizing the second device as being within the same space as the first device corresponds to the second device being physically within the same room of a building as the first device, and categorizing the second device as not being within the same space as the first device corresponds to the second device not being physically within the same room of the building as the first device.

9. The method according to any one of claims 1 to 7, characterized in that, Categorizing the second device as being within the same space as the first device corresponds to the second device and the first device both being physically within a contiguous interior space of a vehicle, and categorizing the second device as not being within the same space as the first device corresponds to neither the second device nor the first device being physically within a contiguous interior space of a vehicle.

10. The method according to any one of claims 1 to 9, characterized in that, The sound sample has a frequency greater than or equal to 17.5 KHz.

11. The method according to any one of claims 1 to 9, characterized in that, The sound sample has a frequency approximately between 20 KHz and 24 KHz.

12. The method according to claim 10 or 11, characterized in that, The sound sample includes a fade-in tone portion, a constant-amplitude chirp portion, and a fade-out tone portion.

13. The method according to any one of claims 1 to 12, characterized in that, Performing the operation on the first device based on the categorization includes causing media content to be automatically streamed for playback through a speaker of the second device when the categorization categorizes the physical location of the second device as being within the same space as the first device.

14. The method according to any one of claims 1 to 12, characterized in that, Performing the operation on the first device based on the categorization includes causing the first device to generate a notification output indicating that the second device is not present when the categorization categorizes the physical location of the second device as not being within the same space as the first device.

15. The method according to any one of claims 1 to 12, characterized in that, Performing the operation on the first device based on the categorization includes causing the first device to establish a connection with the second device to share media content with the second device when the categorization categorizes the physical location of the second device as being within the same space as the first device.

16. The method of any one of claims 1 to 12, wherein, The first device, the second device, and one or more other devices are all associated with a common wireless network, The sound recording includes a received multipath version of the sound sample played by the second device and one or more other sound samples respectively played by the one or more other devices; processing the sound recording includes extracting other features, wherein the other features include respective features corresponding to the one or more other sound samples; the categorizing further includes categorizing a physical location of each of the one or more other devices as one of: (i) the other device being within a same space as the first device, or (ii) the other device not being within the same space as the first device, based on the other features.

17. The method of claim 16, wherein, comprising: the first device sends a request over the common wireless network for the second device to play the sound sample and for the one or more other devices to respectively play a respective one of the one or more other sound samples, wherein the sound sample and the one or more other sound samples have unique waveform properties that enable the sound sample and the one or more other sound samples to be uniquely identified.

18. The method of claim 16 or 17, wherein, the one or more other sound samples each have a frequency greater than or equal to 17.5 KHz.

19. The method of claim 16 or 17, wherein, the one or more other sound samples each have a frequency approximately between 20 KHz and 24 KHz.

20. The method of any one of claims 16-19, wherein, comprising: determining a total number of the second device and the one or more other devices that are physically within a same space as the first device based on the categorizing, wherein performing the operation on the first device is based on the total number.

21. The method of claim 20, wherein, performing the operation on the first device includes: performing the operation on the first device includes: (i) when the total number is 1, causing media content to be automatically streamed or shared for playback by the second device or the one or more other devices that are categorized as being within the same space as the first device, (ii) when the total number is greater than 1, the first device presenting user-selectable device options, wherein the user-selectable device options identify devices that are categorized as being within the same space as the first device, (iii) when the total number is 0, the first device generating an output, wherein the output indicates to a user that no devices are categorized as being within the same space as the first device.

22. A system, comprising: comprising: one or more processors; one or more memories storing machine executable instructions that, when executed by the one or more processors, cause the system to perform the method of any of claims 1-21.

23. A non-transitory processor-readable medium storing machine-executable instructions to perform operations comprising: the machine executable instructions, when executed by one or more processors, cause the one or more processors to perform the method of any of claims 1-21.

24. A computer program, characterized in that, the computer program configures a computer system to perform the method of any of claims 1-21.

25. An apparatus comprising: a computer program for performing the method of any of claims 1-21. a computer program for performing the method of any of claims 1-21.