Patent Application
20210142059
This application claims the priority benefit of the Korean Patent Application No. 10-2019-0143200, filed in the Republic of Korea on Nov. 11, 2019, which is hereby incorporated by reference as if fully set forth herein.
The present disclosure relates to an XR device for providing an augmented reality (AR) mode and a virtual reality (VR) mode and a method for controlling the same, and more particularly, is applicable to a 5G communication technology field, a robot technology field, an autonomous technology field and an artificial intelligence (AI) technology field. The present disclosure relates to an XR device for determining whether to include a virtual object in a screen, through AI, when a user captures an object, and capturing an object using a camera in accordance with the determined result.
Virtual Reality (VR) technology provides an object or background of a real world through a computer graphic (CG) image, Augmented Reality (AR) technology provides a CG image virtually made on a real object image, and Mixed Reality (MR) technology is a computer graphic technology that provides virtual objects to a real world by mixing and combining the virtual objects. The aforementioned VR, AR, MR, etc. will simply be referred to as extended reality (XR) technology.
The AR technology is to allow a virtual digital image to be worn in the real world. The AR technology is different from the VR technology for displaying a graphic image in a state that eyes of a user are covered with something in that the user may view the real world through his/her eyes. Unlike a VR device that may only be used indoors, AR glasses are used in various ways (e.g., the user may use the AR glasses while walking).
In the related art, when a user takes a photo using AR classes, the user should manipulate whether to include a virtual object. Therefore, if a mode that does not include a virtual object is set, the virtual object is not included in an image or video even in the situation that the user desires to include the virtual object, whereby a problem occurs in that the user feels inconvenience.
Accordingly, the present disclosure is directed to an XR device for providing an augmented reality (AR) mode and a virtual reality (VR) mode and a method for controlling the same, which substantially obviates one or more problems due to limitations and disadvantages of the related art.
An object of the present disclosure is to provide an XR device and a method for controlling the same, which identifies a state near a user by using a reinforcement learning model and determines whether to include a virtual object in a screen by predicting a behavior which will be performed by the user.
Another object of the present disclosure is to provide an XR device and a method for controlling the same, which independently determines whether to include a plurality of virtual objects having different attributes in a screen when the plurality of virtual objects are simultaneously displayed on a screen.
In addition to the objects of the present disclosure as mentioned above, additional objects and features of the present disclosure will be clearly understood by those skilled in the art from the following description of the present disclosure.
To achieve these objects and other advantages and in accordance with the purpose of the invention, as embodied and broadly described herein, an XR device according to one aspect of the present disclosure comprises a wireless communication module transmitting or receiving data to or from an external device; a camera capturing an image of an object in front of the XR device; a display including a transparent portion and displaying the captured image; a sensor module sensing peripheral environment information surrounding the XR device; and a controller entering a camera mode if a selection input for executing the camera mode is received from a user, controlling the display to display a menu that includes a first button for displaying at least one virtual object, a second button for executing capturing and a third button for storing at least one virtual object, determining whether to include some of at least one virtual object in a screen when capturing an image based on the peripheral environment information and the object, and controlling the camera to capture the image in accordance with the determined result.
In another aspect of the present disclosure, a method for controlling an XR device comprises sensing peripheral environment information surrounding the XR device; entering a camera mode if a selection input for executing the camera mode is received from a user; controlling the display to display a menu that includes a first button for displaying at least one virtual object, a second button for executing capturing and a third button for storing at least one virtual object; determining whether to include some of at least one virtual object in a screen when capturing an image based on the peripheral environment information and the object; and controlling the camera to capture the image in accordance with the determined result.
According to one embodiment of the present disclosure, when a user views an object in front of him/her in a state that the user wears AR glasses, whether to include a virtual object in a screen may be determined based on peripheral environment information and objects, whereby the user does not manipulate a function for including a virtual object manually and therefore user convenience may be improved.
According to another embodiment of the present disclosure, a state near a user may be identified using a reinforcement learning model and whether to include a virtual object in a screen may be determined by predicting a behavior which will be performed by the user, whereby a feedback of the user may be reflected more exactly and therefore user convenience may be improved.
According to still another embodiment of the present disclosure, when a plurality of virtual objects having different attributes are simultaneously displayed on a screen, whether to include the plurality of virtual objects in the screen may independently be determined, whereby intention of the user may be reflected more exactly and therefore user convenience may be improved.
In addition to the effects of the present disclosure as mentioned above, additional effects and features of the present disclosure will be clearly understood by those skilled in the art from the following description of the present disclosure.
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the principle of the invention. In the drawings:
Reference will now be made in detail to embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts, and a redundant description will be avoided. The terms “module” and “unit” are interchangeably used only for easiness of description and thus they should not be considered as having distinctive meanings or roles. Further, a detailed description of well-known technology will not be given in describing embodiments of the present disclosure lest it should obscure the subject matter of the embodiments. The attached drawings are provided to help the understanding of the embodiments of the present disclosure, not limiting the scope of the present disclosure. It is to be understood that the present disclosure covers various modifications, equivalents, and/or alternatives falling within the scope and spirit of the present disclosure.
The following embodiments of the present disclosure are intended to embody the present disclosure, not limiting the scope of the present disclosure. What could easily be derived from the detailed description of the present disclosure and the embodiments by a person skilled in the art is interpreted as falling within the scope of the present disclosure.
The above embodiments are therefore to be construed in all aspects as illustrative and not restrictive. The scope of the disclosure should be determined by the appended claims and their legal equivalents, not by the above description, and all changes coming within the meaning and equivalency range of the appended claims are intended to be embraced therein.
In the disclosure, downlink (DL) refers to communication from a base station (BS) to a user equipment (UE), and uplink (UL) refers to communication from the UE to the BS. On DL, a transmitter may be a part of the BS and a receiver may be a part of the UE, whereas on UL, a transmitter may be a part of the UE and a receiver may be a part of the BS. A UE may be referred to as a first communication device, and a BS may be referred to as a second communication device in the present disclosure. The term BS may be replaced with fixed station, Node B, evolved Node B (eNB), next generation Node B (gNB), base transceiver system (BTS), access point (AP), network or 5th generation (5G) network node, artificial intelligence (AI) system, road side unit (RSU), robot, augmented reality/virtual reality (AR/VR) system, and so on. The term UE may be replaced with terminal, mobile station (MS), user terminal (UT), mobile subscriber station (MSS), subscriber station (SS), advanced mobile station (AMS), wireless terminal (WT), device-to-device (D2D) device, vehicle, robot, AI device (or module), AR/VR device (or module), and so on.
The following technology may be used in various wireless access systems including code division multiple access (CDMA), frequency division multiple access (FDMA), time division multiple access (TDMA), orthogonal frequency division multiple access (OFDMA), and single carrier FDMA (SC-FDMA).
For the convenience of description, the present disclosure is described in the context of a 3rd generation partnership project (3GPP) communication system (e.g., long term evolution-advanced (LTE-A) and new radio or new radio access technology (NR)), which should not be construed as limiting the present disclosure. For reference, 3GPP LTE is part of evolved universal mobile telecommunications system (E-UMTS) using evolved UMTS terrestrial radio access (E-UTRA), and LTE-A/LTE-A pro is an evolution of 3GPP LTE. 3GPP NR is an evolution of 3GPP/LTE-A/LTE-A pro.
In the present disclosure, a node refers to a fixed point capable of transmitting/receiving wireless signals by communicating with a UE. Various types of BSs may be used as nodes irrespective of their names. For example, any of a BS, an NB, an eNB, a pico-cell eNB (PeNB), a home eNB (HeNB), a relay, and a repeater may be a node. At least one antenna is installed in one node. The antenna may refer to a physical antenna, an antenna port, a virtual antenna, or an antenna group. A node is also referred to as a point.
In the present disclosure, a cell may refer to a certain geographical area or radio resources, in which one or more nodes provide a communication service. A “cell” as a geographical area may be understood as coverage in which a service may be provided in a carrier, while a “cell” as radio resources is associated with the size of a frequency configured in the carrier, that is, a bandwidth (BW). Because a range in which a node may transmit a valid signal, that is, DL coverage and a range in which the node may receive a valid signal from a UE, that is, UL coverage depend on a carrier carrying the signals, and thus the coverage of the node is associated with the “cell” coverage of radio resources used by the node. Accordingly, the term “cell” may mean the service overage of a node, radio resources, or a range in which a signal reaches with a valid strength in the radio resources, under circumstances.
In the present disclosure, communication with a specific cell may amount to communication with a BS or node that provides a communication service to the specific cell. Further, a DL/UL signal of a specific cell means a DL/UL signal from/to a BS or node that provides a communication service to the specific cell. Particularly, a cell that provides a UL/DL communication service to a UE is called a serving cell for the UE. Further, the channel state/quality of a specific cell refers to the channel state/quality of a channel or a communication link established between a UE and a BS or node that provides a communication service to the specific cell.
A “cell” associated with radio resources may be defined as a combination of DL resources and UL resources, that is, a combination of a DL component carrier (CC) and a UL CC. A cell may be configured with DL resources alone or both DL resources and UL resources in combination. When carrier aggregation (CA) is supported, linkage between the carrier frequency of DL resources (or a DL CC) and the carrier frequency of UL resources (or a UL CC) may be indicated by system information transmitted in a corresponding cell. A carrier frequency may be identical to or different from the center frequency of each cell or CC. Hereinbelow, a cell operating in a primary frequency is referred to as a primary cell (Pcell) or PCC, and a cell operating in a secondary frequency is referred to as a secondary cell (Scell) or SCC. The Scell may be configured after a UE and a BS perform a radio resource control (RRC) connection establishment procedure and thus an RRC connection is established between the UE and the BS, that is, the UE is RRC_SONNECTED. The RRC connection may mean a path in which the RRC of the UE may exchange RRC messages with the RRC of the BS. The Scell may be configured to provide additional radio resources to the UE. The Scell and the Pcell may form a set of serving cells for the UE according to the capabilities of the UE. Only one serving cell configured with a Pcell exists for an RRC_CONNECTED UE which is not configured with CA or does not support CA.
A cell supports a unique radio access technology (RAT). For example, LTE RAT-based transmission/reception is performed in an LTE cell, and 5G RAT-based transmission/reception is performed in a 5G cell.
CA aggregates a plurality of carriers each having a smaller system BW than a target BW to support broadband. CA differs from OFDMA in that DL or UL communication is conducted in a plurality of carrier frequencies each forming a system BW (or channel BW) in the former, and DL or UL communication is conducted by loading a basic frequency band divided into a plurality of orthogonal subcarriers in one carrier frequency in the latter. In OFDMA or orthogonal frequency division multiplexing (OFDM), for example, one frequency band having a certain system BW is divided into a plurality of subcarriers with a predetermined subcarrier spacing, information/data is mapped to the plurality of subcarriers, and the frequency band in which the information/data has been mapped is transmitted in a carrier frequency of the frequency band through frequency upconversion. In wireless CA, frequency bands each having a system BW and a carrier frequency may be used simultaneously for communication, and each frequency band used in CA may be divided into a plurality of subcarriers with a predetermined subcarrier spacing.
The 3GPP communication standards define DL physical channels corresponding to resource elements (REs) conveying information originated from upper layers of the physical layer (e.g., the medium access control (MAC) layer, the radio link control (RLC) layer, the packet data convergence protocol (PDCP) layer, the radio resource control (RRC) layer, the service data adaptation protocol (SDAP) layer, and the non-access stratum (NAS) layer), and DL physical signals corresponding to REs which are used in the physical layer but do not deliver information originated from the upper layers. For example, physical downlink shared channel (PDSCH), physical broadcast channel (PBCH), physical multicast channel (PMCH), physical control format indicator channel (PCFICH), and physical downlink control channel (PDCCH) are defined as DL physical channels, and a reference signal (RS) and a synchronization signal are defined as DL physical signals. An RS, also called a pilot is a signal in a predefined special waveform known to both a BS and a UE. For example, cell specific RS (CRS), UE-specific RS (UE-RS), positioning RS (PRS), channel state information RS (CSI-RS), and demodulation RS (DMRS) are defined as DL RSs. The 3GPP communication standards also define UL physical channels corresponding to REs conveying information originated from upper layers, and UL physical signals corresponding to REs which are used in the physical layer but do not carry information originated from the upper layers. For example, physical uplink shared channel (PUSCH), physical uplink control channel (PUCCH), and physical random access channel (PRACH) are defined as UL physical channels, and DMRS for a UL control/data signal and sounding reference signal (SRS) used for UL channel measurement are defined.
In the present disclosure, physical shared channels (e.g., PUSCH and PDSCH) are used to deliver information originated from the upper layers of the physical layer (e.g., the MAC layer, the RLC layer, the PDCP layer, the RRC layer, the SDAP layer, and the NAS layer).
In the present disclosure, an RS is a signal in a predefined special waveform known to both a BS and a UE. In a 3GPP communication system, for example, the CRS being a cell common RS, the UE-RS for demodulation of a physical channel of a specific UE, the CSI-RS used to measure/estimate a DL channel state, and the DMRS used to demodulate a physical channel are defined as DL RSs, and the DMRS used for demodulation of a UL control/data signal and the SRS used for UL channel state measurement/estimation are defined as UL RSs.
In the present disclosure, a transport block (TB) is payload for the physical layer. For example, data provided to the physical layer by an upper layer or the MAC layer is basically referred to as a TB. A UE which is a device including an AR/VR module (i.e., an AR/VR device) may transmit a TB including AR/VR data to a wireless communication network (e.g., a 5G network) on a PUSCH. Further, the UE may receive a TB including AR/VR data of the 5G network or a TB including a response to AR/VR data transmitted by the UE from the wireless communication network.
In the present disclosure, hybrid automatic repeat and request (HARQ) is a kind of error control technique. An HARQ acknowledgement (HARQ-ACK) transmitted on DL is used for error control of UL data, and a HARQ-ACK transmitted on UL is used for error control of DL data. A transmitter performing an HARQ operation awaits reception of an ACK after transmitting data (e.g., a TB or a codeword). A receiver performing an HARQ operation transmits an ACK only when data has been successfully received, and a negative ACK (NACK) when the received data has an error. Upon receipt of the ACK, the transmitter may transmit (new) data, and upon receipt of the NACK, the transmitter may retransmit the data.
In the present disclosure, CSI refers to information representing the quality of a radio channel (or link) established between a UE and an antenna port. The CSI may include at least one of a channel quality indicator (CQI), a precoding matrix indicator (PMI), a CSI-RS resource indicator (CRI), a synchronization signal block resource indicator (SSBRI), a layer indicator (L1), a rank indicator (RI), or a reference signal received power (RSRP).
In the present disclosure, frequency division multiplexing (FDM) is transmission/reception of signals/channels/users in different frequency resources, and time division multiplexing (TDM) is transmission/reception of signals/channels/users in different time resources.
In the present disclosure, frequency division duplex (FDD) is a communication scheme in which UL communication is performed in a UL carrier, and DL communication is performed in a DL carrier linked to the UL carrier, whereas time division duplex (TDD) is a communication scheme in which UL communication and DL communication are performed in time division in the same carrier. In the present disclosure, half-duplex is a scheme in which a communication device operates on UL or UL only in one frequency at one time point, and on DL or UL in another frequency at another time point. For example, when the communication device operates in half-duplex, the communication device communicates in UL and DL frequencies, wherein the communication device performs a UL transmission in the UL frequency for a predetermined time, and retunes to the DL frequency and performs a DL reception in the DL frequency for another predetermined time, in time division, without simultaneously using the UL and DL frequencies.
Referring to
Nsize,μgrid may be different between UL and DL as well as a subcarrier spacing configuration μ. For the subcarrier spacing configuration μ, an antenna port p, and a transmission direction (UL or DL), there is one resource grid. Each element of a resource grid for the subcarrier spacing configuration μ and the antenna port p is referred to as an RE, uniquely identified by an index pair (k,l) where k is a frequency-domain index and l is the position of a symbol in a relative time domain with respect to a reference point. A frequency unit used for mapping physical channels to REs, resource block (RB) is defined by 12 consecutive subcarriers (NRBsc=12) in the frequency domain. Considering that a UE may not support a wide BW supported by the 5G system at one time, the UE may be configured to operate in a part (referred to as a bandwidth part (BWP)) of the frequency BW of a cell.
For the background technology, terminology, and abbreviations used in the present disclosure, standard specifications published before the present disclosure may be referred to. For example, the following documents may be referred to.
3GPP LTE
3GPP NR (e.g. 5G)
Referring to
After the initial cell search, the UE may receive a PBCH from the BS and acquire broadcast information within a cell from the PBCH. During the initial cell search, the UE may check a DL channel state by receiving a DL RS.
Upon completion of the initial cell search, the UE may acquire more specific system information by receiving a PDCCH and receiving a PDSCH according to information carried on the PDCCH (S202).
When the UE initially accesses the BS or has no radio resources for signal transmission, the UE may perform a random access procedure with the BS (S203 to S206). For this purpose, the UE may transmit a predetermined sequence as a preamble on a PRACH (S203 and S205) and receive a PDCCH, and a random access response (RAR) message in response to the preamble on a PDSCH corresponding to the PDCCH (S204 and S206). If the random access procedure is contention-based, the UE may additionally perform a contention resolution procedure. The random access procedure will be described below in greater detail.
After the above procedure, the UE may then perform PDCCH/PDSCH reception (S207) and PUSCH/PUCCH transmission (S208) in a general UL/DL signal transmission procedure. Particularly, the UE receives DCI on a PDCCH.
The UE monitors a set of PDCCH candidates in monitoring occasions configured for one or more control element sets (CORESETs) in a serving cell according to a corresponding search space configuration. The set of PDCCH candidates to be monitored by the UE is defined from the perspective of search space sets. A search space set may be a common search space set or a UE-specific search space set. A CORESET includes a set of (physical) RBs that last for a time duration of one to three OFDM symbols. The network may configure a plurality of CORESETs for the UE. The UE monitors PDCCH candidates in one or more search space sets. Herein, monitoring is attempting to decode PDCCH candidate(s) in a search space. When the UE succeeds in decoding one of the PDCCH candidates in the search space, the UE determines that a PDCCH has been detected from among the PDCCH candidates and performs PDSCH reception or PUSCH transmission based on DCI included in the detected PDCCH.
The PDCCH may be used to schedule DL transmissions on a PDSCH and UL transmissions on a PUSCH. DCI in the PDCCH includes a DL assignment (i.e., a DL grant) including at least a modulation and coding format and resource allocation information for a DL shared channel, and a UL grant including a modulation and coding format and resource allocation information for a UL shared channel.
Initial Access (IA) Procedure
Synchronization Signal Block (SSB) Transmission and Related Operation
Referring to
Cell Search
Cell search is a process of acquiring the time/frequency synchronization of a cell and detecting the cell ID (e.g., physical cell ID (PCI)) of the cell by a UE. The PSS is used to detect a cell ID in a cell ID group, and the SSS is used to detect the cell ID group. The PBCH is used for SSB (time) index detection and half-frame detection.
In the 5G system, there are 336 cell ID groups each including 3 cell IDs. Therefore, a total of 1008 cell IDs are available. Information about a cell ID group to which the cell ID of a cell belongs is provided/acquired by/from the SSS of the cell, and information about the cell ID among 336 cells within the cell ID is provided/acquired by/from the PSS.
The SSB is periodically transmitted with an SSB periodicity. The UE assumes a default SSB periodicity of 20 ms during initial cell search. After cell access, the SSB periodicity may be set to one of {5 ms, 10 ms, 20 ms, 40 ms, 80 ms, 160 ms} by the network (e.g., a BS). An SSB burst set is configured at the start of an SSB period. The SSB burst set is composed of a 5-ms time window (i.e., half-frame), and the SSB may be transmitted up to L times within the SSB burst set. The maximum number L of SSB transmissions may be given as follows according to the frequency band of a carrier.
The possible time positions of SSBs in a half-frame are determined by a subcarrier spacing, and the periodicity of half-frames carrying SSBs is configured by the network. The time positions of SSB candidates are indexed as 0 to L−1 (SSB indexes) in a time order in an SSB burst set (i.e., half-frame). Other SSBs may be transmitted in different spatial directions (by different beams spanning the coverage area of the cell) during the duration of a half-frame. Accordingly, an SSB index (SSBI) may be associated with a BS transmission (Tx) beam in the 5G system.
The UE may acquire DL synchronization by detecting an SSB. The UE may identify the structure of an SSB burst set based on a detected (time) SSBI and hence a symbol/slot/half-frame boundary. The number of a frame/half-frame to which the detected SSB belongs may be identified by using system frame number (SFN) information and half-frame indication information.
Specifically, the UE may acquire the 10-bit SFN of a frame carrying the PBCH from the PBCH. Subsequently, the UE may acquire 1-bit half-frame indication information. For example, when the UE detects a PBCH with a half-frame indication bit set to 0, the UE may determine that an SSB to which the PBCH belongs is in the first half-frame of the frame. When the UE detects a PBCH with a half-frame indication bit set to 1, the UE may determine that an SSB to which the PBCH belongs is in the second half-frame of the frame. Finally, the UE may acquire the SSBI of the SSB to which the PBCH belongs based on a DMRS sequence and PBCH payload delivered on the PBCH.
System Information (SI) Acquisition
SI is divided into a master information block (MIB) and a plurality of system information blocks (SIBs). The SI except for the MIB may be referred to as remaining minimum system information (RMSI). For details, the following may be referred to.
Random Access Procedure
The random access procedure serves various purposes. For example, the random access procedure may be used for network initial access, handover, and UE-triggered UL data transmission. The UE may acquire UL synchronization and UL transmission resources in the random access procedure. The random access procedure may be contention-based or contention-free.
First, a UE may transmit a random access preamble as a first message (Msg1) of the random access procedure on a PRACH. In the present disclosure, a random access procedure and a random access preamble are also referred to as a RACH procedure and a RACH preamble, respectively.
A plurality of preamble formats are defined by one or more RACH OFDM symbols and different cyclic prefixes (CPs) (and/or guard times). A RACH configuration for a cell is included in system information of the cell and provided to the UE. The RACH configuration includes information about a subcarrier spacing, available preambles, a preamble format, and so on for a PRACH. The RACH configuration includes association information between SSBs and RACH (time-frequency) resources, that is, association information between SSBIs and RACH (time-frequency) resources. The SSBIs are associated with Tx beams of a BS, respectively. The UE transmits a RACH preamble in RACH time-frequency resources associated with a detected or selected SSB. The BS may identify a preferred BS Tx beam of the UE based on time-frequency resources in which the RACH preamble has been detected.
An SSB threshold for RACH resource association may be configured by the network, and a RACH preamble transmission (i.e., PRACH transmission) or retransmission is performed based on an SSB in which an RSRP satisfying the threshold has been measured. For example, the UE may select one of SSB(s) satisfying the threshold and transmit or retransmit the RACH preamble in RACH resources associated with the selected SSB.
Upon receipt of the RACH preamble from the UE, the BS transmits an RAR message (a second message (Msg2)) to the UE. A PDCCH that schedules a PDSCH carrying the RAR message is cyclic redundancy check (CRC)-masked by an RA radio network temporary identifier (RNTI) (RA-RNTI) and transmitted. When the UE detects the PDCCH masked by the RA-RNTI, the UE may receive the RAR message on the PDSCH scheduled by DCI delivered on the PDCCH. The UE determines whether RAR information for the transmitted preamble, that is, Msg1 is included in the RAR message. The UE may determine whether random access information for the transmitted Msg1 is included by checking the presence or absence of the RACH preamble ID of the transmitted preamble. If the UE fails to receive a response to Msg1, the UE may transmit the RACH preamble a predetermined number of or fewer times, while performing power ramping. The UE calculates the PRACH transmission power of a preamble retransmission based on the latest pathloss and a power ramping counter.
Upon receipt of the RAR information for the UE on the PDSCH, the UE may acquire timing advance information for UL synchronization, an initial UL grant, and a UE temporary cell RNTI (C-RNTI). The timing advance information is used to control a UL signal transmission timing. To enable better alignment between PUSCH/PUCCH transmission of the UE and a subframe timing at a network end, the network (e.g., BS) may measure the time difference between PUSCH/PUCCH/SRS reception and a subframe and transmit the timing advance information based on the measured time difference. The UE may perform a UL transmission as a third message (Msg3) of the RACH procedure on a PUSCH. Msg3 may include an RRC connection request and a UE ID. The network may transmit a fourth message (Msg4) in response to Msg3, and Msg4 may be treated as a contention solution message on DL. As the UE receives Msg4, the UE may enter an RRC_CONNECTED state.
The contention-free RACH procedure may be used for handover of the UE to another cell or BS or performed when requested by a BS command. The contention-free RACH procedure is basically similar to the contention-based RACH procedure. However, compared to the contention-based RACH procedure in which a preamble to be used is randomly selected among a plurality of RACH preambles, a preamble to be used by the UE (referred to as a dedicated RACH preamble) is allocated to the UE by the BS in the contention-free RACH procedure. Information about the dedicated RACH preamble may be included in an RRC message (e.g., a handover command) or provided to the UE by a PDCCH order. When the RACH procedure starts, the UE transmits the dedicated RACH preamble to the BS. When the UE receives the RACH procedure from the BS, the RACH procedure is completed.
DL and UL Transmission/Reception Operations
DL Transmission/Reception Operation
DL grants (also called DL assignments) may be classified into (1) dynamic grant and (2) configured grant. A dynamic grant is a data transmission/reception method based on dynamic scheduling of a BS, aiming to maximize resource utilization.
The BS schedules a DL transmission by DCI. The UE receives the DCI for DL scheduling (i.e., including scheduling information for a PDSCH) (referred to as DL grant DCI) from the BS. The DCI for DL scheduling may include, for example, the following information: a BWP indicator, a frequency-domain resource assignment, a time-domain resource assignment, and a modulation and coding scheme (MCS).
The UE may determine a modulation order, a target code rate, and a TB size (TBS) for the PDSCH based on an MCS field in the DCI. The UE may receive the PDSCH in time-frequency resources according to the frequency-domain resource assignment and the time-domain resource assignment.
The DL configured grant is also called semi-persistent scheduling (SPS). The UE may receive an RRC message including a resource configuration for DL data transmission from the BS. In the situation of DL SPS, an actual DL configured grant is provided by a PDCCH, and the DL SPS is activated or deactivated by the PDCCH. When DL SPS is configured, the BS provides the UE with at least the following parameters by RRC signaling: a configured scheduling RNTI (CS-RNTI) for activation, deactivation, and retransmission; and a periodicity. An actual DL grant (e.g., a frequency resource assignment) for DL SPS is provided to the UE by DCI in a PDCCH addressed to the CS-RNTI. If a specific field in the DCI of the PDCCH addressed to the CS-RNTI is set to a specific value for scheduling activation, SPS associated with the CS-RNTI is activated. The DCI of the PDCCH addressed to the CS-RNTI includes actual frequency resource allocation information, an MCS index, and so on. The UE may receive DL data on a PDSCH based on the SPS.
UL Transmission/Reception Operation
UL grants may be classified into (1) dynamic grant that schedules a PUSCH dynamically by UL grant DCI and (2) configured grant that schedules a PUSCH semi-statically by RRC signaling.
In the situation of a UL dynamic grant, the BS transmits DCI including UL scheduling information to the UE. The UE receives DCI for UL scheduling (i.e., including scheduling information for a PUSCH) (referred to as UL grant DCI) on a PDCCH. The DCI for UL scheduling may include, for example, the following information: a BWP indicator, a frequency-domain resource assignment, a time-domain resource assignment, and an MCS. For efficient allocation of UL radio resources by the BS, the UE may transmit information about UL data to be transmitted to the BS, and the BS may allocate UL resources to the UE based on the information. The information about the UL data to be transmitted is referred to as a buffer status report (BSR), and the BSR is related to the amount of UL data stored in a buffer of the UE.
Referring to
Referring to (a) of
Referring to (b) of
Each of the blocks illustrated in
Each of the above processes will be described in greater detail. For one codeword, the transmission device may scramble coded bits of the codeword by a scrambler and then transmit the scrambled bits on a physical channel. The codeword is obtained by encoding a TB. The scrambled bits are modulated to complex-valued modulation symbols by a modulation mapper. The modulation mapper may modulate the scrambled bits in a predetermined modulation scheme and arrange the modulated bits as complex-valued modulation symbols representing positions on a signal constellation. Pi/2-binay phase shift keying (pi/2-BPSK), m-phase shift keying (m-PSK), m-quadrature amplitude modulation (m-QAM), or the like is available for modulation of the coded data. The complex-valued modulation symbols may be mapped to one or more transmission layers by a layer mapper. A complexed-value modulation symbol on each layer may be precoded by a precoder, for transmission through an antenna port. If transform precoding is possible for UL transmission, the precoder may perform precoding after the complex-valued modulation symbols are subjected to transform precoding, as illustrated in
A signal processing procedure of the reception device is performed in a reverse order of the signal processing procedure of the transmission device. For details, refer to the above description and
Now, a description will be given of the PUCCH.
The PUCCH is used for UCI transmission. UCI includes an SR requesting UL transmission resources, CSI representing a UE-measured DL channel state based on a DL RS, and/or an HARQ-ACK indicating whether a UE has successfully received DL data.
The PUCCH supports multiple formats, and the PUCCH formats are classified according to symbol durations, payload sizes, and multiplexing or non-multiplexing. [Table 1] below lists exemplary PUCCH formats.
The BS configures PUCCH resources for the UE by RRC signaling. For example, to allocate PUCCH resources, the BS may configure a plurality of PUCCH resource sets for the UE, and the UE may select a specific PUCCH resource set corresponding to a UCI (payload) size (e.g., the number of UCI bits). For example, the UE may select one of the following PUCCH resource sets according to the number of UCI bits, NUCI.
. . .
Herein, K represents the number of PUCCH resource sets (K>1), and Ni represents the maximum number of UCI bits supported by PUCCH resource set #i. For example, PUCCH resource set #1 may include resources of PUCCH format 0 to PUCCH format 1, and the other PUCCH resource sets may include resources of PUCCH format 2 to PUCCH format 4.
Subsequently, the BS may transmit DCI to the UE on a PDCCH, indicating a PUCCH resource to be used for UCI transmission among the PUCCH resources of a specific PUCCH resource set by an ACK/NACK resource indicator (ARI) in the DCI. The ARI may be used to indicate a PUCCH resource for HARQ-ACK transmission, also called a PUCCH resource indicator (PRI).
enhanced Mobile Broadband communication (eMBB)
In the NR system, a massive MIMO environment in which the number of Tx/Rx antennas is significantly increased is under consideration. On the other hand, in an NR system operating at or above 6 GHz, beamforming is considered, in which a signal is transmitted with concentrated energy in a specific direction, not omni-directionally, to compensate for rapid propagation attenuation. Accordingly, there is a need for hybrid beamforming with analog beamforming and digital beamforming in combination according to a position to which a beamforming weight vector/precoding vector is applied, for the purpose of increased performance, flexible resource allocation, and easiness of frequency-wise beam control.
Hybrid Beamforming
In hybrid beamforming, a BS or a UE may form a narrow beam by transmitting the same signal through multiple antennas, using an appropriate phase difference and thus increasing energy only in a specific direction.
Beam Management (BM)
BM is a series of processes for acquiring and maintaining a set of BS (or transmission and reception point (TRP)) beams and/or UE beams available for DL and UL transmissions/receptions. BM may include the following processes and terminology.
The BM procedure may be divided into (1) a DL BM procedure using an SSB or CSI-RS and (2) a UL BM procedure using an SRS. Further, each BM procedure may include Tx beam sweeping for determining a Tx beam and Rx beam sweeping for determining an Rx beam. The following description will focus on the DL BM procedure using an SSB.
The DL BM procedure using an SSB may include (1) transmission of a beamformed SSB from the BS and (2) beam reporting of the UE. An SSB may be used for both of Tx beam sweeping and Rx beam sweeping. SSB-based Rx beam sweeping may be performed by attempting SSB reception while changing Rx beams at the UE.
SSB-based beam reporting may be configured, when CSI/beam is configured in the RRC_CONNECTED state.
The BS may determine a BS Tx beam for use in DL transmission to the UE based on a beam report received from the UE.
Beam Failure Recovery (BFR) Procedure
In a beamforming system, radio link failure (RLF) may often occur due to rotation or movement of a UE or beamforming blockage. Therefore, BFR is supported to prevent frequent occurrence of RLF in NR.
For beam failure detection, the BS configures beam failure detection RSs for the UE. If the number of beam failure indications from the physical layer of the UE reaches a threshold configured by RRC signaling within a period configured by RRC signaling of the BS, the UE declares beam failure.
After the beam failure is detected, the UE triggers BFR by initiating a RACH procedure on a Pcell, and performs BFR by selecting a suitable beam (if the BS provides dedicated RACH resources for certain beams, the UE performs the RACH procedure for BFR by using the dedicated RACH resources first of all). Upon completion of the RACH procedure, the UE considers that the BFR has been completed.
Ultra-Reliable and Low Latency Communication (URLLC)
A URLLC transmission defined in NR may mean a transmission with (1) a relatively small traffic size, (2) a relatively low arrival rate, (3) an extremely low latency requirement (e.g., 0.5 ms or 1 ms), (4) a relatively short transmission duration (e.g., 2 OFDM symbols), and (5) an emergency service/message.
Pre-Emption Indication
Although eMBB and URLLC services may be scheduled in non-overlapped time/frequency resources, a URLLC transmission may take place in resources scheduled for on-going eMBB traffic. To enable a UE receiving a PDSCH to determine that the PDSCH has been partially punctured due to URLLC transmission of another UE, a preemption indication may be used. The preemption indication may also be referred to as an interrupted transmission indication.
In relation to a preemption indication, the UE receives DL preemption RRC information (e.g., a DownlinkPreemption IE) from the BS by RRC signaling.
The UE receives DCI format 2_1 based on the DL preemption RRC information from the BS. For example, the UE attempts to detect a PDCCH conveying preemption indication-related DCI, DCI format 2_1 by using an int-RNTI configured by the DL preemption RRC information.
Upon detection of DCI format 2_1 for serving cell(s) configured by the DL preemption RRC information, the UE may assume that there is no transmission directed to the UE in RBs and symbols indicated by DCI format 2_1 in a set of RBs and a set of symbols during a monitoring interval shortly previous to a monitoring interval to which DCI format 2_1 belongs. For example, the UE decodes data based on signals received in the remaining resource areas, considering that a signal in a time-frequency resource indicated by a preemption indication is not a DL transmission scheduled for the UE.
Massive MTC (mMTC)
mMTC is one of 5G scenarios for supporting a hyper-connectivity service in which communication is conducted with multiple UEs at the same time. In this environment, a UE intermittently communicates at a very low transmission rate with low mobility. Accordingly, mMTC mainly seeks long operation of a UE with low cost. In this regard, MTC and narrow band-Internet of things (NB-IoT) handled in the 3GPP will be described below.
The following description is given with the appreciation that a transmission time interval (TTI) of a physical channel is a subframe. For example, a minimum time interval between the start of transmission of a physical channel and the start of transmission of the next physical channel is one subframe. However, a subframe may be replaced with a slot, a mini-slot, or multiple slots in the following description.
Machine Type Communication (MTC)
MTC is an application that does not require high throughput, applicable to machine-to-machine (M2M) or IoT. MTC is a communication technology which the 3GPP has adopted to satisfy the requirements of the IoT service.
While the following description is given mainly of features related to enhanced MTC (eMTC), the same thing is applicable to MTC, eMTC, and MTC to be applied to 5G (or NR), unless otherwise mentioned. The term MTC as used herein may be interchangeable with eMTC, LTE-M1/M2, bandwidth reduced low complexity (BL)/coverage enhanced (CE), non-BL UE (in enhanced coverage), NR MTC, enhanced BL/CE, and so on.
MTC General
(1) MTC operates only in a specific system BW (or channel BW).
MTC may use a predetermined number of RBs among the RBs of a system band in the legacy LTE system or the NR system. The operating frequency BW of MTC may be defined in consideration of a frequency range and a subcarrier spacing in NR. A specific system or frequency BW in which MTC operates is referred to as an MTC narrowband (NB) or MTC subband. In NR, MTC may operate in at least one BWP or a specific band of a BWP.
While MTC is supported by a cell having a much larger BW (e.g., 10 MHz) than 1.08 MHz, a physical channel and signal transmitted/received in MTC is always limited to 1.08 MHz or 6 (LTE) RBs. For example, a narrowband is defined as 6 non-overlapped consecutive physical resource blocks (PRBs) in the frequency domain in the LTE system.
In MTC, some DL and UL channels are allocated restrictively within a narrowband, and one channel does not occupy a plurality of narrowbands in one time unit.
An MTC narrowband may be configured for a UE by system information or DCI transmitted by a BS.
(2) MTC does not use a channel (defined in legacy LTE or NR) which is to be distributed across the total system BW of the legacy LTE or NR. For example, because a legacy LTE PDCCH is distributed across the total system BW, the legacy PDCCH is not used in MTC. Instead, a new control channel, MTC PDCCH (MPDCCH) is used in MTC. The MPDCCH is transmitted/received in up to 6 RBs in the frequency domain. In the time domain, the MPDCCH may be transmitted in one or more OFDM symbols starting with an OFDM symbol of a starting OFDM symbol index indicated by an RRC parameter from the BS among the OFDM symbols of a subframe.
(3) In MTC, PBCH, PRACH, MPDCCH, PDSCH, PUCCH, and PUSCH may be transmitted repeatedly. The MTC repeated transmissions may make these channels decodable even when signal quality or power is very poor as in a harsh condition like basement, thereby leading to the effect of an increased cell radius and signal penetration.
MTC Operation Modes and Levels
For CE, two operation modes, CE Mode A and CE Mode B and four different CE levels are used in MTC, as listed in [Table 2] below.
An MTC operation mode is determined by a BS and a CE level is determined by an MTC UE.
MTC Guard Period
The position of a narrowband used for MTC may change in each specific time unit (e.g., subframe or slot). An MTC UE may tune to different frequencies in different time units. A certain time may be required for frequency retuning and thus used as a guard period for MTC. No transmission and reception take place during the guard period.
MTC Signal Transmission/Reception Method
Apart from features inherent to MTC, an MTC signal transmission/reception procedure is similar to the procedure illustrated in
After acquiring synchronization with a BS by using the PSS/SSS, an MTC UE may acquire broadcast information within a cell by receiving a PBCH signal from the BS. The broadcast information transmitted on the PBCH is an MIB. In MTC, reserved bits among the bits of the legacy LTE MIB are used to transmit scheduling information for a new system information block 1 bandwidth reduced (SIB1-BR). The scheduling information for the SIB1-BR may include information about a repetition number and a TBS for a PDSCH conveying SIB1-BR. A frequency resource assignment for the PDSCH conveying SIB-BR may be a set of 6 consecutive RBs within a narrowband. The SIB-BR is transmitted directly on the PDSCH without a control channel (e.g., PDCCH or MPDCCH) associated with SIB-BR.
After completing the initial cell search, the MTC UE may acquire more specific system information by receiving an MPDCCH and a PDSCH based on information of the MPDCCH (S202).
Subsequently, the MTC UE may perform a RACH procedure to complete connection to the BS (S203 to S206). A basic configuration for the RACH procedure of the MTC UE may be transmitted in SIB2. Further, SIB2 includes paging-related parameters. In the 3GPP system, a paging occasion (PO) means a time unit in which a UE may attempt to receive paging. Paging refers to the network's indication of the presence of data to be transmitted to the UE. The MTC UE attempts to receive an MPDCCH based on a P-RNTI in a time unit corresponding to its PO in a narrowband configured for paging, paging narrowband (PNB). When the UE succeeds in decoding the MPDCCH based on the P-RNTI, the UE may check its paging message by receiving a PDSCH scheduled by the MPDCCH. In the presence of its paging message, the UE accesses the network by performing the RACH procedure.
In MTC, signals and/or messages (Msg1, Msg2, Msg3, and Msg4) may be transmitted repeatedly in the RACH procedure, and a different repetition pattern may be set according to a CE level.
For random access, PRACH resources for different CE levels are signaled by the BS. Different PRACH resources for up to 4 respective CE levels may be signaled to the MTC UE. The MTC UE measures an RSRP using a DL RS (e.g., CRS, CSI-RS, or TRS) and determines one of the CE levels signaled by the BS based on the measurement. The UE selects one of different PRACH resources (e.g., frequency, time, and preamble resources for a PARCH) for random access based on the determined CE level and transmits a PRACH. The BS may determine the CE level of the UE based on the PRACH resources that the UE has used for the PRACH transmission. The BS may determine a CE mode for the UE based on the CE level that the UE indicates by the PRACH transmission. The BS may transmit DCI to the UE in the CE mode.
Search spaces for an RAR for the PRACH and contention resolution messages are signaled in system information by the BS.
After the above procedure, the MTC UE may receive an MPDCCH signal and/or a PDSCH signal (S207) and transmit a PUSCH signal and/or a PUCCH signal (S208) in a general UL/DL signal transmission procedure. The MTC UE may transmit UCI on a PUCCH or a PUSCH to the BS.
Once an RRC connection for the MTC UE is established, the MTC UE attempts to receive an MDCCH by monitoring an MPDCCH in a configured search space in order to acquire UL and DL data allocations.
In legacy LTE, a PDSCH is scheduled by a PDCCH. Specifically, the PDCCH may be transmitted in the first N (N=1, 2 or 3) OFDM symbols of a subframe, and the PDSCH scheduled by the PDCCH is transmitted in the same subframe.
Compared to legacy LTE, an MPDCCH and a PDSCH scheduled by the MPDCCH are transmitted/received in different subframes in MTC. For example, an MPDCCH with a last repetition in subframe #n schedules a PDSCH starting in subframe #n+2. The MPDCCH may be transmitted only once or repeatedly. A maximum repetition number of the MPDCCH is configured for the UE by RRC signaling from the BS. DCI carried on the MPDCCH provides information on how many times the MPDCCH is repeated so that the UE may determine when the PDSCH transmission starts. For example, if DCI in an MPDCCH starting in subframe #n includes information indicating that the MPDCCH is repeated 10 times, the MPDCCH may end in subframe #n+9 and the PDSCH may start in subframe #n+11. The DCI carried on the MPDCCH may include information about a repetition number for a physical data channel (e.g., PUSCH or PDSCH) scheduled by the DCI. The UE may transmit/receive the physical data channel repeatedly in the time domain according to the information about the repetition number of the physical data channel scheduled by the DCI. The PDSCH may be scheduled in the same or different narrowband as or from a narrowband in which the MPDCCH scheduling the PDSCH is transmitted. When the MPDCCH and the PDSCH are in different narrowbands, the MTC UE needs to retune to the frequency of the narrowband carrying the PDSCH before decoding the PDSCH. For UL scheduling, the same timing as in legacy LTE may be followed. For example, an MPDCCH ending in subframe #n may schedule a PUSCH transmission starting in subframe #n+4. If a physical channel is repeatedly transmitted, frequency hopping is supported between different MTC subbands by RF retuning. For example, if a PDSCH is repeatedly transmitted in 32 subframes, the PDSCH is transmitted in the first 16 subframes in a first MTC subband, and in the remaining 16 subframes in a second MTC subband. MTC may operate in half-duplex mode.
Narrowband-Internet of Things (NB-IoT)
NB-IoT may refer to a system for supporting low complexity, low power consumption, and efficient use of frequency resources by a system BW corresponding to one RB of a wireless communication system (e.g., the LTE system or the NR system). NB-IoT may operate in half-duplex mode. NB-IoT may be used as a communication scheme for implementing IoT by supporting, for example, an MTC device (or UE) in a cellular system.
In NB-IoT, each UE perceives one RB as one carrier. Therefore, an RB and a carrier as mentioned in relation to NB-IoT may be interpreted as the same meaning.
While a frame structure, physical channels, multi-carrier operations, and general signal transmission/reception in relation to NB-IoT will be described below in the context of the legacy LTE system, the description is also applicable to the next generation system (e.g., the NR system). Further, the description of NB-IoT may also be applied to MTC serving similar technical purposes (e.g., low power, low cost, and coverage enhancement).
NB-IoT Frame Structure and Physical Resources
A different NB-IoT frame structure may be configured according to a subcarrier spacing. For example, for a subcarrier spacing of 15 kHz, the NB-IoT frame structure may be identical to that of a legacy system (e.g., the LTE system). For example, a 10-ms NB-IoT frame may include 10 1-ms NB-IoT subframes each including two 0.5-ms slots. Each 0.5-ms NB-IoT slot may include 7 OFDM symbols. In another example, for a BWP or cell/carrier having a subcarrier spacing of 3.75 kHz, a 10-ms NB-IoT frame may include five 2-ms NB-IoT subframes each including 7 OFDM symbols and one guard period (GP). Further, a 2-ms NB-IoT subframe may be represented in NB-IoT slots or NB-IoT resource units (RUs). The NB-IoT frame structures are not limited to the subcarrier spacings of 15 kHz and 3.75 kHz, and NB-IoT for other subcarrier spacings (e.g., 30 kHz) may also be considered by changing time/frequency units.
NB-IoT DL physical resources may be configured based on physical resources of other wireless communication systems (e.g., the LTE system or the NR system) except that a system BW is limited to a predetermined number of RBs (e.g., one RB, that is, 180 kHz). For example, if the NB-IoT DL supports only the 15-kHz subcarrier spacing as described before, the NB-IoT DL physical resources may be configured as a resource area in which the resource grid illustrated in
Like the NB-IoT DL physical resources, NB-IoT UL resources may also be configured by limiting a system BW to one RB. In NB-IoT, the number of UL subcarriers NULsc and a slot duration Tslot may be given as illustrated in [Table 3] below. In NB-IoT of the LTE system, the duration of one slot, Tslot is defined by 7 SC-FDMA symbols in the time domain.
In NB-IoT, RUs are used for mapping to REs of a PUSCH for NB-IoT (referred to as an NPUSCH). An RU may be defined by NULsymb*NULslot SC-FDMA symbols in the time domain by NRUsc consecutive subcarriers in the frequency domain. For example, NRUsc and NULsymb are listed in [Table 4] for a cell/carrier having an FDD frame structure and in [Table 5] for a cell/carrier having a TDD frame structure.
NB-IoT Physical Channels
OFDMA may be adopted for NB-IoT DL based on the 15-kHz subcarrier spacing. Because OFDMA provides orthogonality between subcarriers, co-existence with other systems (e.g., the LTE system or the NR system) may be supported efficiently. The names of DL physical channels/signals of the NB-IoT system may be prefixed with “N (narrowband)” to be distinguished from their counterparts in the legacy system. For example, DL physical channels may be named NPBCH, NPDCCH, NPDSCH, and so on, and DL physical signals may be named NPSS, NSSS, narrowband reference signal (NRS), narrowband positioning reference signal (NPRS), narrowband wake up signal (NWUS), and so on. The DL channels, NPBCH, NPDCCH, NPDSCH, and so on may be repeatedly transmitted to enhance coverage in the NB-IoT system. Further, new defined DCI formats may be used in NB-IoT, such as DCI format N0, DCI format N1, and DCI format N2.
SC-FDMA may be applied with the 15-kHz or 3.75-kHz subcarrier spacing to NB-IoT UL. As described in relation to DL, the names of physical channels of the NB-IoT system may be prefixed with “N (narrowband)” to be distinguished from their counterparts in the legacy system. For example, UL channels may be named NPRACH, NPUSCH, and so on, and UL physical signals may be named NDMRS and so on. NPUSCHs may be classified into NPUSCH format 1 and NPUSCH format 2. For example, NPUSCH format 1 may be used to transmit (or deliver) an uplink shared channel (UL-SCH), and NPUSCH format 2 may be used for UCI transmission such as HARQ ACK signaling. A UL channel, NPRACH in the NB-IoT system may be repeatedly transmitted to enhance coverage. In this situation, the repeated transmissions may be subjected to frequency hopping.
Multi-Carrier Operation in NB-IoT
NB-IoT may be implemented in multi-carrier mode. A multi-carrier operation may refer to using multiple carriers configured for different usages (i.e., multiple carriers of different types) in transmitting/receiving channels and/or signals between a BS and a UE.
In the multi-carrier mode in NB-IoT, carriers may be divided into anchor type carrier (i.e., anchor carrier or anchor PRB) and non-anchor type carrier (i.e., non-anchor carrier or non-anchor PRB).
The anchor carrier may refer to a carrier carrying an NPSS, an NSSS, and an NPBCH for initial access, and an NPDSCH for a system information block, N-SIB from the perspective of a BS. That is, a carrier for initial access is referred to as an anchor carrier, and the other carrier(s) is referred to as a non-anchor carrier in NB-IoT.
NB-IoT Signal Transmission/Reception Process
In NB-IoT, a signal is transmitted/received in a similar manner to the procedure illustrated in
Upon completion of the initial cell search, the NB-IoT UE may acquire more specific system information by receiving an NPDCCH and receiving an NPDSCH corresponding to the NPDCCH (S202). In other words, the BS may transmit more specific system information to the NB-IoT UE which has completed the initial call search by transmitting an NPDCCH and an NPDSCH corresponding to the NPDCCH.
The NB-IoT UE may then perform a RACH procedure to complete a connection setup with the BS (S203 to S206). For this purpose, the NB-IoT UE may transmit a preamble on an NPRACH to the BS (S203). As described before, it may be configured that the NPRACH is repeatedly transmitted based on frequency hopping, for coverage enhancement. In other words, the BS may (repeatedly) receive the preamble on the NPRACH from the NB-IoT UE. The NB-IoT UE may then receive an NPDCCH, and a RAR in response to the preamble on an NPDSCH corresponding to the NPDCCH from the BS (S204). In other words, the BS may transmit the NPDCCH, and the RAR in response to the preamble on the NPDSCH corresponding to the NPDCCH to the NB-IoT UE. Subsequently, the NB-IoT UE may transmit an NPUSCH to the BS, using scheduling information in the RAR (S205) and perform a contention resolution procedure by receiving an NPDCCH and an NPDSCH corresponding to the NPDCCH (S206).
After the above process, the NB-IoT UE may perform an NPDCCH/NPDSCH reception (S207) and an NPUSCH transmission (S208) in a general UL/DL signal transmission procedure. In other words, after the above process, the BS may perform an NPDCCH/NPDSCH transmission and an NPUSCH reception with the NB-IoT UE in the general UL/DL signal transmission procedure.
In NB-IoT, the NPBCH, the NPDCCH, and the NPDSCH may be transmitted repeatedly, for coverage enhancement. A UL-SCH (i.e., general UL data) and UCI may be delivered on the PUSCH in NB-IoT. It may be configured that the UL-SCH and the UCI are transmitted in different NPUSCH formats (e.g., NPUSCH format 1 and NPUSCH format 2).
In NB-IoT, UCI may generally be transmitted on an NPUSCH. Further, the UE may transmit the NPUSCH periodically, aperiodically, or semi-persistently according to request/indication of the network (e.g., BS).
Wireless Communication Apparatus
Referring to
Each of the first communication device 910 and the second communication device 920 includes a processor 911 or 921, a memory 914 or 924, one or more Tx/Rx RF modules 915 or 925, a Tx processor 912 or 922, an Rx processor 913 or 923, and antennas 916 or 926. A Tx/Rx module may also be called a transceiver. The processor performs the afore-described functions, processes, and/or methods. More specifically, on DL (communication from the first communication device 910 to the second communication device 920), a higher-layer packet from a core network is provided to the processor 911. The processor 911 implements Layer 2 (i.e., L2) functionalities. On DL, the processor 911 is responsible for multiplexing between a logical channel and a transport channel, provisioning of a radio resource assignment to the second communication device 920, and signaling to the second communication device 920. The Tx processor 912 executes various signal processing functions of L1 (i.e., the physical layer). The signal processing functions facilitate forward error correction (FEC) of the second communication device 920, including coding and interleaving. An encoded and interleaved signal is modulated to complex-valued modulation symbols after scrambling and modulation. For the modulation, BPSK, QPSK, 16QAM, 64QAM, 246QAM, and so on are available according to channels. The complex-valued modulation symbols (hereinafter, referred to as modulation symbols) are divided into parallel streams. Each stream is mapped to OFDM subcarriers and multiplexed with an RS in the time and/or frequency domain. A physical channel is generated to carry a time-domain OFDM symbol stream by subjecting the mapped signals to IFFT. The OFDM symbol stream is spatially precoded to multiple spatial streams. Each spatial stream may be provided to a different antenna 916 through an individual Tx/Rx module (or transceiver) 915. Each Tx/Rx module 915 may upconvert the frequency of each spatial stream to an RF carrier, for transmission. In the second communication device 920, each Tx/Rx module (or transceiver) 925 receives a signal of the RF carrier through each antenna 926. Each Tx/Rx module 925 recovers the signal of the RF carrier to a baseband signal and provides the baseband signal to the Rx processor 923. The Rx processor 923 executes various signal processing functions of L1 (i.e., the physical layer). The Rx processor 923 may perform spatial processing on information to recover any spatial stream directed to the second communication device 920. If multiple spatial streams are directed to the second communication device 920, multiple Rx processors may combine the multiple spatial streams into a single OFDMA symbol stream. The Rx processor 923 converts an OFDM symbol stream being a time-domain signal to a frequency-domain signal by FFT. The frequency-domain signal includes an individual OFDM symbol stream on each subcarrier of an OFDM signal. Modulation symbols and an RS on each subcarrier are recovered and demodulated by determining most likely signal constellation points transmitted by the first communication device 910. These soft decisions may be based on channel estimates. The soft decisions are decoded and deinterleaved to recover the original data and control signal transmitted on physical channels by the first communication device 910. The data and control signal are provided to the processor 921.
On UL (communication from the second communication device 920 to the first communication device 910), the first communication device 910 operates in a similar manner as described in relation to the receiver function of the second communication device 920. Each Tx/Rx module 925 receives a signal through an antenna 926. Each Tx/Rx module 925 provides an RF carrier and information to the Rx processor 923. The processor 921 may be related to the memory 924 storing a program code and data. The memory 924 may be referred to as a computer-readable medium.
Artificial Intelligence (AI)
Artificial intelligence is a field of studying AI or methodologies for creating AI, and machine learning is a field of defining various issues dealt with in the AI field and studying methodologies for addressing the various issues. Machine learning is defined as an algorithm that increases the performance of a certain operation through steady experiences for the operation.
An artificial neural network (ANN) is a model used in machine learning and may refer to a model having a problem-solving ability, which is composed of artificial neurons (nodes) forming a network via synaptic connections. The ANN may be defined by a connection pattern between neurons in different layers, a learning process for updating model parameters, and an activation function for generating an output value.
The ANN may include an input layer, an output layer, and optionally, one or more hidden layers. Each layer includes one or more neurons, and the ANN may include a synapse that links between neurons. In the ANN, each neuron may output the function value of the activation function, for the input of signals, weights, and deflections through the synapse.
Model parameters refer to parameters determined through learning and include a weight value of a synaptic connection and deflection of neurons. A hyperparameter means a parameter to be set in the machine learning algorithm before learning, and includes a learning rate, a repetition number, a mini batch size, and an initialization function.
The purpose of learning of the ANN may be to determine model parameters that minimize a loss function. The loss function may be used as an index to determine optimal model parameters in the learning process of the ANN.
Machine learning may be classified into supervised learning, unsupervised learning, and reinforcement learning according to learning methods.
Supervised learning may be a method of training an ANN in a state in which a label for training data is given, and the label may mean a correct answer (or result value) that the ANN should infer with respect to the input of training data to the ANN. Unsupervised learning may be a method of training an ANN in a state in which a label for training data is not given. Reinforcement learning may be a learning method in which an agent defined in a certain environment is trained to select a behavior or a behavior sequence that maximizes cumulative compensation in each state.
Machine learning, which is implemented by a deep neural network (DNN) including a plurality of hidden layers among ANNs, is also referred to as deep learning, and deep learning is part of machine learning. The following description is given with the appreciation that machine learning includes deep learning.
<Robot>
A robot may refer to a machine that automatically processes or executes a given task by its own capabilities. Particularly, a robot equipped with a function of recognizing an environment and performing an operation based on its decision may be referred to as an intelligent robot.
Robots may be classified into industrial robots, medical robots, consumer robots, military robots, and so on according to their usages or application fields.
A robot may be provided with a driving unit including an actuator or a motor, and thus perform various physical operations such as moving robot joints. Further, a movable robot may include a wheel, a brake, a propeller, and the like in a driving unit, and thus travel on the ground or fly in the air through the driving unit.
<Self-Driving>
Self-driving refers to autonomous driving, and a self-driving vehicle refers to a vehicle that travels with no user manipulation or minimum user manipulation.
For example, self-driving may include a technology of maintaining a lane while driving, a technology of automatically adjusting a speed, such as adaptive cruise control, a technology of automatically traveling along a predetermined route, and a technology of automatically setting a route and traveling along the route when a destination is set.
Vehicles may include a vehicle having only an internal combustion engine, a hybrid vehicle having both an internal combustion engine and an electric motor, and an electric vehicle having only an electric motor, and may include not only an automobile but also a train, a motorcycle, and the like.
Herein, a self-driving vehicle may be regarded as a robot having a self-driving function.
<eXtended Reality (XR)>
Extended reality is a term covering virtual reality (VR), augmented reality (AR), and mixed reality (MR). VR provides a real-world object and background only as a computer graphic (CG) image, AR provides a virtual CG image on a real object image, and MR is a computer graphic technology that mixes and combines virtual objects into the real world.
MR is similar to AR in that the real object and the virtual object are shown together. However, in AR, the virtual object is used as a complement to the real object, whereas in MR, the virtual object and the real object are handled equally.
XR may be applied to a head-mounted display (HMD), a head-up display (HUD), a portable phone, a tablet PC, a laptop computer, a desktop computer, a TV, a digital signage, and so on. A device to which XR is applied may be referred to as an XR device.
The AI device 1000 illustrated in
Referring to
The communication unit 1010 may transmit and receive data to and from an external device such as another AI device or an AI server by wired or wireless communication. For example, the communication unit 1010 may transmit and receive sensor information, a user input, a learning model, and a control signal to and from the external device.
Communication schemes used by the communication unit 1010 include global system for mobile communication (GSM), CDMA, LTE, 5G, wireless local area network (WLAN), wireless fidelity (Wi-Fi), Bluetooth™, radio frequency identification (RFID), infrared data association (IrDA), ZigBee, near field communication (NFC), and so on. Particularly, the 5G technology described before with reference to
The input unit 1020 may acquire various types of data. The input unit 1020 may include a camera for inputting a video signal, a microphone for receiving an audio signal, and a user input unit for receiving information from a user. The camera or the microphone may be treated as a sensor, and thus a signal acquired from the camera or the microphone may be referred to as sensing data or sensor information.
The input unit 1020 may acquire training data for model training and input data to be used to acquire an output by using a learning model. The input unit 1020 may acquire raw input data. In this situation, the processor 1080 or the learning processor 1030 may extract an input feature by preprocessing the input data.
The learning processor 1030 may train a model composed of an ANN by using training data. The trained ANN may be referred to as a learning model. The learning model may be used to infer a result value for new input data, not training data, and the inferred value may be used as a basis for determination to perform a certain operation.
The learning processor 1030 may perform AI processing together with a learning processor of an AI server.
The learning processor 1030 may include a memory integrated or implemented in the AI device 1000. Alternatively, the learning processor 1030 may be implemented by using the memory 1070, an external memory directly connected to the AI device 1000, or a memory maintained in an external device.
The sensing unit 1040 may acquire at least one of internal information about the AI device 1000, ambient environment information about the AI device 1000, and user information by using various sensors.
The sensors included in the sensing unit 1040 may include a proximity sensor, an illumination sensor, an accelerator sensor, a magnetic sensor, a gyro sensor, an inertial sensor, a red, green, blue (RGB) sensor, an IR sensor, a fingerprint recognition sensor, an ultrasonic sensor, an optical sensor, a microphone, a light detection and ranging (LiDAR), and a radar.
The output unit 1050 may generate a visual, auditory, or haptic output.
Accordingly, the output unit 1050 may include a display unit for outputting visual information, a speaker for outputting auditory information, and a haptic module for outputting haptic information.
The memory 1070 may store data that supports various functions of the AI device 1000. For example, the memory 1070 may store input data acquired by the input unit 1020, training data, a learning model, a learning history, and so on.
The processor 1080 may determine at least one executable operation of the AI device 100 based on information determined or generated by a data analysis algorithm or a machine learning algorithm. The processor 1080 may control the components of the AI device 1000 to execute the determined operation.
To this end, the processor 1080 may request, search, receive, or utilize data of the learning processor 1030 or the memory 1070. The processor 1080 may control the components of the AI device 1000 to execute a predicted operation or an operation determined to be desirable among the at least one executable operation.
When the determined operation needs to be performed in conjunction with an external device, the processor 1080 may generate a control signal for controlling the external device and transmit the generated control signal to the external device.
The processor 1080 may acquire intention information with respect to a user input and determine the user's requirements based on the acquired intention information.
The processor 1080 may acquire the intention information corresponding to the user input by using at least one of a speech to text (STT) engine for converting a speech input into a text string or a natural language processing (NLP) engine for acquiring intention information of a natural language.
At least one of the STT engine or the NLP engine may be configured as an ANN, at least part of which is trained according to the machine learning algorithm. At least one of the STT engine or the NLP engine may be trained by the learning processor, a learning processor of the AI server, or distributed processing of the learning processors. For reference, specific components of the AI server are illustrated in
The processor 1080 may collect history information including the operation contents of the AI device 1000 or the user's feedback on the operation and may store the collected history information in the memory 1070 or the learning processor 1030 or transmit the collected history information to the external device such as the AI server. The collected history information may be used to update the learning model.
The processor 1080 may control at least a part of the components of AI device 1000 to drive an application program stored in the memory 1070. Furthermore, the processor 1080 may operate two or more of the components included in the AI device 1000 in combination to drive the application program.
Referring to
The AI server 1120 may include a communication unit 1121, a memory 1123, a learning processor 1122, a processor 1126, and so on.
The communication unit 1121 may transmit and receive data to and from an external device such as the AI device 1100.
The memory 1123 may include a model storage 1124. The model storage 1124 may store a model (or an ANN 1125) which has been trained or is being trained through the learning processor 1122.
The learning processor 1122 may train the ANN 1125 by training data. The learning model may be used, while being loaded on the AI server 1120 of the ANN, or on an external device such as the AI device 1110.
The learning model may be implemented in hardware, software, or a combination of hardware and software. If all or part of the learning model is implemented in software, one or more instructions of the learning model may be stored in the memory 1123.
The processor 1126 may infer a result value for new input data by using the learning model and may generate a response or a control command based on the inferred result value.
Referring to
The cloud network 1200 may refer to a network that forms part of cloud computing infrastructure or exists in the cloud computing infrastructure. The cloud network 1200 may be configured by using a 3G network, a 4G or LTE network, or a 5G network.
That is, the devices 1210 to 1260 included in the AI system may be interconnected via the cloud network 1200. In particular, each of the devices 1210 to 1260 may communicate with each other directly or through a BS.
The AI server 1260 may include a server that performs AI processing and a server that performs computation on big data.
The AI server 1260 may be connected to at least one of the AI devices included in the AI system, that is, at least one of the robot 1210, the self-driving vehicle 1220, the XR device 1230, the smartphone 1240, or the home appliance 1250 via the cloud network 1200, and may assist at least part of AI processing of the connected AI devices 1210 to 1250.
The AI server 1260 may train the ANN according to the machine learning algorithm on behalf of the AI devices 1210 to 1250, and may directly store the learning model or transmit the learning model to the AI devices 1210 to 1250.
The AI server 1260 may receive input data from the AI devices 1210 to 1250, infer a result value for received input data by using the learning model, generate a response or a control command based on the inferred result value, and transmit the response or the control command to the AI devices 1210 to 1250.
Alternatively, the AI devices 1210 to 1250 may infer the result value for the input data by directly using the learning model, and generate the response or the control command based on the inference result.
Hereinafter, various embodiments of the AI devices 1210 to 1250 to which the above-described technology is applied will be described. The AI devices 1210 to 1250 illustrated in
<AI+XR>
The XR device 1230, to which AI is applied, may be configured as a HMD, a HUD provided in a vehicle, a TV, a portable phone, a smartphone, a computer, a wearable device, a home appliance, a digital signage, a vehicle, a fixed robot, a mobile robot, or the like.
The XR device 1230 may acquire information about a surrounding space or a real object by analyzing 3D point cloud data or image data acquired from various sensors or an external device and thus generating position data and attribute data for the 3D points, and may render an XR object to be output. For example, the XR device 1230 may output an XR object including additional information about a recognized object in correspondence with the recognized object.
The XR device 1230 may perform the above-described operations by using the learning model composed of at least one ANN. For example, the XR device 1230 may recognize a real object from 3D point cloud data or image data by using the learning model, and may provide information corresponding to the recognized real object. The learning model may be trained directly by the XR device 1230 or by the external device such as the AI server 1260.
While the XR device 1230 may operate by generating a result by directly using the learning model, the XR device 1230 may operate by transmitting sensor information to the external device such as the AI server 1260 and receiving the result.
<AI+Robot+XR>
The robot 1210, to which AI and XR are applied, may be implemented as a guide robot, a delivery robot, a cleaning robot, a wearable robot, an entertainment robot, a pet robot, an unmanned flying robot, a drone, or the like.
The robot 1210, to which XR is applied, may refer to a robot to be controlled/interact within an XR image. In this situation, the robot 1210 may be distinguished from the XR device 1230 and interwork with the XR device 1230.
When the robot 1210 to be controlled/interact within an XR image acquires sensor information from sensors each including a camera, the robot 1210 or the XR device 1230 may generate an XR image based on the sensor information, and the XR device 1230 may output the generated XR image. The robot 1210 may operate based on the control signal received through the XR device 1230 or based on the user's interaction.
For example, the user may check an XR image corresponding to a view of the robot 1210 interworking remotely through an external device such as the XR device 1210, adjust a self-driving route of the robot 1210 through interaction, control the operation or driving of the robot 1210, or check information about an ambient object around the robot 1210.
<AI+Self-Driving+XR>
The self-driving vehicle 1220, to which AI and XR are applied, may be implemented as a mobile robot, a vehicle, an unmanned flying vehicle, or the like.
The self-driving driving vehicle 1220, to which XR is applied, may refer to a self-driving vehicle provided with a means for providing an XR image or a self-driving vehicle to be controlled/interact within an XR image. Particularly, the self-driving vehicle 1220 to be controlled/interact within an XR image may be distinguished from the XR device 1230 and interwork with the XR device 1230.
The self-driving vehicle 1220 provided with the means for providing an XR image may acquire sensor information from the sensors each including a camera and output the generated XR image based on the acquired sensor information. For example, the self-driving vehicle 1220 may include an HUD to output an XR image, thereby providing a passenger with an XR object corresponding to a real object or an object on the screen.
When the XR object is output to the HUD, at least part of the XR object may be output to be overlaid on an actual object to which the passenger's gaze is directed. When the XR object is output to a display provided in the self-driving vehicle 1220, at least part of the XR object may be output to be overlaid on the object within the screen. For example, the self-driving vehicle 1220 may output XR objects corresponding to objects such as a lane, another vehicle, a traffic light, a traffic sign, a two-wheeled vehicle, a pedestrian, a building, and so on.
When the self-driving vehicle 1220 to be controlled/interact within an XR image acquires sensor information from the sensors each including a camera, the self-driving vehicle 1220 or the XR device 1230 may generate the XR image based on the sensor information, and the XR device 1230 may output the generated XR image. The self-driving vehicle 1220 may operate based on a control signal received through an external device such as the XR device 1230 or based on the user's interaction.
VR, AR, and MR technologies of the present disclosure are applicable to various devices, particularly, for example, a HMD, a HUD attached to a vehicle, a portable phone, a tablet PC, a laptop computer, a desktop computer, a TV, and a signage. The VR, AR, and MR technologies may also be applicable to a device equipped with a flexible or rollable display.
The above-described VR, AR, and MR technologies may be implemented based on CG and distinguished by the ratios of a CG image in an image viewed by the user.
That is, VR provides a real object or background only in a CG image, whereas AR overlays a virtual CG image on an image of a real object.
MR is similar to AR in that virtual objects are mixed and combined with a real world. However, a real object and a virtual object created as a CG image are distinctive from each other and the virtual object is used to complement the real object in AR, whereas a virtual object and a real object are handled equally in MR. More specifically, for example, a hologram service is an MR representation.
These days, VR, AR, and MR are collectively called XR without distinction among them. Therefore, embodiments of the present disclosure are applicable to all of VR, AR, MR, and XR.
For example, wired/wireless communication, input interfacing, output interfacing, and computing devices are available as hardware (HW)-related element techniques applied to VR, AR, MR, and XR. Further, tracking and matching, speech recognition, interaction and user interfacing, location-based service, search, and AI are available as software (SW)-related element techniques.
Particularly, the embodiments of the present disclosure are intended to address at least one of the issues of communication with another device, efficient memory use, data throughput decrease caused by inconvenient user experience/user interface (UX/UI), video, sound, motion sickness, or other issues.
The communication module 1360 may communicate with an external device or a server, in a wired manner or wirelessly. The communication module 1360 may use, for example, Wi-Fi, Bluetooth, or the like, for short-range wireless communication, and for example, a 3GPP communication standard for long-range wireless communication. LTE is a technology beyond 3GPP TS 36.xxx Release 8. Specifically, LTE beyond 3GPP TS 36.xxx Release 10 is referred to as LTE-A, and LTE beyond 3GPP TS 36.xxx Release 13 is referred to as LTE-A pro. 3GPP 5G refers to a technology beyond TS 36.xxx Release 15 and a technology beyond TS 38.XXX Release 15. Specifically, the technology beyond TS 38.xxx Release 15 is referred to as 3GPP NR, and the technology beyond TS 36.xxx Release 15 is referred to as enhanced LTE. “xxx” represents the number of a technical specification. LTE/NR may be collectively referred to as a 3GPP system.
The camera 1310 may capture an ambient environment of the XR device 1300 and convert the captured image to an electric signal. The image, which has been captured and converted to an electric signal by the camera 1310, may be stored in the memory 1350 and then displayed on the display 1320 through the processor 1340. Further, the image may be displayed on the display 1320 by the processor 1340, without being stored in the memory 1350. Further, the camera 110 may have a field of view (FoV). The FoV is, for example, an area in which a real object around the camera 1310 may be detected. The camera 1310 may detect only a real object within the FoV. When a real object is located within the FoV of the camera 1310, the XR device 1300 may display an AR object corresponding to the real object. Further, the camera 1310 may detect an angle between the camera 1310 and the real object.
The sensor 1330 may include at least one sensor. For example, the sensor 1330 includes a sensing means such as a gravity sensor, a geomagnetic sensor, a motion sensor, a gyro sensor, an accelerator sensor, an inclination sensor, a brightness sensor, an altitude sensor, an olfactory sensor, a temperature sensor, a depth sensor, a pressure sensor, a bending sensor, an audio sensor, a video sensor, a global positioning system (GPS) sensor, and a touch sensor. Further, although the display 1320 may be of a fixed type, the display 1320 may be configured as a liquid crystal display (LCD), an organic light emitting diode (OLED) display, an electroluminescent display (ELD), or a micro LED (M-LED) display, to have flexibility. Herein, the sensor 1330 is designed to detect a bending degree of the display 1320 configured as the afore-described LCD, OLED display, ELD, or M-LED display.
The memory 1350 is equipped with a function of storing all or a part of result values obtained by wired/wireless communication with an external device or a service as well as a function of storing an image captured by the camera 1310. Particularly, considering the trend toward increased communication data traffic (e.g., in a 5G communication environment), efficient memory management is required. In this regard, a description will be given below with reference to
When swapping out AR/VR page data from a RAM 1410 to a flash memory 1420, a controller 1430 may swap out only one of two or more AR/VR page data of the same contents among AR/VR page data to be swapped out to the flash memory 1420.
That is, the controller 1430 may calculate an identifier (e.g., a hash function) that identifies each of the contents of the AR/VR page data to be swapped out, and determine that two or more AR/VR page data having the same identifier among the calculated identifiers contain the same contents. Accordingly, the problem that the lifetime of an AR/VR device including the flash memory 1420 as well as the lifetime of the flash memory 1420 is reduced because unnecessary AR/VR page data is stored in the flash memory 1420 may be overcome.
The operations of the controller 1430 may be implemented in software or hardware without departing from the scope of the present disclosure. More specifically, the memory illustrated in
A device according to embodiments of the present disclosure may process 3D point cloud data to provide various services such as VR, AR, MR, XR, and self-driving to a user.
A sensor collecting 3D point cloud data may be any of, for example, a LiDAR, a red, green, blue depth (RGB-D), and a 3D laser scanner. The sensor may be mounted inside or outside of a HMD, a vehicle, a portable phone, a tablet PC, a laptop computer, a desktop computer, a TV, a signage, or the like.
Referring to
The device or processor according to embodiments of the present disclosure may acquire one or more bit streams and related metadata by decapsulating the received video data, and recover 3D point cloud data by decoding the acquired bit streams in V-PCC or G-PCC (S1540). A renderer may render the decoded point cloud data and provide content suitable for VR/AR/MR/service to the user on a display (S1550).
As illustrated in
Referring to
According to embodiments of the present disclosure, a learning processor 1670 may be coupled communicably to a processor 1640, and repeatedly train a model including ANNs by using training data. An ANN is an information processing system in which multiple neurons are linked in layers, modeling an operation principle of biological neurons and links between neurons. An ANN is a statistical learning algorithm inspired by a neural network (particularly the brain in the central nervous system of an animal) in machine learning and cognitive science. Machine learning is one field of AI, in which the ability of learning without an explicit program is granted to a computer. Machine learning is a technology of studying and constructing a system for learning, predicting, and improving its capability based on empirical data, and an algorithm for the system. Therefore, according to embodiments of the present disclosure, the learning processor 1670 may infer a result value from new input data by determining optimized model parameters of an ANN. Therefore, the learning processor 1670 may analyze a device use pattern of a user based on device use history information about the user. Further, the learning processor 1670 may be configured to receive, classify, store, and output information to be used for data mining, data analysis, intelligent decision, and a machine learning algorithm and technique.
According to embodiments of the present disclosure, the processor 1640 may determine or predict at least one executable operation of the device based on data analyzed or generated by the learning processor 1670. Further, the processor 1640 may request, search, receive, or use data of the learning processor 1670, and control the XR device 1600 to perform a predicted operation or an operation determined to be desirable among the at least one executable operation. According to embodiments of the present disclosure, the processor 1640 may execute various functions of realizing intelligent emulation (i.e., knowledge-based system, reasoning system, and knowledge acquisition system). The various functions may be applied to an adaptation system, a machine learning system, and various types of systems including an ANN (e.g., a fuzzy logic system). That is, the processor 1640 may predict a user's device use pattern based on data of a use pattern analyzed by the learning processor 1670, and control the XR device 1600 to provide a more suitable XR service to the UE. Herein, the XR service includes at least one of the AR service, the VR service, or the MR service.
According to embodiments of the present disclosure, the processor 1670 may store device use history information about a user in the memory 1650 (S1710). The device use history information may include information about the name, category, and contents of content provided to the user, information about a time at which a device has been used, information about a place in which the device has been used, time information, and information about use of an application installed in the device.
According to embodiments of the present disclosure, the learning processor 1670 may acquire device use pattern information about the user by analyzing the device use history information (S1720). For example, when the XR device 1600 provides specific content A to the user, the learning processor 1670 may learn information about a pattern of the device used by the user using the corresponding terminal by combining specific information about content A (e.g., information about the ages of users that generally use content A, information about the contents of content A, and content information similar to content A), and information about the time points, places, and number of times in which the user using the corresponding terminal has consumed content A.
According to embodiments of the present disclosure, the processor 1640 may acquire the user device pattern information generated based on the information learned by the learning processor 1670, and generate device use pattern prediction information (S1730). Further, when the user is not using the device 1600, if the processor 1640 determines that the user is located in a place where the user has frequently used the device 1600, or it is almost time for the user to usually use the device 1600, the processor 1640 may indicate the device 1600 to operate. In this situation, the device according to embodiments of the present disclosure may provide AR content based on the user pattern prediction information (S1740).
When the user is using the device 1600, the processor 1640 may check information about content currently provided to the user, and generate device use pattern prediction information about the user in relation to the content (e.g., when the user requests other related content or additional data related to the current content). Further, the processor 1640 may provide AR content based on the device use pattern prediction information by indicating the device 1600 to operate (S1740). The AR content according to embodiments of the present disclosure may include an advertisement, navigation information, danger information, and so on.
Component modules of an XR device 1800 according to an embodiment of the present disclosure have been described before with reference to the previous drawings, and thus a redundant description is not provided herein.
The outer appearance of a robot 1810 illustrated in
The robot 1810 may be provided, on the exterior thereof, with various sensors to identify ambient objects. Further, to provide specific information to a user, the robot 1810 may be provided with an interface unit 1811 on top or the rear surface 1812 thereof.
To sense movement of the robot 1810 and an ambient object, and control the robot 1810, a robot control module 1850 is mounted inside the robot 1810. The robot control module 1850 may be implemented as a software module or a hardware chip with the software module implemented therein. The robot control module 1850 may include a deep learner 1851, a sensing information processor 1852, a movement path generator 1853, and a communication module 1854.
The sensing information processor 1852 collects and processes information sensed by various types of sensors (e.g., a LiDAR sensor, an IR sensor, an ultrasonic sensor, a depth sensor, an image sensor, and a microphone) arranged in the robot 1810.
The deep learner 1851 may receive information processed by the sensing information processor 1851 or accumulative information stored during movement of the robot 1810, and output a result required for the robot 1810 to determine an ambient situation, process information, or generate a moving path.
The moving path generator 1852 may calculate a moving path of the robot 1810 by using the data calculated by the deep learner 8151 or the data processed by the sensing information processor 1852.
Because each of the XR device 1800 and the robot 1810 is provided with a communication module, the XR device 1800 and the robot 1810 may transmit and receive data by short-range wireless communication such as Wi-Fi or Bluetooth, or 5G long-range wireless communication. A technique of controlling the robot 1810 by using the XR device 1800 will be described below with reference to
The XR device and the robot are connected communicably to a 5G network (S1901). Obviously, the XR device and the robot may transmit and receive data by any other short-range or long-range communication technology without departing from the scope of the present disclosure.
The robot captures an image/video of the surroundings of the robot by means of at least one camera installed on the interior or exterior of the robot (S1902) and transmits the captured image/video to the XR device (S1903). The XR device displays the captured image/video (S1904) and transmits a command for controlling the robot to the robot (S1905). The command may be input manually by a user of the XR device or automatically generated by AI without departing from the scope of the disclosure.
The robot executes a function corresponding to the command received in step S1905 (S1906) and transmits a result value to the XR device (S1907). The result value may be a general indicator indicating whether data has been successfully processed or not, a current captured image, or specific data in which the XR device is considered. The specific data is designed to change, for example, according to the state of the XR device. If a display of the XR device is in an off state, a command for turning on the display of the XR device is included in the result value in step S1907. Therefore, when an emergency situation occurs around the robot, even though the display of the remote XR device is turned off, a notification message may be transmitted.
AR/VR content is displayed according to the result value received in step S1907 (S1908).
According to another embodiment of the present disclosure, the XR device may display position information about the robot by using a GPS module attached to the robot.
The XR device 1300 described with reference to
According to embodiments of the present disclosure, a vehicle 2010 may include a car, a train, and a motor bike as transportation means traveling on a road or a railway. According to embodiments of the present disclosure, the vehicle 2010 may include all of an internal combustion engine vehicle provided with an engine as a power source, a hybrid vehicle provided with an engine and an electric motor as a power source, and an electric vehicle provided with an electric motor as a power source.
According to embodiments of the present disclosure, the vehicle 2010 may include the following components in order to control operations of the vehicle 2010: a user interface device, an object detection device, a communication device, a driving maneuver device, a main electronic control unit (ECU), a drive control device, a self-driving device, a sensing unit, and a position data generation device.
Each of the user interface device, the object detection device, the communication device, the driving maneuver device, the main ECU, the drive control device, the self-driving device, the sensing unit, and the position data generation device may generate an electric signal, and be implemented as an electronic device that exchanges electric signals.
The user interface device may receive a user input and provide information generated from the vehicle 2010 to a user in the form of a UI or UX. The user interface device may include an input/output (I/O) device and a user monitoring device. The object detection device may detect the presence or absence of an object outside of the vehicle 2010, and generate information about the object. The object detection device may include at least one of, for example, a camera, a LiDAR, an IR sensor, or an ultrasonic sensor. The camera may generate information about an object outside of the vehicle 2010. The camera may include one or more lenses, one or more image sensors, and one or more processors for generating object information. The camera may acquire information about the position, distance, or relative speed of an object by various image processing algorithms. Further, the camera may be mounted at a position where the camera may secure an FoV in the vehicle 2010, to capture an image of the surroundings of the vehicle 1020, and may be used to provide an AR/VR-based service. The LiDAR may generate information about an object outside of the vehicle 2010. The LiDAR may include a light transmitter, a light receiver, and at least one processor which is electrically coupled to the light transmitter and the light receiver, processes a received signal, and generates data about an object based on the processed signal.
The communication device may exchange signals with a device (e.g., infrastructure such as a server or a broadcasting station), another vehicle, or a terminal) outside of the vehicle 2010. The driving maneuver device is a device that receives a user input for driving. In manual mode, the vehicle 2010 may travel based on a signal provided by the driving maneuver device. The driving maneuver device may include a steering input device (e.g., a steering wheel), an acceleration input device (e.g., an accelerator pedal), and a brake input device (e.g., a brake pedal).
The sensing unit may sense a state of the vehicle 2010 and generate state information. The position data generation device may generate position data of the vehicle 2010. The position data generation device may include at least one of a GPS or a differential global positioning system (DGPS). The position data generation device may generate position data of the vehicle 2010 based on a signal generated from at least one of the GPS or the DGPS. The main ECU may provide overall control to at least one electronic device provided in the vehicle 2010, and the drive control device may electrically control a vehicle drive device in the vehicle 2010.
The self-driving device may generate a path for the self-driving service based on data acquired from the object detection device, the sensing unit, the position data generation device, and so on. The self-driving device may generate a driving plan for driving along the generated path, and generate a signal for controlling movement of the vehicle according to the driving plan. The signal generated from the self-driving device is transmitted to the drive control device, and thus the drive control device may control the vehicle drive device in the vehicle 2010.
As illustrated in
If the XR device 2000 is connected to the vehicle 2010 in a manner that allows wired/wireless communication. The XR device 2000 may receive/process AR/VR service-related content data that may be provided along with the self-driving service, and transmit the received/processed AR/VR service-related content data to the vehicle 2010. Further, when the XR device 2000 is mounted on the vehicle 2010, the XR device 2000 may receive/process AR/VR service-related content data according to a user input signal received through the user interface device and provide the received/processed AR/VR service-related content data to the user. In this situation, the processor 2001 may receive/process the AR/VR service-related content data based on data acquired from the object detection device, the sensing unit, the position data generation device, the self-driving device, and so on. According to embodiments of the present disclosure, the AR/VR service-related content data may include entertainment content, weather information, and so on which are not related to the self-driving service as well as information related to the self-driving service such as driving information, path information for the self-driving service, driving maneuver information, vehicle state information, and object information.
According to embodiments of the present disclosure, a vehicle or a user interface device may receive a user input signal (S2110). According to embodiments of the present disclosure, the user input signal may include a signal indicating a self-driving service. According to embodiments of the present disclosure, the self-driving service may include a full self-driving service and a general self-driving service. The full self-driving service refers to perfect self-driving of a vehicle to a destination without a user's manual driving, whereas the general self-driving service refers to driving a vehicle to a destination through a user's manual driving and self-driving in combination.
It may be determined whether the user input signal according to embodiments of the present disclosure corresponds to the full self-driving service (S2120). When it is determined that the user input signal corresponds to the full self-driving service, the vehicle according to embodiments of the present disclosure may provide the full self-driving service (S2130). Because the full self-driving service does not need the user's manipulation, the vehicle according to embodiments of the present disclosure may provide VR service-related content to the user through a window of the vehicle, a side mirror of the vehicle, an HMD, or a smartphone (S2130). The VR service-related content according to embodiments of the present disclosure may be content related to full self-driving (e.g., navigation information, driving information, and external object information), and may also be content which is not related to full self-driving according to user selection (e.g., weather information, a distance image, a nature image, and a voice call image).
If it is determined that the user input signal does not correspond to the full self-driving service, the vehicle according to embodiments of the present disclosure may provide the general self-driving service (S2140). Because the FoV of the user should be secured for the user's manual driving in the general self-driving service, the vehicle according to embodiments of the present disclosure may provide AR service-related content to the user through a window of the vehicle, a side mirror of the vehicle, an HMD, or a smartphone (S2140).
The AR service-related content according to embodiments of the present disclosure may be content related to full self-driving (e.g., navigation information, driving information, and external object information), and may also be content which is not related to self-driving according to user selection (e.g., weather information, a distance image, a nature image, and a voice call image).
While the present disclosure is applicable to all the fields of 5G communication, robot, self-driving, and AI as described before, the following description will be given mainly of the present disclosure applicable to an XR device with reference to following figures.
The HMD-type XR device 100a shown in
Referring to
Although the frame may be formed in a shape of glasses worn on the face of the user 10 as shown in
The frame may include a front frame 110 and first and second side frames.
The front frame 110 may include at least one opening, and may extend in a first horizontal direction (i.e., an X-axis direction). The first and second side frames may extend in the second horizontal direction (i.e., a Y-axis direction) perpendicular to the front frame 110, and may extend in parallel to each other.
The control unit 200 may generate an image to be viewed by the user 10 or may generate the resultant image formed by successive images. The control unit 200 may include an image source configured to create and generate images, a plurality of lenses configured to diffuse and converge light generated from the image source, and the like. The images generated by the control unit 200 may be transferred to the optical display unit 300 through a guide lens P200 disposed between the control unit 200 and the optical display unit 300.
The controller 200 may be fixed to any one of the first and second side frames. For example, the control unit 200 may be fixed to the inside or outside of any one of the side frames, or may be embedded in and integrated with any one of the side frames.
The optical display unit 300 may be formed of a translucent material, so that the optical display unit 300 can display images created by the control unit 200 for recognition of the user 10 and can allow the user to view the external environment through the opening.
The optical display unit 300 may be inserted into and fixed to the opening contained in the front frame 110, or may be located at the rear surface (interposed between the opening and the user 10) of the opening so that the optical display unit 300 may be fixed to the front frame 110. For example, the optical display unit 300 may be located at the rear surface of the opening, and may be fixed to the front frame 110 as an example.
Referring to the XR device shown in
Accordingly, the user 10 may view the external environment through the opening of the frame 100, and at the same time may view the images created by the control unit 200.
In addition, hereinafter, the XR device shown in
Referring to
The wireless communication module 110 transmits or receives data to or from an external device 300. The external device 300 includes an external server and a database server.
The camera 121 captures an image of a subject in front of the XR device 100. The camera 121 captures an image of an object in front of the XR device 100. The camera 121 includes a TOF camera 121.
The TOF camera 121 means a general camera to which a TOF sensor is attached. In detail, an image sensor which captures a scene devises a 2D based result. In this situation, if the TOF sensor is used, depth may be measured and therefore a 3D result may be obtained. Time of Flight (TOF) means the time required to transmit sound waves or light sources to a subject and then return them to the original position through the subject. The TOF sensor is a kind of a component that may sense such roles.
A depth map means one image having information related to a distance from a viewpoint to an object surface in a 3D computer graphic.
The TOF camera 121 captures a first image that includes a depth map. If the TOF camera 121 captures images, the original image, a depth map, and a result image obtained by applying the depth map to the original image.
The microphone 122 receives a voice of a user. In detail, the microphone 122 processes an external sound signal to an electrical audio data. The microphone 122 is abbreviated from a microphone. The processed audio data may be used in various ways in accordance with a function or an application program, which is currently in service. Various noise removal algorithms may be implemented in the microphone 122 to remove noise generated when the external sound signal is input.
The microphone 122 is configured to receive a voice of a user, other sound, etc. The microphone 122 may be provided in a plurality of places and configured to receive stereo sounds.
The user input module 123 is to receive information from a user. If information is input through the user input module 123, the controller 180 may control the operation of the XR device 100 to correspond to the input information. The user input module 123 may include a mechanical input means (or mechanical key, for example, button, dome switch, jog wheel, jog switch, etc. located on a front or rear surface or a side of the XR device 100) and a touch input means. As an example, the touch input means may be provided as a virtual key, a soft key, or a visual key, which is displayed on a touch screen through software processing, or may be provided as a touch key arranged on a portion other than the touch screen. The virtual key or the visual key may be displayed on the touch screen in various shapes, and for example, may be configured in the form of graphic, text, icon, video or their combination.
The sensor module 140 senses peripheral environment information surrounding the XR device. In this situation, the peripheral environment information includes displacement of the XR device 100.
The sensor module 140 senses at least one of information inside the XR device, peripheral environment information surrounding the XR device and user information, and generates a sensing signal corresponding to the sensed information. The controller 180 may control driving or operation of the XR device 100 based on the sensing signal, or may perform data processing, function or operation related to an application program installed in the XR device 100. Main sensors of various sensors, which may be included in the sensor module 140, will be described in more detail.
First of all, the proximity sensor 141 means a sensor that detects the presence of an object approaching a predetermined detection surface or an object existing near the predetermined detection surface by using an electric field force or infrared rays without mechanical contact. The proximity sensor 141 may be arranged in an inner area of a mobile terminal surrounded by the touch screen or near the touch screen.
Examples of the proximity sensor 141 may include a transmissive photoelectric sensor, a direct reflective photoelectric sensor, a mirror reflective photoelectric sensor, a high-frequency oscillation-type proximity sensor, a capacitance proximity sensor, a magnetic type proximity sensor, and an infrared proximity sensor. If the touch screen is a capacitance type, the proximity sensor 141 may be configured to detect proximity of an object having conductivity due to a change of an electric field according to proximity of the object. In this situation, the touch screen (or touch sensor) may be categorized as a proximity sensor.
For convenience of description, the term “proximity touch” will be referred to herein to denote the behavior in which an object is positioned to be proximate to the touch screen without contacting the touch screen. The term “contact touch” will be referred to herein to denote the behavior in which an object makes physical contact with the touch screen. For the position corresponding to the proximity touch of the object relative to the touch screen, such position means a position where the object is perpendicular to the touch screen. The proximity sensor 141 may sense a proximity touch, and proximity touch patterns (for example, proximity touch distance, proximity touch direction, proximity touch speed, proximity touch time, proximity touch position, proximity touch moving status, and the like). The controller 180 may process data (or information) corresponding to proximity touches and proximity touch patterns sensed by the proximity sensor 141 and output visual information corresponding to the processed data on the touch screen. In addition, the controller 180 may control the XR device 100 to execute different operations or process different data (or information) depending on whether a touch with respect to a point on the touch screen is either a proximity touch or a contact touch.
A touch sensor senses a touch (or touch input) applied to the touch screen (or display 151), using at least one of a variety of touch methods such as a resistive type, a capacitance type, an infrared type, an ultrasonic type, and a magnetic field type.
As one example, the touch sensor may be configured to convert changes of pressure applied to a specific part of the touch screen or capacitance occurring at a specific part into electric input signals. The touch sensor may also be configured to sense not only a touched position and a touched area of a touch target but also touch pressure and/or touch capacitance. The touch target is an object generally used to apply a touch input to the touch sensor. Examples of the touch target include a finger, a touch pen, a stylus pen, and a pointer.
When a touch input is sensed by the touch sensor, corresponding signal(s) may be transmitted to a touch controller. The touch controller may process the received signals and then transmit corresponding data to the controller 180. Therefore, the controller 180 may sense which region of the display 151 has been touched. In this situation, the touch controller may be a component separate from the controller 180, or may be the controller 180.
The display 151 includes a transparent portion, and displays the captured image in accordance with a control command from the controller 180. The display 151 may project the image into a user's eyes by using prisms. Also, the prisms may be formed in a light-transmissive type such that the user may together view the projected image and a normal view (range that the user views through his/her eyes) at the front.
In this way, the image output through the display 151 may be viewed to be overlapped with the normal view. The XR device 100 may provide augmented reality (AR) that displays one image by overlapping a real image or background with a virtual image using the display property.
The controller 180 enters a camera mode if a selection input for executing the camera mode is received from the user.
The controller 180 controls the display 151 to display a menu that includes a first button for displaying at least one virtual object, a second button for executing a capture, and a third button for storing at least one virtual object. The controller 180 determines whether to include a portion of at least one virtual object in a screen when capturing an image or video based on the peripheral environment information and the object and controls the camera to capture the image or video in accordance with the determined result. The controller 180 may control the camera to capture a video or still image.
Peripheral environment information surrounding the XR device is sensed (S2510).
If a selection input for executing a camera mode is received from a user S(2520), the controller enters the camera mode (S2530).
The controller controls the display 151 to display a menu that includes a first button for displaying at least one virtual object, a second button for executing a capture and a third button for storing at least one virtual object (S2540).
The controller determines whether to include a portion of at least one virtual object in a screen when capturing an image based on the peripheral environment information and object (S2550).
The controller controls the camera to capture an image in accordance with the determined result (S2560).
If the determined result includes all of at least one virtual object, the controller may capture the image by including all of at least one virtual object in the screen. If the determined result includes some of at least one virtual object, the controller may capture the image by including some of at least one virtual object in the screen.
If the determined result excludes all of at least one virtual object, the controller may capture the image by excluding all of at least one virtual object from the screen.
The controller 180 enters the camera mode (S2610).
The controller 180 collects status data which will be used as an input for a reinforcement learning model (S2615). In this situation, the reinforcement learning model means an entity that may determine a behavior taken by a user. For example, the reinforcement learning model determines whether to include a virtual object in a preview screen.
The controller 180 inputs status data to the reinforcement learning model (2620).
The controller 180 acquire a user behavior value (S2625).
The controller 180 determines whether to a virtual object in the screen based on the status data and the acquired user behavior value when capturing an image (S26630).
The controller 180 identifies whether to include the virtual object in the screen (S2635).
If the virtual object is included in the screen (S2635), the controller 180 configures a preview screen by including the virtual object in the screen (S2640).
If the virtual object is not included in the screen (S2635), the controller 180 configures a preview screen without including the virtual object in the screen (S2645).
The controller 180 displays a menu for manual manipulation on the preview screen (S2650).
The controller 180 identifies whether manual manipulation occurs (S2655).
If manual manipulation does not occur (S2655), the controller 180 provides the reinforcement learning model with an addition point (S2660). Since the determination of the reinforcement learning model is right, the reinforcement learning model is re-learned based on the determined result.
If manual manipulation occurs (S2655), the controller 180 provides the reinforcement learning model with a subtraction point (S2665). Since the determination of the reinforcement learning model is false, the reinforcement learning model is re-learned based on the determined result.
The controller 180 controls the camera to capture an image (S2670).
Referring to
The frame portion is supported in the head portion of the user, and provides a space where various components are arranged. As shown, electronic components such as a controller 480 and a sound output module 452 may be arranged in the frame portion. Also, a lens 403 for covering at least one of a left eye and a right eye may detachably be provided in the frame portion. The sound output module 452 includes a speaker.
The controller 480 is configured to control various electronic components provided in the XR device 400. The controller 480 may be understood as a component corresponding to the aforementioned controller 180. In this drawing, the controller 480 is provided in the frame portion on the head portion at one side. However, the position of the controller 480 is not limited to this example.
The display 451 may be implemented in the form of a head mounted display (HMD). The HMD means a display that displays an image in front of a user. When the user wears the XR device 400 of the AR glasses type, the display 451 may be arranged to correspond to at least one of a left eye and a right eye. In this drawing, the display 451 is located at a part corresponding to the right eye such that the image may be output toward the user's right eye.
The display 451 may project to the user's eyes by using prisms. Also, the prisms may be formed in a light-transmissive type such that the user may view the projected image and a normal view (range that the user views through his/her eyes) at the front.
In this way, the image output through the display 451 may be viewed to be overlapped with the normal view. The XR device 400 may provide augmented reality (AR) that displays one image by overlapping a real image or background with a virtual image using the display property.
The camera 421 is arranged to be adjacent to at least one of the left eye and the right eye and formed to capture a front image. Since the camera 421 is arranged to be adjacent to eyes, the camera 421 may acquire a scene viewed by the user as an image.
In this drawing, although the camera 421 is provided in the controller 480, the camera 421 is not limited to the example of the drawing. The camera 421 may be provided in the frame portion, and may be provided in a plural number to acquire a stereoscopic image.
The XR device 400 of the glasses type may include user input modules 423a and 423b manipulated to receive a control command. All tactile manners that the user may manipulate together with a tactile action such as touch and push may be used for the user input modules 423a and 423b. In this drawing, the user input modules 423a and 423b of push and touch input manners are respectively provided in the frame portion and the controller 480 as an example.
Also, the XR device 400 of the glasses type may be provided with a microphone that processes an input sound as electrical audio data and a sound output module 452 that outputs the sound. The sound output module 452 may be configured to transfer the sound in a normal sound output manner or osteophony manner. If the sound output module 452 is implemented in an osteophony manner, the sound output module 452 is closely attached to the head portion and transfers the sound by vibrating skull when the user wears the XR device 400.
Referring to an upper end of
As shown in the upper end of
Referring to an upper end of
As shown in the upper end of
According to one embodiment of the present disclosure, as shown in the lower end of
Referring to
For example, the preview screen 2910 includes a first object 11, a second object 12, a third object 13, and a fourth object 14.
The controller 180 senses the first object 11, the second object 12, the third object 13 and the fourth object 14 by analyzing the preview screen 290 and determines that all of the objects 11, 12, 13 and 14 smoke.
According to the present disclosure, two-dimensional image based object sensing technology may be used. For example, an object existing in an actual screen viewed by AR glasses may be identified by a deep learning object search algorithm, such as CNN, YOLO v3 (You only look once), and SSD, in real time, and a user's current status may be identified based on a corresponding object.
In the situation of YOLO v3, 280 objects may basically be analyzed, and learning may be performed by adding an object to the 280 objects.
Referring to
A horizontal axis means an inference time required for inference. In the situation of Yolo v3-320 model, 50 frames may be detected at 22 msec, 1 second.
A vertical axis is COCO mAP50, and means a performance index of object sensing. In detail, the vertical axis means a ratio detected normally among all detected results.
The reinforcement learning (RL) model 300 is one of issues handled by mechanical learning and means a method for selecting a behavior for maximizing a reward 304 of an action 305, which may be selected, by recognizing a current state 303 by means of an agent 301 defined in a random environment 302.
The reinforcement learning model 300 includes the agent 301, the environment 302, the state 303, the reward 304, and the action 305.
The agent 301 observes the state 303, selects the action 305, and aims at a target. The agent 301 includes a learner and a decision maker.
The environment 302 means the other except the agent 301.
The state 303 means information indicating a current status of a system user.
The action 305 means what the agent 301 does at a current status.
The reward 304 means information indicating whether the action 305 is good or bad.
Referring to
For example, the person image 31 has exactness of 99.3%, and the rock image 35 has exactness of 82.8%.
Referring to
Referring to
For example, the face of the person 14 is analyzed, whereby the gender 21 is determined as a female, the eye's state is determined as eye's opening state, the feeling state 23 is determined as happy, and the lip shape 24 is determined as smile.
In detail, the controller 180 may identify happiness, age, whether the person opens his/her eyes, glasses, facial hair, etc. by analyzing attributes of a face displayed in an image or video. Also, the controller 180 may measure how the attributes are changed on the video as the time passes. For example, the controller 180 may configure a time zone when emotion of an actor is changed.
Referring to
For example, if a user drafts a camera execution text on a paper, the camera captures a camera execution text image and captures an image based on the camera execution text.
For example, if the user drafts a camera recording execution text on a paper by excluding a virtual screen, the camera captures a text image and executes capturing by excluding a virtual object from the screen based on the camera recording execution text without excluding the virtual screen.
The embodiment in which the camera mode is executed through a user's voice will be described with reference to
For example, if the user utters a camera start 20, the controller 180 executes the camera mode since the camera start includes a camera start which is a preset first keyword.
The embodiment in which the camera mode is executed through a user's gesture will be described with reference to
For example, if the user takes a gesture for moving his/her head up and down in a state that the user wears the XR device 100, the controller 180 executes the camera mode. If the user takes a gesture for moving his/her from left to right or vice versa, the controller 180 does not execute the camera mode.
The embodiment in which the camera mode is executed through a finger gesture will be described.
For example, if the user 10 takes a gesture for moving a forefinger up and down in front of the camera in a state that the user wear the XR device 100, the controller 180 executes the camera mode. If the user 10 takes a gesture for moving a forefinger from left to right in front of the camera, the controller 180 does not execute the camera mode.
Referring to
Referring to
The right drawing of
According to the present disclosure, the preview screen 3430 of the XR device 100 is the same as the preview screen 3420 of the mobile device 200, and capturing, camera setup and focus of the XR device 100 may be adjusted through a screen touch in the mobile device 200.
Referring to
Referring to
Referring to
The menu screen that includes the virtual ON icon 31, the capturing icon 32, and the full-storage icon 33 is arranged at a lower end of the preview screen 3510. This is to allow the user's finger not to cover the preview screen when the user clicks the icon.
Referring to
The preview screen 3510 includes a virtual OFF icon 41, a capturing icon 32, and a full-storage icon 33.
Referring to
Referring to
For example, the controller 180 analyzes an object image 11 of a preview screen 3610, and the analyzed object image 11 is an Eiffel tower image in Paris. The current position is a tourist attraction, and the status determination is the status that the user is sightseeing near the Eiffel tower in Paris. The user's behavior is sightseeing, and the virtual object 20 is an information popup image 20.
The controller 180 determines whether to include the virtual object 20 in the screen, when capturing an image, based on the object image 11, the user's behavior, current position and status determination. If the virtual object 20 is provided in a plural number, the controller 180 determines whether to include the virtual object in the screen while capturing the image individually.
Additionally, the controller 180 may determine whether to the virtual object in the screen when capturing the image, by using a reinforcement learning model.
In this situation, the inputs of the reinforcement learning model include the object image 11, the user's behavior, current position and status determination. The controller 180 determines whether to include the virtual object 20 in the screen when capturing the image based on the inputs of the reinforcement learning model.
For example, the object image 11 is an Eiffel tower image in Paris. The user's behavior is sightseeing in Paris. The current position is Paris, and the status determination is the status that the user is currently sightseeing near an Eiffel tower in Paris. The virtual object 20 is an information popup image 20.
The controller 180 provides an addition point if the determination based on the reinforcement learning model is right and provides a subtraction point if the determination based on the reinforcement learning model is false. That is, it is possible to enhance exactness in prediction by reflecting the user's feedback in the reinforcement learning model.
For example, if the user uses a photo configuration as it is without correction, determination based on the reinforcement learning model is right. If the user uses a photo configuration by correction, determination based on the reinforcement learning model is false.
According to the present disclosure, since it is possible to determine whether to include the virtual object in the screen when capturing the virtual object, exactness in prediction may be enhanced in a photo configuration based on the reinforcement learning model obtained by reflecting the user's feedback.
Referring to
A preview screen 3710 includes a virtual object 20 such as AR writing.
Referring to
Referring to
Referring to
Referring to
For example, the controller 180 identifies attributes of the virtual object as note writing based on the virtual object that includes contents such as important expected questions 21, an asterisk 22, and memorizing things 23.
The controller 180 identifies state information of the user who wears the XR device 100. In detail, the controller identifies the state information of the user based on a sensed object displayed on the current screen, GPS position information and peripheral sound information received by the microphone.
The controller 180 determines whether to include the virtual object in the screen, when capturing a virtual object, based on the state information of the user and the attributes of the virtual object.
For example, the objects displayed on the current screen are a blackboard, person, and book. The GPS position information is an academy. The peripheral sound information received by the microphone is “take notes”. Therefore, it is noted that the state information of the user is “in class at academy”. The attribute of the virtual object is note writing.
Therefore, the controller determines that the virtual object including attributes of note writing is included in capturing because the user is in class at academy.
Referring to
The virtual object 20 means a virtual object where a user takes notes, and the virtual OFF icon 31 is an icon for executing a function of disappearing the virtual object from the screen. The individual selection icon 32 is an icon that allows an icon capable of selecting some of a plurality of virtual objects to be generated near the virtual object. The capturing icon 33 is an icon that executes capturing. Any one of video capturing and still image capturing may be executed. The two-type storage icon 34 stores both the situation that the virtual object is included in the screen and the situation that the virtual object is excluded from the screen.
The embodiment in which a feedback is provided to a reinforcement learning model and a virtual object which will be included in capturing is determined using the reinforcement learning model that reflects the feedback will be described with reference to
The controller 180 receives the feedback from the user and determines at least one virtual object, which will be included in the screen, by using the reinforcement learning model that reflects the received feedback.
The controller 180 provides a reward point to the reinforcement learning model if the determined virtual object 20 is right, and provides a subtraction point to the reinforcement learning model if the determined virtual object 20 is false.
For example, if the virtual object 20 of a note writing type is maintained and a prediction result for capturing occurs, the user includes the virtual object 20 in capturing. However, since a wrong prediction may occur, the controller 180 may display the icon 31, which may exclude the virtual object 20.
If the user captures an image including the virtual object 20 as a setup value predicted by the reinforcement learning model without additional manual manipulation, since the determined result is right, the controller 180 provides an addition point to the reinforcement learning model. That is, the situation that the user does not perform manual manipulation means a positive feedback.
The controller 180 provides an addition point to the reinforcement learning model if the determined virtual object 20 is right.
According to the present disclosure, since the determined result is right, the controller 180 provides an addition point of +10 points to the reinforcement learning model. The controller 180 performs additional learning indicating that determination of the reinforcement learning model is right. Therefore, as the corresponding process is repeated, determination of the reinforcement learning model may be more exact.
Referring to
The virtual object 20 means a virtual object where a user 10 takes notes, and the virtual OFF icon 31 is an icon for executing a function of disappearing the virtual object from the screen. The individual selection icon 32 is an icon that allows an icon capable of selecting some of a plurality of virtual objects to be generated near the virtual object. The capturing icon 33 is an icon that executes capturing. Any one of video capturing and still image capturing may be executed. The two-type storage icon 34 stores both the situation that the virtual object is included in the screen and the situation that the virtual object is excluded from the screen.
As shown in
As shown in
The controller 180 provides a subtraction point to the reinforcement learning model because determination of the virtual object 20 to be included in the screen is false.
According to the present disclosure, since the determined result is false, the controller 180 provides a subtraction point of −10 points to the reinforcement learning model. The controller 180 performs additional learning indicating that determination of the reinforcement learning model is false. Therefore, as the corresponding process is repeated, determination of the reinforcement learning model may be more exact. In this situation, −10 points are only exemplary, and the scope of the present disclosure is not limited to the points.
Referring to
The virtual object 20 means a virtual object where a user takes notes, and the virtual OFF icon 31 is an icon for executing a function of disappearing the virtual object from the screen. The individual selection icon 32 is an icon that allows an icon capable of selecting some of a plurality of virtual objects to be generated near the virtual object. The capturing icon 33 is an icon that executes capturing. Any one of video capturing and still image capturing may be executed. The two-type storage icon 34 stores both the situation that the virtual object is included in the screen and the situation that the virtual object is excluded from the screen.
The controller 180 controls the display 151 to display icons 41, 42 and 43 capable of selecting whether to include at least one virtual object in capturing, near at least one virtual object.
For example, if an input for selecting the individual selection icon 32 is received from the user 10, as shown in
For example, the first icon 41 excluded from capturing is displayed near the first virtual icon 21. The second icon 42 excluded from capturing is displayed near the second virtual icon 22. The fourth icon 43 excluded from capturing is displayed near the third virtual icon 23.
The first icon 41 excluded from capturing means that the camera captures an image by excluding the first virtual object 21.
If the user selects the first icon 41 excluded from capturing, the first icon 41 excluded from capturing is changed to the first icon 51 included in capturing.
If the user selects the first icon 51 included in capturing, the first icon 51 included in capturing is changed to the first icon 41 excluded from capturing.
Referring to
The virtual object 20 means a virtual object where a user takes notes, and the virtual OFF icon 31 is an icon for executing a function of disappearing the virtual object from the screen. The individual selection icon 32 is an icon that allows an icon capable of selecting some of a plurality of virtual objects to be generated near the virtual object. The capturing icon 33 is an icon that executes capturing. Any one of video capturing and still image capturing may be executed. The two-type storage icon 34 stores both the situation that the virtual object is included in the screen and the situation that the virtual object is excluded from the screen.
For example, if an input for selecting the individual selection icon 32 is received from the user 10, as shown in
Referring to an upper side of
Referring to
For example, if the user 10 who wears the XR device 100 looks at a front, the virtual object is displayed. The virtual object includes two types, that is, a note memo type 4310 and a portal screen type 4320.
In this situation, the first virtual object 4310 of the note memo type and the second virtual object 4320 of the portal screen type have their respective attributes different from each other. The first virtual object 4310 is a virtual object where the user takes notes, and the second virtual object 4320 is a virtual object generated by executing a portal application through a user.
If the first virtual object 4310 and the second virtual object 4320 having an attribute different from that of the first virtual object 4310 are simultaneously displayed on the screen, the controller 180 independently determines whether to include the first virtual object 4310 and the second virtual object 4320 in the screen.
For example, the controller 180 separately determines whether to include the first virtual object 4310 and the second virtual object 4320 in the screen, based on peripheral environment information and object.
In the situation of the first virtual object 4310, considering that an attribute of the virtual object is note writing and the current status is in class, since note writing corresponds to a class content, the controller 180 determines that the first virtual object is included in the screen when capturing the image.
In detail, since correlation between the peripheral environment information and the first virtual object is 1, the controller 180 determines that the first virtual object is included in the screen when capturing the image.
In this situation, correlation ranges from 0 to 1, wherein 1 means that the virtual object corresponds to peripheral environment information. 0 means that the virtual object is not related with the peripheral environment information.
In the situation of the second virtual object 4320, considering that an attribute of the second virtual object is portal screen and the current status is in class, since portal screen is less related with a class content, the controller 180 determines that the second virtual object 4320 is excluded from the screen when capturing the image.
Referring to
The preview screen includes an individual selection icon 31 for individually selecting a virtual object, a capturing icon 32 for executing capturing, and a two-type storage icon 33 for storing both the situation that the virtual object is included and the situation that the virtual object is excluded.
The embodiment in which an icon indicating that the second virtual object is not included in capturing is displayed if there are virtual objects having different attributes will be described with reference to
The controller 180 controls the display 151 to display an icon excluded from capturing, which indicates that the second virtual object 4420 is not included in capturing, on a portion of the second virtual object 4420 while the first virtual object 4410 is being captured.
The icon 40 excluded from capturing is an image obtained by applying an alphabet X image to a camera shape image. This means that the second virtual object 4420 is not included in capturing. If an input for selecting the icon 40 excluded from capturing is received from the user, the controller 180 controls the display 151 to display the icon 41 included in capturing. The icon 41 included in capturing is an image of a camera shape. This means that the second virtual object 4420 is included in capturing.
If an input for selecting the icon 41 included in capturing is received from the user, the controller 180 controls the display 151 to again display the icon 40 excluded from capturing.
Referring to
Referring to
If an input for selecting a photo capturing icon 32 of the application execution screen 30 is received from the user 10, the controller 180 executes a camera mode.
Referring to
The controller 180 determines that the virtual object 20, which includes the amount of calories, is excluded from the screen, by considering that the SNS application is being executed.
The user's action may include three types. The first type action is to capture an image by including a virtual object in the screen. The second type action is to capture an image by excluding the virtual object from the screen. The third type action is to capture the situation that the virtual object is included in the screen and the situation that the virtual object is excluded from the screen.
That is, since the SNS application is executed, a virtual object, which includes a private content, does not need to be viewed by other users if the virtual object is captured by the camera. Therefore, the controller 180 determines the second type action for excluding the virtual object 20 from the screen.
The controller 180 removes the virtual object 20, which includes the amount of calory, from the screen and activates a camera capturing screen 4610.
The controller 180 displays a virtual ON icon 31, a capturing icon 32, and a two-type storage icon 33 for storing two types of images below the screen to allow the virtual object to be manually displayed below the screen.
Whether to include the virtual object 20 in the screen will be described in detail. The controller 180 determines conditions of SNS application execution state, which are added to a condition input.
For example, the conditions include i) sensing an object from a two-dimensional screen, ii) acquiring a GPS position, iii) sensing a current status through a microphone, iv) checking a type of a virtual object displayed on a current screen, and v) entering a camera mode through SNS application.
According to one embodiment of the present disclosure, when a user views an object in front of him/her in a state that the user wears AR glasses, whether to include a virtual object in a screen may be determined based on peripheral environment information and objects, whereby the user does not manipulate a function for including a virtual object manually and therefore user convenience may be improved.
According to another embodiment of the present disclosure, a state near a user may be identified using a reinforcement learning model and whether to include a virtual object in a screen may be determined by predicting a behavior which will be performed by the user, whereby a feedback of the user may be reflected more exactly and therefore user convenience may be improved.
According to still another embodiment of the present disclosure, when a plurality of virtual objects having different attributes are simultaneously displayed on a screen, whether to include the plurality of virtual objects in the screen may independently be determined, whereby intention of the user may be reflected more exactly and therefore user convenience may be improved.
The XR device and the method for controlling the same according to the present disclosure are not limited to the aforementioned embodiments, and all or some of the aforementioned embodiments may selectively be configured in combination so that various modifications may be made in the aforementioned embodiments.
Although the present specification has been described with reference to the accompanying drawing, it will be apparent to those skilled in the art that the present specification can be embodied in other specific forms without departing from the spirit and characteristics of the specification. The scope of the specification should be determined by reasonable interpretation of the appended claims and all change which comes within the equivalent scope of the specification are included in the scope of the specification.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10-2019-0143200 | Nov 2019 | KR | national |
