TW202109076A - Wireless communication with enhanced maximum permissible exposure (mpe) compliance - Google Patents

Abstract

Aspects of the disclosure relate to classifying a target object. An electronic device may transmit a detection signal and receive a reflection signal reflected from the target object. The electronic device then determines, based on one or more features of the reflection signal, a category of the target object and adjusts at least one transmission parameter based on the category. The electronic device then transmits an adjust signal using the transmission parameter. Other aspects, embodiments, and features are also claimed and described.

Description

Wireless communications with enhanced maximum allowable exposure (MPE) compliance

Related matters: This patent application relates to the US patent application No. 62/890,514 filed on the same day as the case and named "WIRELESS COMMUNICATION WITH ENHANCED MAXIMUM PERMISSIBLE EXPOSURE (MPE) COMPLIANCE BASED ON VITAL SIGNS DETECTION", the entire contents of which are incorporated by reference The method is incorporated into this article.

The technology discussed below relates to wireless communication and/or object classification systems as a whole, and specifically, to wireless transmission control based on machine learning and object classification for controlling the maximum allowable exposure. The embodiments can provide and implement technologies for classifying nearby objects and/or target objects (for example, objects detected by wireless proximity sensors or other communication enabling components) and controlling the maximum allowable exposure.

Next-generation wireless telecommunication systems, such as fifth-generation (5G) or new radio (NR) technologies, for example, are deployed using millimeter wave (mmW) signals. These signals can operate in the 28GHz and 39GHz spectrum, for example. Although higher frequency signals provide greater bandwidth to effectively transmit large amounts of information/data, mmW signals may suffer from high path loss (for example, path attenuation). To compensate for the path loss, the transmit power level can be increased, or beamforming can concentrate energy in a specific direction.

As with various types of electronic signal transmission, there are usually regulatory rules governing the intensity of transmission. For example, for mmW signals, the Federal Communications Commission (FCC) and other regulatory agencies set strict RF exposure requirements. These rules ensure that the maximum allowable exposure (MPE) on human skin does not exceed a power density of 1 mW/cm2. In order to meet the target guidelines, electronic equipment is responsible for balancing performance and transmission power and other constraints. The realization of this balancing act can be challenging, especially for devices with cost, size, and other concerns.

The following presents a simplified summary of one or more aspects of the content of this case to provide a basic understanding of these aspects. This summary is not a broad overview of all expected aspects of the content of the case, neither intends to identify the key or important elements of all aspects of the content of the case, nor does it describe the scope of any or all aspects of the content of the case. Its sole purpose is to present some concepts of one or more aspects of the content of this case in a simplified form as a prelude to the more detailed description presented later.

According to some aspects, wireless communication equipment, methods and systems are provided to achieve MPE compliance and/or human target object perception and detection. For example, a device embodiment (for example, a mobile device) may include a wireless communication enabling element (for example, a mmW signal interface). Wireless communication enabling components (for example, transceivers) can not only facilitate wireless communication by receiving and transmitting radio frequency signals (for example, mmW signals), transceivers can also use mmW signal transmission to detect objects. The device embodiment can use the object detection feature to determine the type of the object. If it is determined that the object is human, non-human, animate, inanimate, etc., the device can adjust the operating parameters of the communication interface (for example, mmW transceiver) to comply with MPE (for example, power-on or power-off signal transmission). According to some aspects, signal transmission power adjustments can occur instantly or according to various desired timing arrangements.

In some aspects, the content of this case provides a method for classifying target objects. The method includes sending a detection signal and receiving a reflection signal reflected from the target object. The method also includes: determining the type of the target object based on one or more characteristics of the reflected signal, and adjusting at least one transmission parameter based on the type of the target object. The method also includes sending the adjusted signal using the transmission parameters.

In another aspect, the content of this case provides an electronic device configured to classify target objects. The electronic device includes a processor, a transceiver communicatively coupled to the processor, and a data storage medium communicatively coupled to the processor. Here, the processor is configured to send a detection signal via the transceiver and receive a reflection signal via the transceiver, and the reflection signal is reflected from the target object. The processor is also configured to determine the type of the target object based on one or more characteristics of the reflected signal, and adjust at least one transmission parameter based on the type of the target object. The processor is also configured to send the adjusted signal via the transceiver using the transmission parameters.

In another aspect, the content of this case provides an electronic device configured to classify target objects. The electronic device includes a component for sending a detection signal and a component for receiving a reflection signal, the reflection signal being reflected from a target object. The electronic device also includes a component for determining the type of the target object based on one or more characteristics of the reflected signal, and a component for adjusting at least one transmission parameter based on the type of the target object. The electronic device also includes means for sending adjusted signals using transmission parameters.

In another aspect, the content of this case provides a non-transitory computer-readable medium that stores computer-executable code. The code includes an instruction for the electronic device to send a detection signal and an instruction for the electronic device to receive the reflected signal reflected from the target object. The code also includes instructions for the electronic device to determine the type of the target object based on one or more characteristics of the reflected signal, and instructions for the electronic device to adjust at least one transmission parameter based on the type of the target object. The code also includes instructions for the electronic device to use the transmission parameters to send the adjusted signal.

In another aspect, the content of the present case provides a wireless communication device, including a housing, the shape and size of which are suitable for carrying one or more components, and the components include a memory, a wireless transceiver, a power amplifier, and at least one processor. The wireless transceiver is configured to send and/or receive millimeter wave signals via a wireless channel. The wireless transceiver is also configured to sense objects relative to and outside the housing via millimeter wave signal transmission, and is configured to provide object sensing information to at least one processor. The at least one processor is configured to control the power amplifier based on the object sensing information to adjust transmission parameters associated with the wireless transceiver sending and/or receiving millimeter waves, and is configured to communicate and sense positioning relative to the housing and outside the housing Information associated with the object of.

In another aspect, the content of this case provides a method for providing information to wireless communication devices in a system for providing information between multiple wireless communication devices (information can evaluate the object classification of the object being observed). Here, the method includes configuring the data storage device to perform electrical and wireless communication with one or more unique wireless communication devices among a plurality of wireless communication devices operating in a wireless network. The method also includes receiving from one or more unique wireless communication devices the jog information sent via the wireless network, the jog information including data observation indicating the jog associated with one or more target objects. The method also includes determining object classification information for one or more target objects based at least in part on the received micro-motion information and other stored information. The method also includes sending object classification information to one or more wireless communication devices among the wireless communication devices in the wireless network, so that any one of the wireless communication devices can adjust the wireless communication device associated with its transmission and reception operations. Transmission parameters.

In another aspect, the content of the present case provides a wireless communication device configured as a vehicle, the vehicle including a body configured to carry at least one of a payload or passengers. The vehicle includes a wireless communication interface, the size and shape of which are suitable for being placed close to or inside the vehicle body, wherein the wireless communication interface is configured to send and/or receive millimeter wave signals via a wireless channel. The wireless communication interface is also configured to transmit object sensing with respect to the vehicle body via millimeter wave signals, and is configured to provide object sensing information to at least one processor. The at least one processor is configured to control transmission parameters associated with the wireless communication interface sending and/or receiving millimeter waves based on the object sensing information, and is configured to convey information associated with sensing the object relative to the vehicle body.

In another aspect, the content of this case provides a wireless communication device configured for use in games, wherein the wireless communication device includes a housing, the size and shape of which are suitable for the game, to allow users to participate in the electronic game environment. The wireless communication device includes a wireless communication interface, the size and shape of which are suitable for being placed close to or in the housing, wherein the wireless communication interface is configured to send and/or receive millimeter wave signals via a wireless channel. The wireless communication interface is also configured to sense objects relative to the housing via millimeter wave signal transmission, and is configured to provide object sensing information to at least one processor. The at least one processor is configured to control transmission parameters associated with the wireless communication interface sending and/or receiving millimeter waves based on the object sensing information, and is configured to convey information associated with sensing the object relative to the housing.

These and other aspects of the present invention will be more fully understood by reading the specific embodiments below. By reading the following description of specific exemplary embodiments of the present invention in conjunction with the accompanying drawings, other aspects, features, and embodiments of the present invention will become apparent to those skilled in the art. Although the features of the present invention may be discussed below with respect to certain embodiments and drawings, all embodiments of the present invention may include one or more of the advantageous features discussed herein. That is, although one or more embodiments may be discussed as having certain advantageous features, one or more of such features may also be used in accordance with the various embodiments discussed herein. In a similar manner, although exemplary embodiments may be discussed below as device, system, or method embodiments, it should be understood that such exemplary embodiments may be implemented in various devices, systems, and methods.

The specific embodiments set forth below in conjunction with the accompanying drawings are intended as descriptions of various configurations, and are not intended to represent the only configurations in which the concepts described herein can be practiced. The detailed description includes specific details for the purpose of providing a thorough understanding of various concepts. However, it is obvious to those skilled in the art that these concepts can be practiced without such specific details. In some cases, well-known structures and elements are shown in block diagram form to avoid making these concepts difficult to understand.

Although various aspects and embodiments have been described in this case through the description of some examples, those skilled in the art will understand that other implementations and use cases can be implemented in many different arrangements and scenarios. The innovations described in this article can be implemented across many different platform types, devices, systems, shapes, sizes, and packaging arrangements. For example, the embodiments and/or uses can be implemented by integrating chip embodiments and other equipment based on non-module components (for example, end user equipment, vehicles, communication equipment, computing equipment, industrial equipment, retail/purchasing equipment, medical equipment, Support AI equipment, etc.) to achieve. Although some examples may or may not be specific to use cases or applications, a wide variety of applicability of the described innovations may arise. Implementations can range from wafer-level or modular components to non-modular, non-wafer-level implementations, and further to aggregate, distributed, or OEM equipment that includes one or more aspects of the innovation described Or the scope of the system. In some actual settings, the device including the described aspects and features may also include additional elements and features for implementing and practicing the claimed and described embodiments. For example, the transmission and reception of wireless signals must include multiple components for analog and digital purposes (for example, including antennas, RF chains, power amplifiers, modulators, buffers, processors, interleavers, adders/summers Hardware components). It is intended that the innovations described herein can be implemented in a wide variety of devices, wafer-level components, systems, decentralized arrangements, end user devices, etc. of different sizes, shapes, and configurations.

Electronic wireless communication devices that utilize millimeter wave (mmW) signals can use high transmit power to compensate for the path loss associated with signals at these frequencies. Many of these electronic devices, such as mobile user equipment (UE), can be physically operated by the user. The physical proximity of such electronic devices provides opportunities for radiation to exceed given guidelines, such as the maximum allowable exposure (MPE) limit determined by the Federal Communications Commission (FCC) or other regulatory agencies. Due to these problems, it is advantageous to enable the device to adjust one or more transmission parameters based on the proximity of the user, including but not limited to transmit power.

Some proximity detection technologies can use dedicated sensors, such as cameras, infrared (IR) sensors, radar sensors, etc., to detect users. However, these sensors can be bulky and expensive. In addition, a single electronic device may include multiple antennas located on different surfaces (eg, on the top, bottom, or opposite sides). Considering each of the antennas, according to some aspects, it may be necessary to install multiple cameras or sensors near each of the antennas, which further increases the cost and size of the electronic device.

In another aspect and/or example, the same wireless transceiver used for wireless communication can also perform proximity detection. For example, the local oscillator (LO) circuit in the wireless transceiver can generate one or more reference signals, which can realize proximity detection and wireless communication. The LO circuit can enable the transmission of a frequency modulated continuous wave (FMCW) signal or a multi-tone signal for radar-based proximity detection. By analyzing the reflection from any one of these signals, the distance to the object (for example, the distance or the tilt range) can be determined, and in some instances, the material composition of the object can be determined.

In order to ensure compliance with MPE requirements, according to some aspects, proximity detectors (including but not limited to radar functions based on integrated FMCW) can detect the presence of objects and/or nearby targets. Objects can be located around the device, and some objects can be objects of interest. For example, the detector can determine whether the target is within 20 cm of the radiating element of the device. The detection of multiple objects can be used to create a virtual map of items or topics that have a spatial relationship with the device. Based on such proximity detection, the device can adjust one or more transmission parameters for wireless communication accordingly, such as by reducing the transmission power, by switching to a different transmitting antenna, and so on. By actively measuring the distance to one or more objects, the electronic device can continuously monitor its surrounding environment, and can incrementally adjust one or more transmission parameters to take into account the movement of the object (for example, adjusting mm waves that can be beamforming) Or RF waves usually increase or decrease the transmission power in a specific direction).

Usually, radar signal processing is customized to extract location-based information of the target, such as its distance, speed, angle, position, and so on. However, typical radars do not provide information about the nature of the target, such as whether the target is a living animal/creature, or human. According to the content of this case, it may be advantageous to adjust one or more transmission parameters for wireless communication not only based on the proximity detection of nearby targets, but also based on the classification of the targets. That is, the MPE requirements generally only apply to human exposure. If inanimate objects (such as coffee cups or walls) are close to electronic devices, the MPE requirements may not apply, and high transmission parameters (for example) can continue to be used. Therefore, including information about objects spaced apart in space (for example, nearby objects or objects positioned away from the electronic device), or the classification of detected adjacent objects, can help optimize the power transmission of the electronic device .

FIG. 1 illustrates an exemplary electronic device 102 for implementing a machine learning (ML) algorithm to classify a target object detected by a radar-based proximity detector according to some aspects of the content of the present case. In the example environment 100, the electronic device 102 communicates with the base station 104 via a wireless communication link 106 (wireless link 106). For example, the electronic device 102 and the base station 104 may be part of a system for providing information between a plurality of unique wireless communication devices. In FIG. 1, the electronic device 102 is shown as a smart phone, a vehicle, or a game device to provide some examples. However, the electronic device 102 may be any suitable fixed or mobile device including a wireless transceiver. In the content of this case, the term electronic equipment broadly represents a wide variety of equipment and technologies. The electronic device may include multiple hardware structural elements whose size, shape, and arrangement facilitate communication; these elements may include antennas, antenna arrays, RF chains, amplifiers, one or more processors, etc. that are electrically coupled to each other. For example, some non-limiting examples of mobile devices include mobile devices, cellular (cell) phones, smart phones, SIP phones, laptop computers, personal computers (PC), notebooks, small laptops, Smart computers, tablets, personal digital assistants (PDAs) and various embedded systems, for example, correspond to the "Internet of Things" (IoT). Mobile devices can also be cars or other transportation vehicles, remote sensors or actuators, robots or robotic equipment, satellite radio equipment, global positioning system (GPS) equipment, object tracking equipment, drones, multi-axis aircraft, four Axis aircraft, remote control devices, consumer and/or wearable devices, such as glasses, wearable cameras, virtual reality devices, smart watches, health or fitness trackers, digital audio players (such as MP3 players), cameras, games Computer, game device (for example, an interface that enables users to participate in or play electronic games), etc. Mobile devices can additionally be digital home or smart home devices, such as home audio, video and/or multimedia devices, electrical appliances, vending machines, smart lighting, home security systems, smart meters, augmented reality devices, virtual reality devices, mixed reality Equipment, etc. Mobile devices can additionally be smart energy equipment, security equipment, solar panels or solar arrays, municipal infrastructure equipment that controls electricity (such as smart grids), lighting, and water; industrial automation and enterprise equipment; logistics controllers; agricultural equipment ; Military defense equipment, vehicles, aircraft, ships and weapons. In addition, mobile devices can provide connected medical or remote medical support, that is, remote healthcare. Remote healthcare equipment can include remote healthcare monitoring equipment and remote healthcare management equipment, and its communication can be given priority processing or access priority over other types of information, for example, in priority access for the transmission of key service data And/or related QoS aspects for the transmission of key service data. In some examples, the electronic device 102 may be a wireless communication device that includes a housing that is shaped and sized to carry one or more components, such as those described below.

In one example, the electronic device 102 may be a vehicle or part of a vehicle that includes a body that is configured to carry at least one of a payload or passengers. In this example, the size and shape of the wireless transceiver 120 may be suitable for placement near and/or within the vehicle body. In another example, the electronic device 102 may be a game device or a part of a game device that includes a housing that is sized and shaped to allow a user to participate in an electronic game environment. In this example, the size and shape of the wireless transceiver 120 may be adapted to be placed in a position near the housing. In addition, the gaming device may include a visual interface area, which defines a visual display configured to visually convey object sensing information to the user, and/or one or more user interfaces located near the housing for receiving user input , And in response, convey object sensing information to the user.

The base station 104 communicates with the electronic device 102 via a wireless link 106, and the wireless link 106 can be implemented as any suitable type of wireless link. Although shown as a tower of a cellular network, the base station 104 may represent or be implemented as another device, such as a satellite, a cable TV headend, a terrestrial TV broadcast tower, an access point, an inter-level device, a mesh network node, Small cell nodes, fiber optic lines, etc. Therefore, the electronic device 102 can communicate with the base station 104 or another device via a wired connection, a wireless connection, or a combination thereof.

The wireless link 106 may include a downlink of data or control information communicated from the base station 104 to the electronic device 102 and an uplink of other data or control information communicated from the electronic device 102 to the base station 104. The wireless link 106 can be implemented using any suitable communication protocol or standard, such as the Third Generation Partnership Project Long Term Evolution (3GPP LTE), the fifth generation new radio (5G NR), IEEE 802.11, IEEE 802.16, BluetoothTM Wait. In some embodiments, no data link is provided or in addition to a data link, the wireless link 106 may provide power wirelessly, and the base station 104 may include a power source.

The electronic device 102 includes an application processor 108 and a computer-readable storage medium 110 (CRM 110). The application processor 108 may include any type of processor (for example, an application processor, a digital signal processor (DSP), or a multi-core processor) that executes processor-executable code stored by the CRM 110. The CRM 110 may include any suitable type of data storage media, such as volatile memory (e.g., random access memory (RAM)), non-volatile memory (e.g., flash memory), optical media, magnetic media ( For example, disk or tape) and so on. In the context of the content of this case, the CRM 110 is implemented to store the instructions 112, data 114, and other information and software of the electronic device 102. For example, the CRM 110 may include a memory for storing data, and the data is configured to enable the processor/DSP 128 to process object sensing information for the stored data, so that the target object can be classified into different object types. The CRM 110 may be resident in the application processor 108, external to the application processor 108, or distributed on multiple entities including the application processor 108. CRM 110 can be embodied in computer program products. For example, the computer program product may include the computer readable medium in the packaging material. Those skilled in the art will recognize that how best to implement the functions presented throughout the content of this case depends on the specific application and the overall design constraints imposed on the entire system.

One or more processors 108 can execute software. Software should be broadly interpreted as representing instructions, instruction sets, codes, code fragments, code, programs, subprograms, software modules, applications, software applications, packaged software, routines, subroutines, objects, Executable programs, running threads, procedures, functions, etc., whether called software, firmware, intermediary software, microcode, hardware description language, or others.

The electronic device 102 may also include an input/output port 116 (I/O port 116) and a display 118. The I/O port 116 can realize data exchange or interaction with other devices, networks, or users. The I/O port 116 may include a serial port (for example, a universal serial bus (USB) port), a parallel port, an audio port, an infrared (IR) port, and the like. The display 118 presents graphics of the electronic device 102, such as a user interface associated with an operating system, program, or application program. Alternatively or additionally, the display 118 may be implemented as a display port or a virtual interface through which the graphic content of the electronic device 102 is presented.

The wireless transceiver 120 of the electronic device 102 may include a wireless transmitter 122 and a wireless receiver 124. The wireless transceiver 120 provides a connection to the corresponding network and other electronic devices connected to it. In addition, the electronic device 102 may include a wired transceiver, such as an Ethernet or optical fiber interface, for communicating via a local network, an intranet, or the Internet. The wireless transceiver 120 can facilitate communication via any suitable type of wireless network, such as wireless LAN (WLAN), peer-to-peer (P2P) network, mesh network, cellular network, wireless wide area network (WWAN) and/or wireless Personal Area Network (WPAN). In the context of the example environment 100, the wireless transceiver 120 enables the electronic device 102 to communicate with the base station 104 and the network connected to it.

The wireless transceiver 120 includes circuits and logic for transmitting and receiving signals via the antenna 126. For example, the wireless transceiver 120 may be configured to send and/or receive mmW signals via a wireless channel, and further, through the use of mmW signal transmission, to sense objects relative to the housing of the electronic device 192 and objects outside the housing of the electronic device 192 . The wireless transceiver 120 may be configured to substantially simultaneously participate in mmW communication and mmW object sensing. In addition, the wireless transceiver 120 may be configured to sense objects in a range from approximately 1 degree to approximately 360 degrees, so that the processor/DSP 128 can generate an object sensing map of objects outside the housing of the electronic device 102.

The wireless transceiver 120 includes circuits and logic for transmitting and receiving signals via the antenna 126. The components of the wireless transceiver 120 may include amplifiers, mixers, switches, analog-to-digital converters, filters, etc. for conditioning signals. The wireless transceiver 120 may also include logic to perform in-phase/quadrature (I/Q) operations, such as synthesis, encoding, modulation, decoding, demodulation, and so on. The wireless transceiver 120 may include one or more elements or features for adjusting or controlling transmission parameters. For example, the wireless transceiver 120 may provide object sensing information to the processor DSP 128. In some cases, the components of the wireless transceiver 120 are implemented as separate transmitter 122 and receiver 124 entities. Additionally or alternatively, multiple or different parts may be used to implement the wireless transceiver 120 to implement corresponding transmission and reception operations (for example, separate transmission and reception chains). Although the examples described below generally refer to the integrated wireless transceiver 120 that performs wireless communication and object sensing operations, the various aspects of the content of this case are not limited to this case. For example, the electronic device 102 may include an interface circuit for interfacing with auxiliary and/or auxiliary sensing devices, which is spaced apart from the housing of the electronic device 102. The auxiliary and/or auxiliary sensing device may include a remote wireless device capable of communicating with the electronic device 102 (eg, game controller, wearable device, augmented/virtual reality device, and other types of mobile devices mentioned above). Here, the interface circuit (not shown) can realize the communication between the electronic device 102 and the auxiliary sensing device, so that the electronic device 102 receives object sensing information from the auxiliary sensing device. In response to the object sensing information, the electronic device 102 can adjust the transmission parameters associated with the wireless transceiver 120 sending and/or receiving mmW signals. The adjustment of the transmission power may include controlling and/or modifying the transmission power, such as increasing and/or reducing or otherwise changing the transmission power level.

The electronic device 102 also includes a processor/digital signal processor (DSP) 128 which is coupled to the wireless transceiver 120. The processor/DSP 128, which may include a modem, can be implemented inside the wireless transceiver 120 or separately from the wireless transceiver 120. Although not explicitly shown, the processor/DSP 128 can include part of the CRM 110 or can access the CRM 110 to obtain computer-readable instructions. The processor/DSP 128 controls the wireless transceiver 120 and enables wireless communication or proximity detection. For example, the processor/DSP 128 can control the power amplifier at the transceiver 120 to adjust the transmission parameters based on the object sensing information. The processor/DSP 128 can include a baseband circuit to execute high-rate sampling procedures, which may include analog-to-digital conversion, digital-to-analog conversion, Fourier transform, gain correction, skew correction, frequency conversion, and the like. The processor/DSP 128 can provide communication data to the wireless transceiver 120 for transmission. The processor/DSP 128 can also process signals in the baseband form obtained from the wireless transceiver 120 to generate data, which can be provided to other parts of the electronic device 102 via a communication interface for wireless communication or proximity detection.

In some instances, the electronic device 102 (eg, via the processor/DSP 128) may be configured to generate, output, or generate a map. The map can be based on at least object sensing information. The map can include spatial locations and other information (for example, communication status, operating status, relative object location, direction, movement, type, and/or classification) associated with objects around the electronic device 102. In addition, the map can be used to display and/or identify objects, people, and/or animals relative to the electronic device 102. The map can be time-static or time-dynamic. The map can be used to visually display the complete 360-degree arrangement range around the device and other or smaller ranges (for example, from about 1 degree to about 360 degrees) of objects. Maps enable users to access augmented/virtual reality environments (for example, (electronic) games, (remote) healthcare and/or education). According to some aspects, the map may be provided to the user as an output via a display screen (for example, the display screen shown for the electronic device 102) or other user interface. The electronic device 102 can send the map information to other devices on the network. According to some aspects, sharing map information in this way can help disseminate map and/or spatial positioning information within the network.

Although not explicitly shown, the wireless transceiver 120 or the processor/DSP 128 may also include a controller. The controller can include at least one processor and at least one CRM, such as an application processor 108 and a CRM 110. The CRM can store computer-executable instructions, such as instructions 112. The processor and CRM can be located in one module or an integrated circuit chip or can be distributed on multiple modules or chips. The processor and related instructions together can be implemented in separate circuits, fixed logic circuits, hard-coded logic, and so on. The controller can be implemented as part of a wireless transceiver 120, a processor/DSP 128, a dedicated processor configured to perform MPE technology, a general-purpose processor, some combination thereof, and the like.

The processor/DSP 128 may include a feature extraction circuit 130 and an SVM classification circuit 132. The feature extraction circuit 130 may be used to extract the feature of the reflected signal, which indicates the micro-movement characteristics of the human target. The SVM classification circuit 132 may be used to determine the category or classification of the target object based on the characteristics of one or more extracted reflection signals. For example, the SVM classification circuit 132 may apply the extracted features to the classification model, where the SVM classification circuit 132 determines the position of the target object in the feature space relative to the boundary, which divides the objects in the feature space into categories. Based on the position, the SVM classification circuit 132 can identify the type of the target object, including human tissue, non-human object, or a combination thereof.

In some examples, the base station 104 may be a data storage device, or may communicate with one or more data storage devices that receive jog information from a wireless communication device via a wireless network. In this way, the data storage device can transmit object classification information between wireless communication devices in the network. Based on this information, the data storage device can determine object classification information for one or more target objects based at least in part on the received micro-motion information and in some examples based on other stored information. The object classification information can then be transmitted to other wireless communication devices, so that any wireless communication device in the network can adjust the wireless transmission parameters associated with its transmission and reception operations. In another example, the data storage device may transmit information indicating that any one or more wireless communication devices have adjusted their power transmission levels, for example, based on the classification of the target object.

FIG. 2 illustrates an example operating environment 200 for classifying target objects detected by a radar-based proximity detector. In the example environment 200, the user's hand 214 holds the electronic device 102. In one aspect, the electronic device 102 communicates with the base station 104 by transmitting an uplink signal 202 (UL signal 202) or receiving a downlink signal 204 (DL signal 204) via at least one antenna 126. However, the user's thumb may indicate the approaching target object 206 that may be exposed to radiation via the uplink signal 202. In order to determine the distance to the target object 206, the electronic device 102 transmits the proximity detection signal 208-1 via at least one of the antennas 126, and receives the reflected proximity detection signal 208-2 via at least another antenna of the antenna 126.

In one embodiment, the proximity detection signal 208-1 includes a frequency modulated continuous wave (FMCW) signal 216. In general, the frequency of the FMCW signal 216 increases or decreases over a period of time. Different types of frequency modulation can be used, including linear frequency modulation (LFM) (for example, chirp), sawtooth frequency modulation, triangular frequency modulation, and so on. The FMCW signal 216 enables the use of radar-based ranging technology to determine the distance to the target object 206. In order to achieve finer range resolution for short-range applications (for example, on the order of centimeters (cm)), larger bandwidths, such as 1 gigahertz (GHz), 4 GHz, 8 GHz, etc., can be used. For example, the FMCW signal 216 may have a frequency bandwidth of approximately 4 GHz and include frequencies between approximately 26 and 30 GHz. The finer distance resolution improves distance accuracy and enables multiple objects 206 to be distinguished within a distance. The FMCW signal 216 can provide accurate distance measurements for various distances based on the bandwidth (for example, for a 4 GHz bandwidth, between about 4 and 20 cm). The amount of time to perform proximity detection using the FMCW signal 216 can also be relatively short, for example, within about one microsecond.

In another embodiment, the proximity detection signal 208 may be a multi-tone signal 218, which includes at least three tones (eg, frequencies). The multi-tone signal 218 can be generated using existing components in the wireless transceiver 120 that are also used to generate the uplink signal 202. For example, the multi-tone signal 218 can be generated using an existing phase-locked loop (PLL), using orthogonal frequency division multiplexing (OFDM), or using a multi-tone transmission signal generated at a fundamental frequency via a digital signal generator. Depending on the technology used, the amount of time for performing proximity detection via the multi-tone signal 218 can be on the order of about one microsecond and 400 microseconds. The frequency separation between tones can be on the order of megahertz (MHz) or GHz. The bandwidth of the multi-tone signal 218 can be, for example, about 800 MHz or 2 GHz. The distance of the object 206 is determined by analyzing the phase change of each of the tones. In order to improve the distance accuracy, a larger bandwidth (for example, the interval between tones) or a larger number of tones can be used. The multi-tone signal 218 can be used to measure a distance between approximately 0 and 7 cm.

In some electronic devices 102, antenna 126 may include at least two different antennas, at least two antenna elements 212 of antenna array 210, at least two antenna elements 212 associated with different antenna arrays 210, or any combination thereof. As shown in FIG. 2, the antenna 126 corresponds to the antenna element 212 in the antenna array 210. The antenna array 210 can include a plurality of antenna elements 212-1 to 212-N, where N represents a positive integer. Using at least one of the antenna elements 212, the wireless transceiver 120 can transmit the proximity detection signal 208-1, while using at least another antenna of the antenna element 212 to receive the reflected proximity detection signal 208-2. In other words, the wireless transceiver 120 can receive the reflected proximity detection signal 208-2 via the first antenna element 212-1 during a portion of the time of the proximity detection signal 208-1 transmitted via the second antenna element 212. The antenna 124 and/or its components can be implemented using any type of antenna, including patch antennas, dipole antennas, and the like.

If the electronic device 102 includes multiple antennas 126 on different sides (eg, top, bottom, or opposite sides) of the electronic device 102, the described technique may enable the user to be detected with respect to each antenna 126. In this way, the transmission parameters can be adjusted independently in relation to the distance of the object 206 relative to each antenna 126. Therefore, such independent detection enables two or more of the antennas 126 to be configured for different purposes. For example, one of the antennas 126 may be configured for enhanced communication performance, while the other antenna of the antennas 126 is configured to comply with FCC requirements at the same time. As described in further detail with respect to FIG. 3, some components of the wireless transceiver 120 can be used for wireless communication and proximity detection at different times or at the same time. In some instances, the electronic device can perform radar sensing and proximity detection during the unused time slot of the wireless communication protocol. For example, in mmW communication, the communication frame may include one or more unused time slots for random channel access (RACH); the electronic device 102 can perform radar sensing during these unused RACH time slots And proximity detection.

In some examples, when the electronic device 102 is operating in a wireless communication network, the electronic device 102 can transmit the object transmission data to one or more other devices in the network. In this way, the electronic device 102 and other devices can share the target object sensing data with each other. Therefore, the electronic device 102 and other devices can determine the surrounding target object information of the surrounding objects.

FIG. 3 illustrates a wireless transceiver 120 and a processor/DSP circuit of a machine learning (ML) algorithm for classifying detected target objects using a radar-based proximity detector according to some aspects of the content of this case Example implementation of 128. The wireless transceiver 120 may include a transmitter 122 and a receiver 124, which are coupled between the processor/DSP 128 and the antenna array 210, respectively. The transceiver 120 includes a power amplifier (PA) 302 configured to dynamically provide power to selected antenna elements of the antenna elements 210 for adjusting transmission parameters and/or for beamforming. The transceiver 120 also includes a low noise amplifier (LNA) 304 for amplifying the signal received by the receiving antenna 210-2. Local oscillator (LO) circuit 306 is coupled to mixers 308 and 310. The LO circuit 306 generates at least one reference signal, which enables the mixers 308 and 310 to up-convert or down-convert, respectively, to transmit or receive analog signals in the chain. The LO circuit 306 can also be configured to generate one or more different types of reference signals to support proximity detection and wireless communication. In some examples, the LO circuit 306 may be configured to generate one or more in-phase and quadrature (I/Q) reference signals. In this way, the transmission from transmit antenna 210-1 may include I and Q components. And, after receiving the reflected signal from the receiving antenna 210-2, the I and Q components of the reflected signal may be separated from each other for processing.

The transceiver 120 can also include other additional elements not shown in FIG. 3. The additional components can include bandpass filters, additional mixers, switches, and so on. In addition, as mentioned above, the transceiver 120 can not only be configured for the distance measurement and detection of the target object described below, but also can be configured for wireless communication.

Although not explicitly described, the wireless transceiver 120 and/or the processor/DSP 128 can also include a controller. The controller can include at least one processor and at least one CRM, such as an application processor 108 and a CRM 110. The CRM can store computer-executable instructions, such as instructions 112. The processor and CRM can be located on one module or one integrated circuit chip or can be distributed on multiple modules or chips. The processor and related instructions together can be implemented in separate circuits, fixed logic circuits, hard-coded logic, and so on. The controller may be implemented as part of a wireless transceiver 120, a processor/DSP 128, a special-purpose processor configured to perform MPE technology, a general-purpose processor, some combination thereof, and the like.

The voltage controlled oscillator (VCO) 312 may be configured to generate a sinusoidal signal having a frequency that depends on the voltage of the input signal v(t). That is, by appropriately changing the input signal v(t) to the VCO 312, the VCO 312 can generate, for example, a sine wave whose frequency increases with time, commonly referred to as a chirp signal. The chirp signal can be used for FMCW-based radars. Of course, other suitable input signals v(t) and other suitable radar configurations can be used within the scope of the content of this case for proximity detection and target object sampling.

The chirp signal may be amplified by the PA 302 and mixed with the LO signal at the mixer 308 (ie, upconverted) for transmission from the transmitting antenna 210-1. The transmitted signal may be reflected from the target object 314 and reflected back to the receiving antenna 210-2. The reflected signal at the receiving antenna 210-2 may be mixed with the LO signal at the mixer 310 (ie, down-converted) and amplified by the LNA 304.

The output of the LNA 304 (ie, the amplified received signal) may be mixed with the chirp signal at the mixer 316. With FMCW-based radars, this mixing produces a beat signal, which represents the frequency offset between the radio frequency transmit signal and the radio frequency receive signal. Generally, the frequency of the beat signal is proportional to the distance of the target object 314.

The beat signal may be processed by a baseband circuit 318, which is configured to perform various baseband functions, including but not limited to gain correction, skew correction, frequency conversion, and the like. The output from the baseband circuit 318 can be converted to the digital domain using one or more analog-to-digital converters (ADC) 320. In the example in which the radar transmission includes I and Q components, as described above, the output from the baseband circuit 318 may include separate I and Q signals, and the ADC 320 may include two ADCs for dividing the I and Q components respectively Each component is converted to the digital domain. The digital output from ADC 320 may then be provided to processor/DSP circuit 128. In some embodiments, the processor/DSP circuit 128 may be a DSP or any suitable functional component for executing the following programs.

An undesirable side effect of having tightly positioned transmit antenna 210-1 and receive antenna 210-2 (as may occur in small electronic devices) is mutual coupling (MC). That is, part of the transmitted energy can be coupled back to the receiver. Such mutual coupling is a well-known problem in the art. Within the processor/DSP circuit 128, the MC cancellation circuit 322 can provide cancellation of undesirable energy coupled between the transmit antenna 210-1 and the receive antenna 210-2. In order to remove the MC component from the received signal, the MC canceling circuit 322 uses the transmission signal to cancel the MC component. Although not explicitly shown, MC cancellation can be performed in the time domain or the frequency domain via the MC cancellation circuit 322.

After the MC is eliminated, the Discrete Fourier Transform (DFT) circuit 324 can convert the received beat signal to the frequency domain and provide samples of the beat signal in the domain. For example, if 30 measurements of the target object 314 are obtained from 30 consecutive target object reflections, the output x from the DFT circuit 324 includes x _i = [x ₁ , x ₂ , …, x ₃₀ ] as its output. Here, each sample x _i corresponds to the spectrum measured from a single radar reflection. The samples x _i can then be sent to the feature extraction circuit 326. That is, according to the content of the case, one or more features (for example, M features as shown in the figure) can be extracted from the spectrum of the radar sampling sequence of the target object 314. The extracted features can be used to classify the target object as human or non-human, for example, as described further below. That is, the feature indicating the micro-movement characteristics of the human target object can therefore be used to classify the target object.

Then the M extracted features can be provided to the classification circuit 328. In some examples, the classification circuit 328 may be a support vector machine (SVM) that uses machine learning (ML) to classify the target object. As described further below, the SVM classification circuit 328 may determine the distance of the extracted feature relative to the boundary in the defined feature space. Based on the distance from and/or the position relative to such a boundary in the defined feature space, the SVM classification circuit 328 can then provide a decision to classify the target object 314, for example, as a human or non-human. Also as described further below, based on the classification of the target object 314, the processor/DSP circuit 128 can generate transmission parameters that control one or more transmission attributes for wireless communication. By specifying transmission parameters, the processor/DSP circuit 128 can, for example, cause the transceiver 120 to reduce the transmission power if the target object 314 near the electronic device 102 is a human, or if the target object 314 is far away from the electronic device 102 and/or is not a human, then The transceiver 120 increases the transmission power. For example, the power amplifier 302 may be dynamically controlled based on the target object classification. If it is determined that the target object 314 is not a human being, the processor 122 can, for example, keep the transmission parameters unchanged. The transmission parameters can adjust the power level, beam steering angle, frequency, selected antenna or antenna array, or communication protocol used to transmit uplink signals and/or receive downlink signals. The ability to determine the distance to the target object 314 and the type of the target object 314 and the ability to control the transceiver 120 enables the processor 122 to balance the performance of the electronic device 102 with compliance or radiation requirements.

The processor/DSP circuit 128 may also be coupled to the LO circuit 306, which enables the processor/DSP circuit 128 to control the LO circuit 306 via a mode signal. For example, the mode signal can enable the LO circuit 306 to switch between generating a reference signal for proximity detection or generating a reference signal for wireless communication. In other embodiments, the application processor 108 (see FIG. 1) can perform one or more of these functions.

Although the wireless transceiver 120 is shown as a direct conversion transceiver in FIG. 3, the described technique can also be applied to other types of transceivers, such as superheterodyne transceivers. In general, the LO circuit 306 can be used to perform frequency between any frequency level (for example, between the fundamental frequency and the radio frequency, between the intermediate frequency and the radio frequency, or between the fundamental frequency and the intermediate frequency). Conversion.

FIG. 4 illustrates a series of three graphs generated using an exemplary embodiment of the electronic device 102. Each illustrated figure illustrates 30 consecutive acquisitions of data from reflected radar pulses over a period of 9 seconds, with samples taken at 0.3 second intervals. In each corresponding figure, the horizontal axis represents the time (or sampling index), and the vertical axis represents the distance from the electronic device, as determined by the distance detection algorithm generally described above. In addition, the shading at any given point represents the energy content of the received signal reflected from the target object at the corresponding time and distance from the electronic device. For example, the feature extraction circuit 326 may determine parameters, which include the energy content of the received signal at each target distance, and other parameters.

Graph 402 provides a data set corresponding to stationary non-human target objects (such as coffee cups). As shown in the figure, the static object is characterized by relatively static data between samples. Diagram 404 illustrates a data set corresponding to a moving human hand as a target object. This indicates a significant change in the data over time. Graph 406 illustrates a data set corresponding to a human hand as a target object, in which the human keeps their hands still. Even when people try to remain completely still, they cannot eliminate the fretting caused by small muscle movements, breathing, blood vessel pulses, etc. Because the integrated FMCW-based radar that uses the mmW spectrum has a wavelength of about 1 cm or less, it can detect very small movements in the target object, such as 2 or 3 mm.

By observing data from various objects such as these, the inventor realized that when the detected target object has a human nature, the observed data shows changes or fluctuations in various metrics or characteristics, such as the peak energy of the reflected signal, Sidelobe changes, etc. In addition, by analyzing the fluctuations on the appropriate feature set extracted from the target object, the target object can be reliably and accurately classified as a human or non-human target object. According to various aspects of the content of this case, a machine learning (ML) algorithm is provided to use these and other features to classify target objects. In this way, in some instances, the transmission characteristics can be controlled to dynamically meet the MPE requirements for mmW transmission.

Figure 5 provides two diagrams showing an exemplary two-dimensional (2D) feature space. The figures provide examples of how to combine and use appropriate extracted feature sets to improve the reliability of the classification of target objects detected by radar-based proximity detectors as described above. In the context of the content of this case, the feature refers to the decision parameter of the classification problem that is related or specific to the target object. That is, the electronic device 102 can extract one or more features to determine whether the target object is a human.

In the graph shown in Figure 5, each point corresponds to a sample of data collected from the measurement and post-processing of the target object. In the first graph 502, the horizontal axis represents the variance of the peak power of the reflected signal minus the average power on 30 consecutive radar reflections. The vertical axis represents the maximum value of the discrete Fourier transform (DFT) of the reflected signal power minus the average value of the DFT of the reflected signal power. In the second graph 504, the horizontal axis represents the average value of the phase change (Δ) between sample n and sample n-1 on 30 consecutive radar reflections; the vertical axis represents the average value between sample n and sample n-1. The variance of the phase change (Δ).

In each of these figures, each "×" represents a data point from a human target object, and each "○" represents a data point from a non-human target object. It can be clearly seen that the circle "○" representing the sampling of non-human target objects forms a group in the lower left corner. On the other hand, the "×" representing the sampling of human target objects is scattered throughout the figure. Therefore, in these exemplary illustrations, a simple linear boundary separation can be used to distinguish between human and non-human samples.

FIG. 6 illustrates another example of a 2D feature space, and illustrates an example of how the classifier algorithm uses data from known classified target objects to determine the best separation between target objects of different categories. In this illustration, _{the axes labeled x 1} and x ₂ represent features extracted from the target object. The data points displayed with filled circles (●) correspond to human target objects, and the data points displayed with blank circles (○) correspond to non-human target objects.

The graphs in Figures 5 and 6 illustrate a 2D feature space, comparing two extracted features to obtain a classification between human and non-human target objects. However, the content of this case is not limited to this 2D feature space. In general, due to the higher dimensionality of the feature space, a multi-dimensional comparison between any suitable number of features can be constructed. In this case, instead of the line 602 as the boundary between the classifications, a plane or a hyperplane can be used to separate the classification of the target object in a higher-dimensional feature space. That is, the exemplary classification circuit 328 can use SVM to analyze any suitable number of extracted features from the sample set of the target object, and determine the classification of the target object.

Referring again to Figure 6, it can be observed that data points of different categories can be easily separated from each other. However, in order to best separate the different categories in order to most reliably classify new data from the target object in the future, the classification model should identify the best separation between the categories. For example, an unlimited number of lines, such as line 602, can theoretically completely separate the measurement results in a given data set. However, if the classification model uses the line 604 shown to predict the category of the future target object, the prediction may be unreliable. That is, because the line 604 is close to the group that passes through the human target object being measured, even a small change in the extracted feature of the future measurement result of the human target object may cause the measurement result to fall on the opposite side of the line 604. As a result, the model incorrectly classifies the target object.

In one aspect of the content of this case, machine learning (ML) algorithms can be used to establish a reliable separation between sets of target objects that become different categories (for example, human and non-human target objects). That is, the classifier algorithm can be built by constructing a large data set (for example, training data) that includes many human body parts (for example, hands, arms, faces in different postures, etc.) and many non-human objects commonly encountered by electronic devices .

As an example, the classification circuit 328 may be a support vector machine (SVM), which may be used to construct a target object classifier algorithm. SVM is a well-known ML model for data collection classification in the art. In a nutshell, SVM can be used to analyze a data set and identify the boundary between classes by maximizing the minimum distance between the closest point in the sample set of each class and the boundary. These boundaries are called support vectors.

Refer again to Figure 6, which illustrates the data corresponding to the eight realizations of human target objects (●) and the eight realizations of non-human target objects (○). Here, the realization corresponds to the radar acquisition or observation set received after reflection from the target object based on the radar transmission. According to the content of this case, by selecting the appropriate features x ₁ and x ₂ extracted from each realization, it can be observed that the two types of data can be divided into different groups. In another aspect, the classification circuit 328 (eg, SVM) can be used to calculate the best boundary between classes based on the data.

As mentioned above, in the two-dimensional feature space as shown in FIG. 6, the boundary corresponds to a line. In the aspect of the content of this case, the line can be represented by the value of wx-b = 0. Here, w is a weight vector; x is a vector in the feature space> x ₁ , x ₂ >; and b is a scalar deviation or offset. In this equation, the weight vector w is configured such that the product wx of the two vectors results in a scalar value.

的值。可以將邊界602選擇為在該等相應線之間居中。此處，

表示w的範數，計算為向量的所有元素的平方和的根（在圖示的情況下，

）。並且如圖6中進一步所示，

是邊界602距原點的距離或偏移量。As seen in Figure 6, in the two-dimensional feature space, the separation between the lines parallel to the boundary 602 passing through the closest sample in each class has

Value. The boundary 602 may be selected to be centered between the corresponding lines. Here,

Represents the norm of w, calculated as the root of the sum of squares of all elements of the vector (in the case of the illustration,

). And as further shown in Figure 6,

Is the distance or offset of the boundary 602 from the origin.

Based on the analysis of the data set, the classification circuit 328 (eg, SVM) determines the values of w and b for the best boundary 602 between the classes based on the training data given to it. That is, although the illustration shows a total of 16 samples or realizations, of which 8 are from human target objects and 8 are from non-human target objects, any suitable number of samples can be used as the training data set. In general, the larger and more diverse/variable the training data used is in nature, the more reliable the calculation boundary 602 used to determine the classification of the newly entered sample.

In general, the classification circuit 328 can generate boundaries (eg, lines, planes, or hyperplanes, depending on the number of dimensions in the feature space) to separate samples from different categories. That is, the SVM defines a boundary that separates samples from different categories so that the minimum distance between the samples in each category and the boundary is maximized. Using this boundary can provide a robust way to distinguish different categories.

By predetermining such a boundary, the calculation cost for the electronic device 102 to classify the new target object can be reduced. That is, deep learning algorithms or neural network algorithms can potentially determine the separation between categories in real time. However, the use of this algorithm will come at the cost of high calculation requirements, high power usage, long calculation time, and even higher system costs.

The following provides some examples of features that can be used to distinguish human target objects from non-human target objects. According to aspects of the present invention, the feature extraction circuit 326 can analyze the realization (reflected signal) set from the target object as described above to extract M feature sets corresponding to the target object.

其中X = {x₁ , x₂ , …, x_L }且Y = {y₁ , y₂ , …, y_L }。因此，對於X之每一者取樣和Y之每一者取樣，DTW依賴於DFT域中各個值之間的距離的比較。大體上，兩個非常相似的序列的DTW可能非常小，而兩個極為不同的序列的DTW可能很大。

In some examples, the feature extraction circuit 326 may utilize dynamic time warping (DTW). DTW is an algorithm known in the art for determining the similarity between sequences (for example, sequences X and Y, each with L samples), and is defined according to the following equation:

Where X = {x ₁ , x ₂ , …, x _L } and Y = {y ₁ , y ₂ , …, y _L }. Therefore, for each sample of X and each sample of Y, DTW relies on the comparison of the distance between the various values in the DFT domain. In general, the DTW of two very similar sequences may be very small, while the DTW of two very different sequences may be very large.

_{For example, the feature that the feature extraction circuit 326 can extract is the variance (Var DTW} ) of the DTW over a series of realizations (for example, a series of 10 or any suitable number of realizations). In the above equation, Var _DTW corresponds to the calculated variance of DTW over n realizations. For stationary (for example, non-human) objects, the variance of the determined DTW over a series (for example, a series of 10) realizations will be very small, because each realization will provide basically the same data. On the other hand, for human target objects, due to the movement or micro-movement of the human target objects, a series of (for example, a series of 10) realizations will have obvious differences from each other. Therefore, the variance of DTW in a series of realizations will be greater than the variance of non-human target objects. 2.

In another example, the feature that can be extracted by the feature extraction circuit 326 is the maximum value (Max _DTW ) of the DTW over a series of realizations (for example, a series of 10 or any suitable number of realizations). In the above equation, Max _DTW corresponds to the maximum value of the calculated DTW over n realizations. Here, the maximum value of DTW can be considered as an extension between different implementations. By using the maximum DTW, even if the human target object is fairly stable for, for example, 5 realizations, but then exhibits movement, the electronic device can determine that the maximum DTW is relatively high. Therefore, even for a temporarily very stationary target object, strong motion can be used to classify the target object as a human being. 3.

In another example, the feature that the feature extraction circuit 326 can extract is the variance of the difference between the peak power in each realization and the average peak power over the sequence of realizations. For example, after the DFT circuit 324 may calculate the DFT for a given implementation. The DFT can provide the received power of the sample at each frequency in a series of frequencies. In the case of using FMCW-based radars, the frequencies correspond to the distance from the electronic device. Therefore, by plotting the power P versus distance, one or more local maxima or peaks (for example, i local maxima) can appear for each corresponding sample. Among the extracted features, the average peak value can be determined in multiple implementations. Here, the change (Δ) of the detected peak power level for each sample n relative to the average peak value can be determined; and, in multiple implementations, the variance of the change can be determined. According to the aspect of the content of this case, for human target objects, the value of this feature may be higher than the value for non-human target objects. 4.

In another example, the feature that the feature extraction circuit 326 can extract is the variance of the difference between the distance where the peak power is located in each realization and the average distance where the peak power is located on the realization sequence. This feature is very similar to that described above for Var_∆P _{(avg_max).} However, here, the feature is not looking at the measured power, but looking at the distance at which the peak power is captured. Similar to the above, for human target objects, the value of this feature can be higher than for non-human target objects. 5.

In another example, the feature that can be extracted by the feature extraction circuit 326 is the extension (Δ) of the peak power of the realized DFT in a series of n (for example, a series of 10) realizations. That is, as shown in Equation 5 above, the spread is defined as the difference between the maximum (max) peak power of the DFT in a series of implementations and the minimum (min) peak power of the DFT. For stationary non-human target objects, the expansion is expected to be relatively small, while for human target objects that at least exhibit micro-movements, the expansion is expected to be large. 6.

In another example, the feature that the feature extraction circuit 326 can extract is a series of n (for example, a series of 10) implementations of the variance of the peak power of the DFT. For stationary non-human target objects, the variance is expected to be relatively small, while for human target objects that at least exhibit micro-movements, the variance is expected to be relatively large. 7.

In another example, the feature that can be extracted by the feature extraction circuit 326 is the variance of the change (Δ) of the power measured in the continuous extraction. In the above equations 1-6, the extracted features depend on the relationship between the entire realization sequence. However, in Equation 7 (and Equation 8 below), the extracted features depend on the relationship between successive or sequential individual captures or realizations. When the target object is a human, the peak power from one acquisition to the next acquisition may have a relatively large change due to the micro-movements that may occur at any given time. By using the variance of this parameter as the extracted feature, the micro-movement characteristics of the human target object can be identified. 8.

In another example, the feature that can be extracted by the feature extraction circuit 326 is the variance of the variation (Δ) of the distance at which the peak power occurs in continuous or sequential extraction or realization. When the target object is a human, there may be a relatively large change in the peak power distance from one capture to the next due to the micro-movements that may occur at any given time. By using the variance of this parameter as the extracted feature, the micro-movement characteristics of the human target object can be identified.

此外，時域取樣的虛部可以根據以下等式決定：

In some other examples, the feature extraction circuit 326 may extract features based on in-phase and quadrature (I/Q) sampling in the time domain after mutual coupling is removed. As an example, the FMCW-based radar can be used as described above to collect a sequence of n consecutive time-domain samples (for example, 10 samples). The real part of the time domain sampling can be determined according to the following equation:

In addition, the imaginary part of the time domain sampling can be determined according to the following equation:

When collecting samples of human target objects, even in the time domain, there may be "noise" or variations in the measured power of continuous samples, for example due to micro-movements. However, when collecting samples of stationary non-human target objects, the measurement power of continuous sampling can generally be relatively stable. Therefore, the electronic device 102 can use calculation parameters such as the variance of time domain sampling to classify the target object. In another example, the average of the I/Q samples can be removed (mean removal). For example, the calculated average or mean value on the sample set can be subtracted from the value of each sample. In this way, any deviation or offset that may affect the final result can be considered. 9.

Therefore, in an example, the feature that can be extracted by the feature extraction circuit 326 is the variance (Var{}) of the real part (Re()) of the _{time-domain sampling (IQ n ), with the aforementioned mean removal.} 10.

In another similar example, the feature that can be extracted by the feature extraction circuit 326 is the variance of the imaginary part (Im()) of the time domain sampling, with the aforementioned mean removal.

Fig. 7 is a flowchart showing an exemplary procedure for constructing a human target object classifier according to some aspects of the content of the present case. In various instances, some or all of the illustrated features may be omitted in specific implementations within the scope of the content of the present case. In addition, some of the illustrated features may not be required to implement a particular example. In some examples, the program in FIG. 7 may be executed by the manufacturer, supplier, or retailer of the electronic device 102. In some examples, the program in FIG. 7 can be executed via any suitable device or component to perform the functions or algorithms described below.

At block 702, a data collection procedure may be executed. For example, the electronic device 102 may collect multiple human radar captured data sets, such as radar captured data sets of different genders, races, ages, sizes, and so on. Such information can be collected from various body parts of the subject. In addition, the electronic device 102 can collect other radar captured data on multiple non-human target objects of various types and characteristics. Here, it may be advantageous to maximize the size of the data set and the types of target objects.

At block 704, the data set may undergo post-processing to extract features that can be used to distinguish human target objects from non-human target objects. For example, one or more of the features described above in relation to Equations 1-10, and any other suitable features that can be used to distinguish the target object can be extracted from the data set.

At block 706, a classifier model may be constructed and trained based on the extracted features and the collected data set. That is, the extracted features can be used to train and verify the performance of the classifier based on the known classification of the samples in the data set. And at block 708, the SVM can be established by mapping the feature space and calculating the distance from the determined boundary (for example, in the multi-dimensional feature space, hyperplane boundary) to the mapped data point. At block 710, the effectiveness of the constructed human classifier can be tested for accuracy in real time based on the detection of unknown (for the classifier) target objects of humans and non-humans. When the reliability of the classifier is appropriately high, the classifier can be deployed to users.

Fig. 8 is a flowchart showing an exemplary procedure for classifying a target object according to some aspects of the content of the present case. Although humans can be used as examples, any living animal can be the basis for adjusting the transmission power. As described below, some or all of the illustrated features may be omitted in a specific implementation within the scope of the content of the present case, and some of the illustrated features may not be necessary to implement all the embodiments. In some instances, the program may be executed by the electronic device 102 shown in FIG. 1 or 2. In some examples, the program can be executed by various elements of the electronic device, including but not limited to the transceiver 120 and the processor or DSP circuit 122 shown in FIG. 3. In some examples, the program of FIG. 8 may be executed via any suitable device or means for performing the functions or algorithms described below.

At block 802, the electronic device 102 may send a detection signal. For example, the transceiver 120 may utilize one or more antennas, such as the transmitting antenna 210-1, to transmit pulses, FMCW signals, multi-tone signals, or any other suitable signals for radar-based proximity detection. At block 804, the electronic device 102 may receive the reflected signal reflected from the target object. For example, the transceiver 120 may utilize one or more antennas, such as the receiving antenna 210-2, to receive the reflected signal.

At block 806, the electronic device 102 may extract one or more characteristics of the reflected signal. For example, the feature extraction circuit 326 may process the information corresponding to the reflected signal, such as the spectrum of one or more radar samples of the reflected signal, to determine one or more features that can be used to characterize the target object. In some instances, a given feature may correspond to a single realization or reflection from a target object. In other examples, a given feature may correspond to a plurality of realizations, such as a sequence of any suitable number of realizations.

At block 808, the electronic device 102 may apply the extracted features to the classification model. For example, the SVM 810 can be configured based on the collection of training data 812 to establish one or more boundaries 814. The one or more boundaries may be configured to separate the classification of the target object in the feature space based on features extracted from the reflected signal received from the target object. The establishment of the boundary 814 based on the training data 812 may be based on the classification model 816 established using the procedure described above and shown in FIG. 7.

At block 818, the electronic device 102 may determine the category of the target object based on one or more characteristics of the reflected signal. For example, the electronic device 102 may determine the position of the target object relative to the boundary 814 in the feature space of the classification model 816. Using this position, the electronic device 102 can recognize the type of the target object based on the position in the feature space (for example, on which side of the boundary 814 the target object is located in the feature space).

If the category of the target object indicates that the target object is a human, then at block 820, the electronic device 102 may adjust at least one transmission parameter of the transmission signal (such as mmW uplink signal) to provide a signal that is no greater than the mmW signal for the human target object. Maximum allowable exposure (MPE) exposure. For example, the electronic device 102 can adjust the power level of the uplink signal, the beam steering angle of the uplink signal, the frequency of the uplink signal, the selected antenna of the uplink signal, and the communication protocol of the uplink signal. At least one of or a combination of the above, so that the power of the uplink signal at the human target object is not greater than the MPE regulatory requirement. On the other hand, if the category of the target object indicates that the target object is non-human, then at block 822, the electronic device 102 may adjust at least one transmission parameter of the transmission signal without considering the MPE regulatory rules. For example, the electronic device can adjust the transmission parameters in such a way that the power of the transmitted signal can exceed the MPE level at the non-human target object. At block 824, the electronic device 102 may use the adjusted transmission parameters to send an adjustment signal, as described above.

The program shown in FIG. 8 may include other aspects, such as any single aspect or any combination of aspects of one or more other programs described below and/or in combination with other parts of this document.

In the first aspect, an electronic device can adjust one or more transmission parameters based on the determined type of the target object. Here, the transmission parameter may be at least one or some combination of power level, beam steering angle, frequency, selected antenna, and communication protocol.

In the second aspect, alone or in combination with the first aspect, the type of the target object can be one of the following: a human target object or a non-human target object; an animal target object or a non-animal target object; or a living target Object or inanimate target object.

In the third aspect, alone or in combination with one or more of the first and second aspects, the electronic device can determine the type of the target object based on one or more characteristics of the reflected signal. Here, the one or more characteristics of the reflected signal may include one or more characteristics indicating the micro-movement characteristics of the human target object.

In the fourth aspect, alone or in combination with one or more of the first to third aspects, determining the type of the target object may include extracting one or more characteristics of the reflected signal, and combining one or more The extracted features are applied to a classification model equipped with a boundary that divides objects into multiple categories in the feature space, determines the position of the target object relative to the boundary in the feature space, and identifies the target object category based on the position in the feature space .

In the fifth aspect, alone or in combination with one or more of the first to fourth aspects, the classification model corresponds to a support vector machine (SVM). Here, the boundary in the feature space is determined based on the training data set.

In the sixth aspect, alone or in combination with one or more of the first to fifth aspects, the adjusted signal includes a millimeter wave (mmW) signal. Here, the category of the target object is a human target object, and adjusting at least one transmission parameter includes configuring the adjusted signal to provide an exposure that is not greater than a maximum allowable exposure (MPE) of mmW signals to the human target object.

In the seventh aspect, alone or in combination with one or more of the first to sixth aspects, one or more characteristics of the reflected signal include at least one of the following: a series of realizations of the sampling target object The variance of the dynamic time warp of the sampling target object; the maximum value of the dynamic time warp of a series of realizations of the sampling target object; the peak power of each realization of the series of sampling target objects and the average peak value of the series of realizations of the sampling target object The variance of the difference between the powers; the variance of the difference between the distance of the peak power in each realization of a series of realizations of the sampled target object and the average distance of the peak power of the series of realizations of the sampled target object ; The expansion of the peak power of a series of discrete Fourier transform (DFT) of the sampling target object; the variance of the peak power of the DFT of a series of sampling target objects; the change of the measured power in the continuous realization of the sampling target object The variance of the distance at which the peak power occurs in the continuous realization of the sampling target object; the variance of the real part of the time domain sampling of the realization of the sampling target object; the variance of the imaginary part of the time domain sampling of the realization of the sampling target object; or Its combination.

In the eighth aspect, alone or in combination with one or more of the first to seventh aspects, the electronic device or wireless communication device includes a housing, the shape and size of which are suitable for carrying one or more components, It includes memory, wireless transceiver, power amplifier and at least one processor. Here, the memory storage is configured to enable at least one processor to process the data of the object sensing information with respect to the stored data, so that one or more objects can be classified into one or more object type classifications.

In the ninth aspect, alone or in combination with one or more of the first to eighth aspects, the wireless transceiver is configured to substantially simultaneously participate in mmW communication and mmW object sensing.

In the tenth aspect, alone or in combination with one or more of the first to ninth aspects, the wireless communication device includes an antenna module having an array of antenna elements, wherein the power amplifier is configured to dynamically The selected antenna element among the antenna elements provides power for adjusting transmission parameters and/or for beamforming.

In the eleventh aspect, alone or in combination with one or more of the first to tenth aspects, the wireless communication device includes an interface circuit for interface connection with auxiliary equipment spaced apart from the housing. The interface circuit is configured to enable communication between the wireless communication device and the auxiliary device, so that the wireless communication device receives object sensing information from the auxiliary device, and in response, adjusts the wireless transceiver to send and/or receive millimeter waves. The transmission parameters of the link.

In the twelfth aspect, alone or in combination with one or more of the first to eleventh aspects, at least one processor is configured to determine whether the sensed object can be combined with one or more objects The categories are related, and one or more object categories include: non-human tissue, human tissue, or a combination thereof.

In the thirteenth aspect, alone or in combination with one or more of the first to twelfth aspects, the wireless transceiver is configured to sense objects in a range from about 1 degree to about 360 degrees , Enabling at least one processor to generate an object sensing map of an object outside the housing.

In the fourteenth aspect, alone or in combination with one or more of the first to thirteenth aspects, at least one processor is configured to generate a map based on at least object sensing information, wherein the map is relative to Wireless communication equipment identification object.

In the fifteenth aspect, alone or in combination with one or more of the first to fourteenth aspects, the wireless transceiver is configured to repeatedly transmit millimeter wave signals to one or more objects and The mmW signal is repeatedly received from one or more objects and transmitted via the mmW signal to sense objects relative to and outside the housing, so that the wireless transceiver is configured to observe the micro-movements caused by the one or more objects.

In the sixteenth aspect, alone or in combination with one or more of the first to fifteenth aspects, the wireless communication device receives signal transmission from one or more other wireless communication devices, indicating that other wireless communication is in progress Any one of the wireless communication devices adjusts one or more wireless transmission parameters based on the object classification information.

In the seventeenth aspect, alone or in combination with one or more of the first to sixteenth aspects, the wireless communication device determines the surrounding information of objects located around one or more other wireless communication devices.

In the eighteenth aspect, alone or in combination with one or more of the first to seventeenth aspects, the wireless transmission parameter is related to the power of the transmitted or received millimeter wave signal.

In the nineteenth aspect, alone or in combination with one or more of the first to eighteenth aspects, the wireless communication device is configured as a vehicle. Here, the at least one processor is configured to control one or more operating parameters of the vehicle body based at least in part on the object sensing information.

In the twentieth aspect, alone or in combination with one or more of the first to nineteenth aspects, the wireless communication device is configured for gaming. Here, one or more user interfaces are located near the housing of the gaming device, and at least one processor is also configured to receive user input via one or more user interfaces located near the housing, and in response to The object sensing information is communicated to the user.

In the twenty-first aspect, alone or in combination with one or more of the first to twentieth aspects, the game device includes a visual interface area, the definition of which is configured to visually convey a sense of the object to the user Visual display for measuring information.

In one configuration, the electronic device 102 includes a means for sending a detection signal, a means for receiving a reflected signal reflected from a target object, and a means for sending an adjusted signal using transmission parameters. In one aspect, the aforementioned component may be the transceiver 120. In one aspect, the aforementioned component may be the processor or DSP circuit 128 shown in FIGS. 1 and 3, configured to perform the functions described by the aforementioned component. In another aspect, the aforementioned component may be a circuit or any device configured to perform the functions described by the aforementioned component. The electronic device 102 may also include a component for determining the type of the target object based on one or more characteristics of the reflected signal, and a component for adjusting at least one transmission parameter based on the type of the target object. In one aspect, the aforementioned component may be the processor or DSP circuit 128 shown in FIGS. 1 and 3, configured to perform the functions described by the aforementioned component. In another aspect, the aforementioned component may be a circuit or any device configured to perform the functions described by the aforementioned component.

Of course, in the above examples, the circuits included in the processor or DSP circuit 128 are provided as examples only, and other components used to perform the described functions may be included in various aspects of the content of this case, including but not limited to storage. Instructions in the computer-readable storage medium 110, or any other suitable device described in any one of FIGS. 1, 2 and/or 3 and using, for example, the procedures and/or algorithms described herein with respect to FIG. 8 Or component.

In the content of this case, the word "exemplary" is used to mean "used as an example, instance, or illustration." Any embodiment or aspect described herein as "exemplary" is not necessarily construed as preferred or other aspects superior to the content of the case. Similarly, the term "aspect" does not require that all aspects of the content of the case include the discussed features, advantages or modes of operation. The term "coupling" is used herein to represent the direct or indirect coupling between two objects. For example, if object A physically touches object B, and object B touches object C, then objects A and C can still be regarded as coupled to each other-even if they are not in direct physical contact with each other. For example, even if the first object has never been in direct physical contact with the second object, the first object can still be coupled to the second object. The terms "circuit" and "circuit system" are widely used, and are intended to include hardware implementations of electrical equipment and conductors that can achieve the functions described in the content of the case when connected and configured, but not about electronics. The limitation of the circuit type and the software implementation of the information and instructions. The software implementation of the information and instructions can realize the functions described in the content of this case when executed by the processor.

One or more of the elements, steps, features and/or functions shown in Figures 1-8 can be rearranged and/or combined into a single element, step, feature or function or embodied in several elements, steps or functions . Without departing from the novel features disclosed herein, additional elements, components, steps and/or functions can also be added. The apparatuses, devices, and/or elements shown in Figures 1-8 can be configured to perform one or more of the methods, features, or steps described herein. The novel algorithms described in this article can also be effectively implemented in software and/or embedded hardware.

It should be understood that the specific order or hierarchy of the steps in the disclosed method is an illustration of an exemplary procedure. Based on design preferences, it can be understood that the specific order or hierarchy of steps in the method can be rearranged. The attached method claims present the elements of each step in an exemplary order, and are not meant to be limited to the specific order or hierarchy presented, unless specifically pointed out herein.

The previous description is provided to enable any person skilled in the art to practice the various aspects described herein. Various modifications to these aspects will be obvious to those skilled in the art, and the general principles defined herein can be applied to other aspects. Therefore, the claim is not intended to be limited to the various aspects shown in this article, but should be given a full range consistent with the language of the claim. The reference to an element in the singular is not intended to mean "one and only one" unless specifically This description is "one or more." Unless specifically stated otherwise, the term "some" refers to one or more. A phrase referring to "at least one" in the list of items refers to any combination of those items, including individual members. For example, "at least one of a, b, or c" is intended to cover: a; b; c; a and b; a and c; b and c; and a, b, and c. All the structural and functional equivalents of the various aspects of the elements described in the full text of the case that are known or later known to those of ordinary skill in the art are expressly incorporated herein by reference and are intended to be covered by the claim. In addition, regardless of whether the disclosed content is clearly stated in the request, any content disclosed in this article is not intended to be contributed to the public. Therefore, no element of the claim shall be interpreted in accordance with the provisions of Article 19, Item 4 of the Enforcement Rules of the Patent Law, unless the element is clearly expressed by the phrase "component used for...", or in the case of a method claim, Use the phrase "steps for..." to describe this element.

100: Example environment 102: electronic equipment 104: base station 106: wireless link 108: application processor 110: Computer readable storage media 112: instruction 114: Information 116: I/O port 118: display 120: wireless transceiver 122: Launcher 124: Receiver 126: Antenna 128: processor/DSP circuit 130: Feature extraction circuit 132: SVM classification circuit 200: Example operating environment 202: Uplink signal 204: DL signal 206: Target Object 208-1: Proximity detection signal 208-2: Proximity detection signal 210: antenna array 210-1: Antenna element 210-2: Antenna element 212-1: Antenna element 212-2: Antenna element 212-N: Antenna element 214: User's Hand 216: Frequency Modulated Continuous Wave (FMCW) signal 218: Multi-tone signal 302: Power Amplifier (PA) 304: Low Noise Amplifier (LNA) 306: LO circuit 308: Mixer 310: Mixer 312: Voltage Controlled Oscillator (VCO) 314: Target Object 316: Mixer 318: Fundamental Frequency Circuit 320: Analog to Digital Converter (ADC) 322: MC elimination circuit 324: Discrete Fourier Transform (DFT) circuit 326: Feature Extraction Circuit 328: classification circuit 402: Picture 404: Figure 406: figure 502: figure 504: figure 602: Line 604: Line 702: Block 704: Block 706: Block 708: Block 710: Block 802: Block 804: Block 806: Block 808: Block 810: SVM 812: Training Materials 814: border 816: classification model 818: Cube 820: Block 822: Cube 824: Block

Fig. 1 is a block diagram of a wireless electronic device according to some aspects of the content of this case.

Fig. 2 is a schematic diagram of an operating environment of an electronic device using a radar-based proximity detector according to some aspects of the content of this case.

Fig. 3 is a block diagram showing additional details of a part of an electronic device according to some aspects of the content of the present case.

Figure 4 is a series of diagrams showing the radar echo signatures of different target objects according to some aspects of the content of the case.

Figure 5 is a series of diagrams illustrating a two-dimensional feature space according to some aspects of the content of this case, which illustrate how the classification of target objects can be achieved based on the use of appropriate features.

FIG. 6 is a diagram showing how a service vector machine (SVM) according to some aspects of the content of this case can establish the best boundary for separating different categories of target objects.

Fig. 7 is a flowchart showing an exemplary procedure for constructing a target object classifier according to some aspects of the content of the present case.

FIG. 8 is a flowchart showing an exemplary procedure for using a target object classifier to control one or more transmission parameters according to some aspects of the content of the present case.

Domestic deposit information (please note in order of deposit institution, date and number) no

Foreign hosting information (please note in the order of hosting country, institution, date, and number) no

120: wireless transceiver

128: processor /DSP circuit

210: antenna array

210-1: Antenna element

210-2: Antenna element

302: Power Amplifier (PA)

304: Low Noise Amplifier (LNA)

306: LO circuit

308: Mixer

310: Mixer

312: Voltage Controlled Oscillator (VCO)

314: Target Object

316: Mixer

318: Fundamental Frequency Circuit

320: Analog-to-digital converter (ADC)

322: MC elimination circuit

324: Discrete Fourier Transform (DFT) circuit

326: Feature Extraction Circuit

328: classification circuit

Claims (27)

A method for classifying a target object, the method includes: Send a detection signal; Receiving a reflected signal reflected from the target object; Determine a category of the target object based on one or more characteristics of the reflected signal; Adjust at least one transmission parameter based on the category of the target object; and Use this transmission parameter to send an adjusted signal. The method according to claim 1, wherein the transmission parameter includes at least one of a power level, a beam steering angle, a frequency, a selected antenna, a communication protocol, or a combination thereof. According to the method of claim 1, wherein the category of the target object is one of the following: A human target object or a non-human target object; An animal target object or a non-animal target object; or A animate target object or an inanimate target object. The method according to claim 1, wherein the one or more characteristics of the reflected signal include one or more characteristics indicative of a micro-movement characteristic of a human target object. The method according to claim 1, wherein the determining the category of the target object includes: Extract the one or more features of the reflected signal; Applying one or more extracted features to a classification model configured with a boundary that classifies objects into categories in a feature space; Determine a position of the target object relative to the boundary in the feature space; and Identify the category of the target object based on the position in the feature space. According to the method of claim 1, Where the adjusted signal includes a millimeter wave signal, Where the category of the target object is a human target object, and The adjusting at least one transmission parameter includes configuring the adjusted signal to provide an exposure that is not greater than a maximum allowable exposure of the millimeter wave signal to the human target object. The method according to claim 1, wherein the one or more characteristics of the reflected signal include at least one of the following: Sampling the deviation of a dynamic time warp of a series of realizations of the target object; Sampling a maximum value of the dynamic time warp achieved by the series of the target object; Sampling the difference between a peak power in each realization of the series of realizations of the target object and an average peak power of an average peak power of the series of realizations of sampling the target object; The difference between a distance between a peak power in each realization of the series of realizations that sampled the target object and an average distance of the peak power in the series of realizations that sampled the target object One side difference Sampling an extension of the peak power of a discrete Fourier transform (DFT) achieved by the series of the target object; Sampling the difference of the peak power of the DFT achieved by the series of the target object; The difference of a change in the measured power in the continuous realization of sampling the target object; In the continuous realization of sampling the target object, there is a difference in a distance at a peak power; Sampling the deviation of a real part of the real part of the realization of the target object; Sampling the side difference of an imaginary part of the realized time domain sampling of the target object; Or a combination. The method according to claim 1, wherein the method can be operated at an electronic device, and the method also includes: Sensing one or more target objects including the target object relative to the electronic device; and Sending information associated with the one or more target objects located relative to the electronic device. An electronic device configured to classify a target object, the electronic device comprising: A processor A transceiver that is communicatively coupled to the processor; and A data storage medium communicatively coupled to the processor, The processor is configured to: Send a detection signal via the transceiver; Receiving a reflection signal via the transceiver, the reflection signal being reflected from the target object; Determine a category of the target object based on one or more characteristics of the reflected signal; Adjust at least one transmission parameter based on the category of the target object; and The transmission parameter is used to send an adjusted signal via the transceiver. The electronic device according to claim 9, wherein the transmission parameter includes at least one of a power level, a beam steering angle, a frequency, a selected antenna, a communication protocol, or a combination thereof, The one or more features of the reflected signal include one or more features indicating the micro-movement characteristics of a human target object, and The category of the target object is one of the following: A human target object or a non-human target object; An animal target object or a non-animal target object; or A animate target object or an inanimate target object. According to the electronic device of claim 9, the processor configured to determine the category of the target object is also configured to: Extract the one or more features of the reflected signal; Applying one or more extracted features to a classification model configured with a boundary that classifies objects into categories in a feature space; Determine a position of the target object relative to the boundary in the feature space; and Identify the category of the target object based on the position in the feature space. According to the electronic equipment of claim 9, Where the adjusted signal includes a millimeter wave signal, Where the category of the target object is a human target object, and The processor configured to adjust at least one transmission parameter is also configured to adjust the signal to provide an exposure that is not greater than a maximum allowable exposure of the millimeter wave signal to the human target object. The electronic device according to claim 9, wherein the one or more characteristics of the reflected signal include at least one of the following: Sampling the deviation of a dynamic time warp of a series of realizations of the target object; Sampling a maximum value of the dynamic time warp achieved by the series of the target object; Sampling the difference between a peak power in each realization of the series of realizations of the target object and an average peak power of an average peak power of the series of realizations of sampling the target object; The difference between a distance between a peak power in each realization of the series of realizations that sampled the target object and an average distance of the peak power in the series of realizations that sampled the target object One side difference Sampling an extension of the peak power of a discrete Fourier transform (DFT) achieved by the series of the target object; Sampling the difference of the peak power of the DFT achieved by the series of the target object; The difference of a change in the measured power in the continuous realization of sampling the target object; In the continuous realization of sampling the target object, there is a difference in a distance at a peak power; Sampling the difference of a real part of a realized time domain sampling of the target object; Sampling the side difference of an imaginary part of the realized time domain sampling of the target object; Or a combination. The electronic device according to claim 9, wherein the processor is also configured to use the transceiver to sense one or more target objects including the target object relative to the electronic device; and for transmitting and relative to the Information associated with the one or more target objects located by the electronic device. A wireless communication device, including: A housing whose shape and size are suitable for carrying one or more components including a memory, a wireless transceiver, a power amplifier and at least one processor; The wireless transceiver is configured to send and/or receive millimeter wave signals via a wireless channel; The wireless transceiver is also configured to sense objects relative to the housing and outside the housing via millimeter wave signal transmission, and is configured to provide object sensing information to the at least one processor; and The at least one processor is configured to control the power amplifier based on the object sensing information to adjust a transmission parameter associated with the wireless transceiver sending and/or receiving millimeter waves, and It is configured to transmit information associated with sensing relative to the housing and objects located outside the housing. The wireless communication device according to claim 15, wherein the wireless transceiver is configured to substantially simultaneously participate in millimeter wave communication and millimeter wave object sensing. The wireless communication device according to claim 15 also includes an antenna module having an antenna element array, wherein the power amplifier is configured to dynamically provide power to selected antenna elements among the antenna elements for adjusting the transmission Parameters and/or used for beamforming. The wireless communication device according to claim 15 also includes an interface circuit for interface connection with an auxiliary device spaced apart from the housing, and the interface circuit is configured to enable communication between the wireless communication device and the auxiliary device Communication enables the wireless communication device to receive object sensing information from the auxiliary device, and in response, adjust a transmission parameter associated with the wireless transceiver to send and/or receive millimeter waves. The wireless communication device according to claim 15, wherein the at least one processor is configured to determine whether the sensed object can be associated with one or more object categories, wherein the one or more object categories include: non-human tissue, Human tissue or a combination thereof. The wireless communication device according to claim 15, wherein the wireless transceiver is also configured to sense an object in a range from about 1 degree to about 360 degrees, so that the at least one processor can generate an object of an object outside the housing Sensing the map. The wireless communication device according to claim 15, wherein the at least one processor is configured to generate a map based on at least the object sensing information, wherein the map identifies the object with respect to the wireless communication device. The wireless communication device according to claim 15, wherein the wireless transceiver is configured to transmit the millimeter wave signal through the millimeter wave signal by repeatedly transmitting the millimeter wave signal to one or more objects and repeatedly receiving the millimeter wave signal from the one or more objects To sense objects with respect to the housing and outside the housing, so that the wireless transceiver is configured to observe the micro-movements caused by the one or more objects. The wireless communication device according to claim 15, wherein the at least one processor is also configured to: Extract one or more characteristics of the object's sensing information; Applying the one or more extracted features to a classification model configured with a boundary for classifying the object into categories in a feature space; Determine the position of the object relative to the boundary in the characteristic space; and The category is identified based on the positions of the objects in the feature space. A wireless communication device, including: A housing whose shape and size are suitable for carrying one or more components including a memory and at least one processor; A wireless communication interface, the size and shape of which are suitable for being placed in a position close to or in the housing; The wireless communication interface is configured to send and/or receive millimeter wave signals via a wireless channel; The wireless communication interface is also configured to sense objects relative to the housing via millimeter wave signal transmission, and is configured to provide object sensing information to at least one processor; and The at least one processor is configured to control a transmission parameter associated with the wireless communication interface sending and/or receiving millimeter waves based on the object sensing information, and is configured to communicate and sense an object relative to the housing Linked information. The wireless communication device according to claim 24, wherein the wireless communication device is configured as a vehicle, Wherein the shell includes a car body configured to carry at least one of a payload or a passenger, and The at least one processor is configured to control one or more operating parameters of the vehicle body based at least in part on the object sensing information. The wireless communication device according to claim 24, wherein the wireless communication device is configured for gaming, Wherein the size and shape of the casing are suitable for games, so as to allow a user to participate in an electronic game environment; The wireless communication device also includes one or more user interfaces located near the housing, wherein the at least one processor is also configured to pass through the one or more user interfaces located near the housing and respond to the object The sensing information is communicated to the user to receive user input. The wireless communication device according to claim 26 also includes a visual interface area defining a visual display configured to visually convey object sensing information to the user.

Images