Method and process for predictive beam refinement

Abstract

A wireless transmit / receive unit (WTRU) may be configured to receive a configuration for predictive beam refinement. The configuration may include at least one threshold. The at least one threshold may be related to the information. The information may be used to predict and / or select one or more beam pairs. The WTRU may perform measurements on one or more reference signal (RS) sets. The WTRU may determine to store one or more of the measurements in memory, for example as training data samples. The WTRU may determine to store the one or more measurements based on a comparison of the one or more measurements to the at least one threshold. The WTRU may determine to start training or retraining a model for performing predictions related to the one or more beam pairs. The WTRU may send an indication that the stored training data sample has reached the at least one threshold.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of U.S. Provisional Patent Application No. 63 / 395,884, filed on August 8, 2022, the entire contents of which are incorporated herein by reference. Background Art

[0003] Beam management (e.g., in NR) may involve the selection, adjustment, and / or tracking of transmit (Tx) beams and receive (Rx) beams, and may be categorized as a P-1 process, a P-2 process, and a P-3 process. Beam management (e.g., in NR) may identify a suitable (e.g., optimal) beam based on a beam search principle transmitted by a gNB with a downlink signal (e.g., synchronization signal block (SSB) / channel state information-reference signal (CSI-RS), etc.) and / or a WTRU measurement and reporting process. The beam scanning and measurement operations may consume a significant amount of resources from both the Tx side and the Rx side, and / or may also result in latency (e.g., for a large number of beams and / or at high frequencies).

[0004] Artificial intelligence and / or machine learning (AI / ML) may be implemented with respect to an air interface (e.g., an NR air interface) and / or beam management. AI / ML may be used as a framework for improving performance and complexity, for example, with respect to beam management (e.g., including beam prediction in the time domain and / or spatial domain for overhead and latency reduction, beam selection accuracy improvement, etc.). If beam management and / or beam prediction are based on an AI / ML framework, certain challenges associated with beam measurement and reporting (e.g., and training and validation of AI / ML models in scenarios with hierarchical spatial relationships and beam associations in different frequency ranges) may be addressed. For example, data-driven WTRU inferences and behaviors may be used to determine / perform associations, measurements, and reporting associated with beam resources (e.g., and training, validation, activation, and / or deactivation of AI / ML models), which may be used to implement predictive beam establishment / refinement. Summary of the invention

[0005] Data-driven (e.g., via an artificial intelligence / machine learning (AI / ML) model) beam prediction and / or establishment may be performed. For example, configuration information associated with beam scanning may be received. For example, the configuration information may include a buffer size associated with the AI / ML model and / or a threshold quality metric. For example, beam scanning may be performed across one or more beams based on the configuration information. The AI / ML model may be trained based on the beam scanning performed across the one or more beams. For example, the AI / ML model may be implemented at a wireless transmit receive unit (WTRU) and / or a network node (e.g., a gNB). Training the AI / ML model may also include: determining a quality metric associated with each of the one or more beams for which beam scanning is performed; and filling a buffer associated with the AI / ML model based on whether the determined quality metric associated with each of the one or more beams is greater than a threshold (e.g., received via the configuration information). Using the AI / ML model, a beam of the one or more beams may be identified, and an indication of the identified beam may be transmitted. After the beam has been identified, the quality metric associated with the identified beam may be periodically measured. If the quality metric associated with the identified beam is below a second threshold, the AI / ML model may be retrained.

[0006] The WTRU may be configured to receive a configuration for predictive beam refinement. The configuration may include at least one threshold. The at least one threshold may be associated with information. The information may be used to predict and / or select one or more beam pairs. The WTRU may perform measurements on one or more reference signal (RS) sets. The WTRU may determine to store one or more of the measurements in a memory, for example, as training data samples. The WTRU may determine to store the one or more measurements based on a comparison of the one or more measurements with at least one threshold. The WTRU may determine, for example, based on at least one threshold, to start training or retraining a model for performing predictions associated with one or more beam pairs. The WTRU may send an indication. The indication may include that the stored training data samples have reached at least one threshold.

[0007] The WTRU may predict measurements, for example, for one or more beam pairs. For example, the WTRU may predict measurements based on stored training data samples and / or a predictive beam refinement configuration. The WTRU may report one or more of a predictive beam identifier, a codebook identifier, a predictive layer 1 received signal received power (L1-RSRP) value, or a predicted angle of arrival (AoA), for example, based on the predicted measurements. The at least one threshold may include at least one of: a buffer size threshold associated with a memory for indicating a maximum number of samples of beam quality measurements or beam combining weight pairs, a threshold associated with a maximum buffer size or a remaining buffer size, a threshold associated with one or more measurements to be performed on one or more RS sets, a threshold number of comparisons of one or more measurements to one or more thresholds, a completed AI / ML model, and / or one or more thresholds for determining to store samples in a memory. Based on the received configuration, the WTRU may retrieve the stored training data samples from the memory.

[0008] A determination may be made at the WTRU and / or a network entity to begin predicting measurements for one or more beam pairs. The at least one threshold may be associated with information to be used for predicting and / or selecting one or more beam pairs.

[0009] For example, when one or more of the measurements are greater than or equal to at least one threshold, the WTRU may store the one or more measurements in the memory as training data samples. For example, when one or more of the measurements are less than at least one threshold, the WTRU may repeat the determination to store the one or more measurements in the memory as training data samples.

[0010] The WTRU may calculate a prediction accuracy of measurements for one or more beam pairs, for example based on stored training data samples. The at least one threshold may include a value that increases with the stored training data samples. Predicting measurements for one or more beam pairs may be based on a weight vector.

[0011] Embodiments described herein may implement artificial intelligence (AI) and / or machine learning (ML) algorithms (e.g., models). For example, the WTRU and / or the network may use one or more AI / ML models to predict measurements and refine predictions for beam pairs. As used herein, the terms "artificial intelligence" and / or "AI" may include behaviors exhibited by one or more machines that simulate one or more cognitive functions (e.g., to perceive, reason, adapt, and / or act). BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1A is a system diagram illustrating an example communication system in which one or more disclosed embodiments may be implemented.

[0013] Figure 1B is an example of a method that can be performed according to an embodiment of the present invention. Figure 1A A system diagram of an example wireless transmit / receive unit (WTRU) for use within an illustrated communication system.

[0014] Figure 1C is an example of a method that can be performed according to an embodiment of the present invention. Figure 1A System diagram of an example Radio Access Network (RAN) and an example Core Network (CN) for use within the illustrated communication system.

[0015] Figure 1D is an example of a method that can be performed according to an embodiment of the present invention. Figure 1A System diagram of another example RAN and another example CN for use within the illustrated communication system.

[0016] Figure 2A is a schematic diagram of an example system environment in which artificial intelligence (AI) and / or machine learning (ML) models may be implemented.

[0017] Figure 2B An example of a neural network is illustrated.

[0018] Figure 2C is a schematic diagram of an example system environment for training and / or implementing an AI / ML model including a neural network (NN).

[0019] Figure 2D is a schematic diagram of an example system environment for training and / or implementing an AI / ML model that includes an autoencoder.

[0020] Figure 3 is a schematic diagram of an example of training an AI / ML model.

[0021] Figure 4 is a schematic diagram of an example of inferred beam prediction performance.

[0022] Figure 5 is a flow chart illustrating an example process for predictive beam refinement. DETAILED DESCRIPTION

[0023] Figure 1A1 is a diagram illustrating an example communication system 100 in which one or more disclosed embodiments may be implemented. The communication system 100 may be a multiple access system that provides content such as voice, data, video, messaging, broadcast, etc. to multiple wireless users. The communication system 100 may enable multiple wireless users to access such content through sharing of system resources (including wireless bandwidth). For example, the communication system 100 may employ one or more channel access methods such as code division multiple access (CDMA), time division multiple access (TDMA), frequency division multiple access (FDMA), orthogonal FDMA (OFDMA), single carrier FDMA (SC-FDMA), zero tail unique word DFT spread OFDM (ZT UW DTS-s OFDM), unique word OFDM (UW-OFDM), resource block filtered OFDM, and filter bank multi-carrier (FBMC), etc.

[0024] like Figure 1A As shown, the communication system 100 may include wireless transmit / receive units (WTRUs) 102a, 102b, 102c, 102d, RAN 104 / 113, CN 106 / 115, public switched telephone network (PSTN) 108, Internet 110 and other networks 112, but it should be understood that the disclosed embodiments contemplate any number of WTRUs, base stations, networks, and / or network elements. Each of the WTRUs 102a, 102b, 102c, 102d may be any type of device configured to operate and / or communicate in a wireless environment. By way of example, the WTRUs 102a, 102b, 102c, 102d (any of which may be referred to as a “station” and / or “STA”) may be configured to send and / or receive wireless signals and may include a user equipment (WTRU), a mobile station, a fixed or mobile subscriber unit, a subscription-based unit, a pager, a cellular phone, a personal digital assistant (PDA), a smart phone, a laptop, a netbook, a personal computer, a wireless sensor, a hotspot or Mi-Fi device, an Internet of Things (IoT) device, a watch or other wearable device, a head-mounted display (HMD), a vehicle, a drone, medical devices and applications (e.g., remote surgery), industrial devices and applications (e.g., robots and / or other wireless devices operating in an industrial and / or automated process chain environment), a consumer electronic device, a device operating on a commercial and / or industrial wireless network, etc. Any of the WTRUs 102a, 102b, 102c, and 102d may be interchangeably referred to as a WTRU.

[0025] The communication system 100 may also include a base station 114a and / or a base station 114b. Each of the base stations 114a, 114b may be any type of device that is configured to wirelessly interface with at least one of the WTRUs 102a, 102b, 102c, 102d to facilitate access to one or more communication networks, such as the CN 106 / 115, the Internet 110, and / or other networks 112. By way of example, the base stations 114a, 114b may be a base transceiver station (BTS), a Node B, an eNode B, a Home Node B, a Home eNode B, a gNB, an NR Node B, a site controller, an access point (AP), a wireless router, and the like. Although the base stations 114a, 114b are each depicted as a single element, it should be appreciated that the base stations 114a, 114b may include any number of interconnected base stations and / or network elements.

[0026] The base station 114a may be part of the RAN 104 / 113, which may also include other base stations and / or network elements (not shown), such as a base station controller (BSC), a radio network controller (RNC), a relay node, etc. The base station 114a and / or the base station 114b may be configured to send and / or receive wireless signals on one or more carrier frequencies (which may be referred to as cells (not shown)). These frequencies may be in a licensed spectrum, an unlicensed spectrum, or a combination of licensed and unlicensed spectrums. A cell may provide coverage of wireless services to a specific geographic area, which may be relatively fixed or may change over time. The cell may be further divided into cell sectors. For example, a cell associated with the base station 114a may be divided into three sectors. Therefore, in one embodiment, the base station 114a may include three transceivers, i.e., one transceiver for each sector of the cell. In one embodiment, the base station 114a may employ multiple-input multiple-output (MIMO) technology and may utilize multiple transceivers for each sector of the cell. For example, beamforming may be used to send and / or receive signals in a desired spatial direction.

[0027] The base stations 114a, 114b may communicate with one or more of the WTRUs 102a, 102b, 102c, 102d over an air interface 116, which may be any suitable wireless communication link (e.g., radio frequency (RF), microwave, centimeter wave, micrometer wave, infrared (IR), ultraviolet (UV), visible light, etc.). The air interface 116 may be established using any suitable radio access technology (RAT).

[0028] More specifically, as noted above, the communication system 100 may be a multiple access system and may employ one or more channel access schemes such as CDMA, TDMA, FDMA, OFDMA, SC-FDMA, etc. For example, the base station 114a in the RAN 104 / 113 and the WTRUs 102a, 102b, 102c may implement a radio technology such as Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access (UTRA), which may use Wideband CDMA (WCDMA) to establish the air interface 115 / 116 / 117. WCDMA may include communication protocols such as High Speed ​​Packet Access (HSPA) and / or Evolved HSPA (HSPA+). HSPA may include High Speed ​​Downlink (DL) Packet Access (HSDPA) and / or High Speed ​​UL Packet Access (HSUPA).

[0029] In one embodiment, the base station 114a and the WTRUs 102a, 102b, 102c may implement a radio technology such as Evolved UMTS Terrestrial Radio Access (E-UTRA) that may establish the air interface 116 using Long Term Evolution (LTE) and / or Advanced LTE (LTE-A) and / or Advanced LTE Pro (LTE-A Pro).

[0030] In one embodiment, the base station 114a and the WTRUs 102a, 102b, 102c may implement a radio technology such as NR radio access, which may establish the air interface 116 using New Radio (NR).

[0031] In one embodiment, the base station 114a and the WTRUs 102a, 102b, 102c may implement multiple radio access technologies. For example, the base station 114a and the WTRUs 102a, 102b, 102c may together implement LTE radio access and NR radio access, for example using dual connectivity (DC) principles. Thus, the air interface utilized by the WTRUs 102a, 102b, 102c may be characterized by multiple types of radio access technologies and / or transmissions transmitted to / from multiple types of base stations (e.g., eNBs and gNBs).

[0032] In other embodiments, the base station 114a and the WTRUs 102a, 102b, 102c may implement radio technologies such as IEEE 802.11 (i.e., Wireless Fidelity (WiFi)), IEEE 802.16 (i.e., Worldwide Interoperability for Microwave Access (WiMAX)), CDMA2000, CDMA2000 1X, CDMA2000 EV-DO, Interim Standard 2000 (IS-2000), Interim Standard 95 (IS-95), Interim Standard 856 (IS-856), Global System for Mobile Communications (GSM), Enhanced Data rates for GSM Evolution (EDGE), GSM EDGE (GERAN), etc.

[0033] Figure 1A The base station 114b in the example may be, for example, a wireless router, a Home NodeB, a Home eNodeB, or an access point, and may utilize any suitable RAT to facilitate wireless connectivity in a local area, such as a business location, a home, a vehicle, a campus, an industrial facility, an air corridor (e.g., for use by drones), a road, and the like. In one embodiment, the base station 114b and the WTRUs 102c, 102d may implement a radio technology such as IEEE 802.11 to establish a wireless local area network (WLAN). In one embodiment, the base station 114b and the WTRUs 102c, 102d may implement a radio technology such as IEEE 802.15 to establish a wireless personal area network (WPAN). In another embodiment, the base station 114b and the WTRUs 102c, 102d may utilize a cellular-based RAT (e.g., WCDMA, CDMA2000, GSM, LTE, LTE-A, LTE-A Pro, NR, etc.) to establish a micro-micro cell base station or a femto cell base station. As Figure 1A As shown, base station 114b may have a direct connection to Internet 110. Thus, base station 114b may not need to access Internet 110 via CN 106 / 115.

[0034] The RAN 104 / 113 may be in communication with the CN 106 / 115, which may be any type of network configured to provide voice, data, applications and / or Voice over Internet Protocol (VoIP) services to one or more of the WTRUs 102a, 102b, 102c, 102d. The data may have different quality of service (QoS) requirements, such as different throughput requirements, latency requirements, error tolerance requirements, reliability requirements, data throughput requirements, mobility requirements, etc. The CN 106 / 115 may provide call control, billing services, mobile location-based services, prepaid calling, Internet connectivity, video distribution, etc., and / or perform high-level security functions, such as user authentication. Although not described in detail in the text, the CN 106 / 115 may be configured to provide voice, data, applications and / or Voice over Internet Protocol (VoIP) services to one or more of the WTRUs 102a, 102b, 102c, 102d. The data may have different quality of service (QoS) requirements, such as different throughput requirements, latency requirements, error tolerance requirements, reliability requirements, data throughput requirements, mobility requirements, etc. Figure 1AAlthough not shown in the figure, it will be appreciated that the RAN 104 / 113 and / or the CN 106 / 115 may be in direct or indirect communication with other RANs that employ the same RAT as the RAN 104 / 113 or a different RAT. For example, in addition to being connected to the RAN 104 / 113, which may utilize NR radio technology, the CN 106 / 115 may also be in communication with another RAN (not shown) that employs GSM, UMTS, CDMA 2000, WiMAX, E-UTRA, or WiFi radio technology.

[0035] The CN 106 / 115 may also act as a gateway for the WTRUs 102a, 102b, 102c, 102d to access the PSTN 108, the Internet 110, and / or other networks 112. The PSTN 108 may include a circuit-switched telephone network that provides plain old telephone service (POTS). The Internet 110 may include a global system of interconnected computer networks and devices that use common communication protocols, such as the Transmission Control Protocol (TCP), the User Datagram Protocol (UDP), and / or the Internet Protocol (IP) in the TCP / IP Internet protocol suite. The networks 112 may include wired communication networks and / or wireless communication networks owned and / or operated by other service providers. For example, the networks 112 may include another CN connected to one or more RANs, which may employ the same RAT as the RAN 104 / 113 or a different RAT.

[0036] Some or all of the WTRUs 102a, 102b, 102c, 102d in the communication system 100 may include multi-mode capabilities (e.g., the WTRUs 102a, 102b, 102c, 102d may include multiple transceivers for communicating with different wireless networks via different wireless links). Figure 1A The illustrated WTRU 102c may be configured to communicate with the base station 114a, which may employ a cellular-based radio technology, and with the base station 114b, which may employ an IEEE 802 radio technology.

[0037] Figure 1B is a system diagram illustrating an example WTRU 102. Figure 1B As shown, the WTRU 102 may include, among other things, a processor 118, a transceiver 120, a transmit / receive element 122, a speaker / microphone 124, a keypad 126, a display / touchpad 128, non-removable memory 130, removable memory 132, a power source 134, a global positioning system (GPS) chipset 136, and / or other peripherals 138. It will be appreciated that the WTRU 102 may include any sub-combination of the foregoing elements while remaining consistent with an embodiment.

[0038] The processor 118 may be a general purpose processor, a special purpose processor, a conventional processor, a digital signal processor (DSP), a plurality of microprocessors, one or more microprocessors associated with a DSP core, a controller, a microcontroller, an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) circuit, any other type of integrated circuit (IC), a state machine, etc. The processor 118 may perform signal decoding, data processing, power control, input / output processing, and / or any other functionality that enables the WTRU 102 to operate in a wireless environment. The processor 118 may be coupled to the transceiver 120, which may be coupled to the transmit / receive element 122. Although Figure 1B The processor 118 and the transceiver 120 are depicted as separate components, but it is understood that the processor 118 and the transceiver 120 may be integrated together in an electronic package or chip.

[0039] The send / receive element 122 may be configured to send a signal to a base station (e.g., base station 114a) or receive a signal from the base station via an air interface 116. For example, in one embodiment, the send / receive element 122 may be an antenna configured to send and / or receive an RF signal. In one embodiment, the send / receive element 122 may be a transmitter / detector configured to send and / or receive, for example, an IR, UV or visible light signal. In another embodiment, the send / receive element 122 may be configured to send and / or receive both an RF signal and an optical signal. It should be understood that the send / receive element 122 may be configured to send and / or receive any combination of wireless signals.

[0040] Although the transmit / receive element 122 Figure 1B Although depicted as a single element in the figure, the WTRU 102 may include any number of transmit / receive elements 122. More specifically, the WTRU 102 may employ MIMO technology. Thus, in one embodiment, the WTRU 102 may include two or more transmit / receive elements 122 (e.g., multiple antennas) for transmitting and receiving wireless signals over the air interface 116.

[0041] The transceiver 120 may be configured to modulate signals to be transmitted by the transmit / receive element 122 and to demodulate signals received by the transmit / receive element 122. As noted above, the WTRU 102 may have multi-mode capabilities. For example, the transceiver 120 may include multiple transceivers for enabling the WTRU 102 to communicate via multiple RATs, such as NR and IEEE 802.11.

[0042] The processor 118 of the WTRU 102 may be coupled to and may receive user input data from a speaker / microphone 124, a keypad 126, and / or a display / touchpad 128 (e.g., a liquid crystal display (LCD) display unit or an organic light emitting diode (OLED) display unit). The processor 118 may also output user data to the speaker / microphone 124, the keypad 126, and / or the display / touchpad 128. In addition, the processor 118 may access information from and store data in any type of suitable memory, such as a non-removable memory 130 and / or a removable memory 132. The non-removable memory 130 may include a random access memory (RAM), a read-only memory (ROM), a hard disk, or any other type of memory storage device. The removable memory 132 may include a subscriber identity module (SIM) card, a memory stick, a secure digital (SD) memory card, and the like. In other embodiments, the processor 118 may access information from and store data in a memory that is not physically located on the WTRU 102, such as on a server or a home computer (not shown).

[0043] The processor 118 may receive power from the power source 134, and may be configured to distribute and / or control power to the other components in the WTRU 102. The power source 134 may be any suitable device for powering the WTRU 102. For example, the power source 134 may include one or more dry cell batteries (e.g., nickel-cadmium (NiCd), nickel-zinc (NiZn), nickel-metal hydride (NiMH), lithium-ion (Li-ion), etc.), solar cells, fuel cells, etc.

[0044] The processor 118 may also be coupled to the GPS chipset 136, which may be configured to provide location information (e.g., longitude and latitude) regarding the current location of the WTRU 102. In addition to or in lieu of the information from the GPS chipset 136, the WTRU 102 may receive location information from a base station (e.g., base stations 114a, 114b) over the air interface 116 and / or determine its location based on the timing of signals received from two or more nearby base stations. It will be appreciated that the WTRU 102 may acquire location information by any suitable location-determination method while remaining consistent with an embodiment.

[0045] The processor 118 may also be coupled to other peripherals 138, which may include one or more software modules and / or hardware modules that provide additional features, functionality, and / or wired or wireless connectivity. For example, the peripherals 138 may include an accelerometer, an electronic compass, a satellite transceiver, a digital camera (for photos and / or video), a universal serial bus (USB) port, a vibration device, a television transceiver, a hands-free headset, module, a frequency modulation (FM) radio unit, a digital music player, a media player, a video game player module, an Internet browser, a virtual reality and / or augmented reality (VR / AR) device, an activity tracker, etc. The peripheral device 138 may include one or more sensors, which may be one or more of the following: a gyroscope, an accelerometer, a Hall effect sensor, a magnetometer, an orientation sensor, a proximity sensor, a temperature sensor, a time sensor; a geolocation sensor; an altimeter, a light sensor, a touch sensor, a magnetometer, a barometer, a gesture sensor, a biometric sensor, and / or a humidity sensor.

[0046] The WTRU 102 may include a full-duplex radio for which transmission and reception of some or all signals (e.g., associated with specific subframes for both UL (e.g., for transmission) and downlink (e.g., for reception)) may be concurrent and / or simultaneous. The full-duplex radio may include an interference management unit 139 for reducing and / or substantially eliminating self-interference via signal processing performed via hardware (e.g., a choke) or via a processor (e.g., a separate processor (not shown) or via the processor 118). In one embodiment, the WRTU 102 may include a half-duplex radio for which transmission and reception of some or all signals (e.g., associated with specific subframes for both UL (e.g., for transmission) or downlink (e.g., for reception)) may be concurrent and / or simultaneous.

[0047] Figure 1C 1 is a system diagram illustrating the RAN 104 and the CN 106 in accordance with an embodiment. As noted above, the RAN 104 may employ an E-UTRA radio technology to communicate with the WTRUs 102a, 102b, 102c over the air interface 116. The RAN 104 may also be in communication with the CN 106.

[0048] The RAN 104 may include evolved Node-Bs 160a, 160b, 160c, though it will be appreciated that the RAN 104 may include any number of evolved Node-Bs while remaining consistent with an embodiment. The evolved Node-Bs 160a, 160b, 160c may each include one or more transceivers for communicating with the WTRUs 102a, 102b, 102c over the air interface 116. In one embodiment, the evolved Node-Bs 160a, 160b, 160c may implement MIMO technology. Thus, the evolved Node-B 160a, for example, may use multiple antennas to transmit wireless signals to and / or receive wireless signals from the WTRU 102a.

[0049] Each of the evolved Node Bs 160a, 160b, 160c may be associated with a particular cell (not shown) and may be configured to handle radio resource management decisions, handover decisions, scheduling of users in the UL and / or DL, etc. Figure 1C As shown, the eNode-Bs 160a, 160b, 160c may communicate with one another via an X2 interface.

[0050] Figure 1C The illustrated CN 106 may include a mobility management entity (MME) 162, a serving gateway (SGW) 164, and a packet data network (PDN) gateway (or PGW) 166. While each of the foregoing elements is depicted as being part of the CN 106, it should be understood that any of these elements may be owned and / or operated by an entity other than the CN operator.

[0051] The MME 162 may be connected to each of the evolved Node-Bs 162a, 162b, 162c in the RAN 104 via an S1 interface and may serve as a control node. For example, the MME 162 may be responsible for authenticating users of the WTRUs 102a, 102b, 102c, bearer activation / deactivation, selecting a particular serving gateway during an initial attach of the WTRUs 102a, 102b, 102c, etc. The MME 162 may provide a control plane function for switching between the RAN 104 and other RANs (not shown) that employ other radio technologies, such as GSM and / or WCDMA.

[0052] The SGW 164 may be connected to each of the evolved Node-Bs 160a, 160b, 160c in the RAN 104 via an S1 interface. The SGW 164 may generally route and forward user data packets to / from the WTRUs 102a, 102b, 102c. The SGW 164 may perform other functions, such as anchoring the user plane during an inter-evolved Node-B handover, triggering paging when DL data is available for the WTRUs 102a, 102b, 102c, managing and storing the context of the WTRUs 102a, 102b, 102c, and the like.

[0053] The SGW 164 may be connected to the PGW 166, which may provide the WTRUs 102a, 102b, 102c with access to packet-switched networks, such as the Internet 110, to facilitate communications between the WTRUs 102a, 102b, 102c and IP-enabled devices.

[0054] The CN 106 may facilitate communications with other networks. For example, the CN 106 may provide the WTRUs 102a, 102b, 102c with access to circuit-switched networks, such as the PSTN 108, to facilitate communications between the WTRUs 102a, 102b, 102c and traditional land-line communications devices. For example, the CN 106 may include, or may be in communication with, an IP gateway (e.g., an IP Multimedia Subsystem (IMS) server) that acts as an interface between the CN 106 and the PSTN 108. In addition, the CN 106 may provide the WTRUs 102a, 102b, 102c with access to other networks 112, which may include other wired networks and / or wireless networks owned and / or operated by other service providers.

[0055] Although the WTRU Figures 1A to 1D Although described as wireless terminals, it is contemplated that in certain representative embodiments, such terminals may (eg, temporarily or permanently) use a wired communications interface with a communications network.

[0056] In a representative embodiment, the other network 112 may be a WLAN.

[0057] A WLAN in infrastructure basic service set (BSS) mode may have an access point (AP) for the BSS and one or more stations (STAs) associated with the AP. The AP may have access or an interface to a distribution system (DS) or another type of wired / wireless network that carries traffic to and / or out of the BSS. Traffic originating from outside the BSS and destined for the STA may be reached by the AP and may be delivered to the STA. Traffic originating from the STA and destined for a target outside the BSS may be transmitted to the AP to be delivered to the corresponding target. Traffic between STAs within the BSS may be transmitted by the AP, for example, wherein the source STA may transmit traffic to the AP, and the AP may deliver traffic to the target STA. Traffic between STAs within the BSS may be considered and / or referred to as point-to-point traffic. Direct link establishment (DLS) may be utilized to transmit point-to-point traffic between the source STA and the target STA (e.g., directly between them). In certain representative embodiments, the DLS may use 802.11e DLS or 802.11z tunnel DLS (TDLS). A WLAN using an independent BSS (IBSS) mode may not have an AP, and STAs within or using the IBSS (eg, all STAs in the STA) may communicate directly with each other. The IBSS communication mode may sometimes be referred to herein as an "ad hoc" communication mode.

[0058] When using the 802.11ac infrastructure operation mode or a similar operation mode, the AP may send beacons on a fixed channel (such as a primary channel). The primary channel may be a fixed width (e.g., a 20 MHz wide bandwidth) or a width dynamically set via signaling. The primary channel may be a working channel of the BSS and may be used by the STA to establish a connection with the AP. In certain representative embodiments, carrier sense multiple access / collision avoidance (CSMA / CA) may be implemented, for example, in an 802.11 system. For CSMA / CA, a STA (e.g., each STA) (including the AP) may sense the primary channel. If the primary channel is sensed / detected and / or determined to be busy by a particular STA, the particular STA may back off. One STA (e.g., only one station) may transmit at any given time in a given BSS.

[0059] High throughput (HT) STAs may communicate using a 40 MHz wide channel, for example, via a combination of a primary 20 MHz channel and adjacent or non-adjacent 20 MHz channels to form a 40 MHz wide channel.

[0060] Very high throughput (VHT) STA can support 20MHz, 40MHz, 80MHz and / or 160MHz wide channels. 40MHz channels and / or 80MHz channels can be formed by combining continuous 20MHz channels. 160MHz channels can be formed by combining 8 continuous 20MHz channels, or by combining two non-continuous 80MHz channels (this can be called 80+80 configuration). For 80+80 configuration, after channel coding, the data can pass through a segment parser that can divide the data into two streams. Each stream can be processed by inverse fast Fourier transform (IFFT) and time domain processing separately. These streams can be mapped to two 80MHz channels, and the data can be sent by sending STA. At the receiver of the receiving STA, the above-mentioned operation for 80+80 configuration can be reversed, and the combined data can be transmitted to the medium access control (MAC).

[0061] 802.11af and 802.11ah support operating modes below 1GHz. Channel operating bandwidths and carriers are reduced in 802.11af and 802.11ah relative to those used in 802.11n and 802.11ac. 802.11af supports 5MHz, 10MHz, and 20MHz bandwidths in the TV White Space (TVWS) spectrum, and 802.11ah supports 1MHz, 2MHz, 4MHz, 8MHz, and 16MHz bandwidths using non-TVWS spectrum. According to a representative embodiment, 802.11ah can support instrument type control / machine type communications, such as MTC devices in macro coverage areas. MTC devices may have certain capabilities, such as limited capabilities, including support for (e.g., only support for) certain bandwidths and / or limited bandwidths. MTC devices may include batteries with battery life above a threshold (e.g., to maintain very long battery life).

[0062] WLAN systems that can support multiple channels and channel bandwidths (such as 802.11n, 802.11ac, 802.11af, and 802.11ah) include channels that can be designated as primary channels. The primary channel may have a bandwidth equal to the maximum common operating bandwidth supported by all STAs in the BSS. The bandwidth of the primary channel may be set and / or limited by a STA (which supports the minimum bandwidth operation mode) from all STAs operating in the BSS. In the example of 802.11ah, for STAs (e.g., MTC-type devices) that support (e.g., only support) 1MHz mode, the primary channel may be 1MHz wide, even if the AP and other STAs in the BSS support 2MHz, 4MHz, 8MHz, 16MHz, and / or other channel bandwidth operation modes. Carrier sensing and / or network allocation vector (NAV) settings may depend on the state of the primary channel. If the primary channel is busy, for example, because a STA (supporting only the 1MHz operation mode) is sending to the AP, the entire available band may be considered busy even if most of the band remains idle and may be available.

[0063] In the United States, the available frequency band for 802.11ah is 902MHz to 928MHz. In South Korea, the available frequency band is 917.5MHz to 923.5MHz. In Japan, the available frequency band is 916.5MHz to 927.5MHz. The total bandwidth available for 802.11ah is 6MHz to 26MHz, depending on the country code.

[0064] Figure 1D 1 is a system diagram illustrating the RAN 113 and the CN 115 in accordance with an embodiment. As noted above, the RAN 113 may employ NR radio technology to communicate with the WTRUs 102a, 102b, 102c over the air interface 116. The RAN 113 may also be in communication with the CN 115.

[0065] The RAN 113 may include gNBs 180a, 180b, 180c, though it will be appreciated that the RAN 113 may include any number of gNBs while remaining consistent with an embodiment. The gNBs 180a, 180b, 180c may each include one or more transceivers for communicating with the WTRUs 102a, 102b, 102c over the air interface 116. In one embodiment, the gNBs 180a, 180b, 180c may implement MIMO technology. For example, the gNBs 180a, 180b may utilize beamforming to send signals to and / or receive signals from the gNBs 180a, 180b, 180c. Thus, the gNB 180a may, for example, use multiple antennas to send wireless signals to and / or receive wireless signals from the WTRU 102a. In one embodiment, the gNBs 180a, 180b, 180c may implement carrier aggregation technology. For example, the gNB 180a may transmit multiple component carriers to the WTRU 102a (not shown). A subset of these component carriers may be on an unlicensed spectrum, while the remaining component carriers may be on a licensed spectrum. In one embodiment, the gNBs 180a, 180b, 180c may implement coordinated multi-point (CoMP) technology. For example, the WTRU 102a may receive coordinated transmissions from the gNB 180a and gNB 180b (and / or gNB 180c).

[0066] The WTRUs 102a, 102b, 102c may communicate with the gNBs 180a, 180b, 180c using transmissions associated with parameter sets that may be scalable. For example, OFDM symbol spacing and / or OFDM subcarrier spacing may vary for different transmissions, different cells, and / or different portions of the wireless transmission spectrum. The WTRUs 102a, 102b, 102c may communicate with the gNBs 180a, 180b, 180c using subframes or transmit time intervals (TTIs) of varying or scalable lengths (e.g., containing different numbers of OFDM symbols and / or continuously varying absolute time lengths).

[0067] The gNBs 180a, 180b, 180c may be configured to communicate with the WTRUs 102a, 102b, 102c in a standalone configuration and / or a non-standalone configuration. In a standalone configuration, the WTRUs 102a, 102b, 102c may communicate with the gNBs 180a, 180b, 180c while not accessing other RANs (e.g., such as the eNodeBs 160a, 160b, 160c). In a standalone configuration, the WTRUs 102a, 102b, 102c may use one or more of the gNBs 180a, 180b, 180c as mobility anchor points. In a standalone configuration, the WTRUs 102a, 102b, 102c may communicate with the gNBs 180a, 180b, 180c using signals in an unlicensed band. In a non-standalone configuration, the WTRUs 102a, 102b, 102c may communicate / connect with the gNB 180a, 180b, 180c while also communicating / connecting with another RAN, such as the eNodeB 160a, 160b, 160c. For example, the WTRUs 102a, 102b, 102c may implement the DC principle to communicate with one or more gNBs 180a, 180b, 180c and one or more eNodeBs 160a, 160b, 160c substantially simultaneously. In a non-standalone configuration, the eNodeB 160a, 160b, 160c may act as a mobility anchor for the WTRUs 102a, 102b, 102c, and the gNB 180a, 180b, 180c may provide additional coverage and / or throughput for serving the WTRUs 102a, 102b, 102c.

[0068] Each of the gNBs 180a, 180b, 180c may be associated with a specific cell (not shown) and may be configured to handle radio resource management decisions, handover decisions, scheduling of users in UL and / or DL, support of network slicing, dual connectivity, interworking between NR and E-UTRA, routing of user plane data towards a user plane function (UPF) 184a, 184b, routing of control plane information towards an access and mobility management function (AMF) 182a, 182b, etc. Figure 1D As shown, gNBs 180a, 180b, and 180c may communicate with each other via an Xn interface.

[0069] Figure 1DThe illustrated CN 115 may include at least one AMF 182a, 182b, at least one UPF 184a, 184b, at least one session management function (SMF) 183a, 183b, and possible data networks (DNs) 185a, 185b. Although each of the foregoing elements is depicted as part of the CN 115, it should be understood that any of these elements may be owned and / or operated by an entity other than the CN operator.

[0070] The AMF 182a, 182b may be connected to one or more of the gNBs 180a, 180b, 180c in the RAN 113 via the N2 interface and may act as a control node. For example, the AMF 182a, 182b may be responsible for authenticating users of the WTRU 102a, 102b, 102c, support of network slicing (e.g., handling of different PDU sessions with different requirements), selecting a specific SMF 183a, 183b, management of registration areas, termination of NAS signaling, mobility management, etc. The AMF 182a, 182b may use network slicing in order to customize CN support for the WTRU 102a, 102b, 102c based on the type of services utilized by the WTRU 102a, 102b, 102c. For example, different network slices may be established for different use cases, such as services relying on ultra-reliable low latency (URLLC) access, services relying on enhanced mobile broadband (eMBB) access, and / or services for machine type communication (MTC) access, etc. The AMF 162 may provide a control plane function for switching between the RAN 113 and other RANs (not shown) that employ other radio technologies, such as LTE, LTE-A, LTE-A Pro, and / or non-3GPP access technologies, such as WiFi.

[0071] The SMF 183a, 183b may be connected to the AMF 182a, 182b in the CN 115 via the N11 interface. The SMF 183a, 183b may also be connected to the UPF 184a, 184b in the CN 115 via the N4 interface. The SMF 183a, 183b may select and control the UPF 184a, 184b, and configure traffic routing through the UPF 184a, 184b. The SMF 183a, 183b may perform other functions such as managing and allocating WTRU IP addresses, managing PDU sessions, controlling policy enforcement and QoS, providing downlink data notification, etc. The PDU session type may be IP-based, non-IP-based, Ethernet-based, etc.

[0072] The UPF 184a, 184b may be connected to one or more of the gNBs 180a, 180b, 180c in the RAN 113 via the N3 interface, which may provide the WTRUs 102a, 102b, 102c with access to packet-switched networks, such as the Internet 110, to facilitate communications between the WTRUs 102a, 102b, 102c and IP-enabled devices. The UPF 184, 184b may perform other functions, such as routing and forwarding packets, enforcing user plane policies, supporting multi-homed PDU sessions, handling user plane QoS, buffering downlink packets, and providing mobility anchoring, etc.

[0073] The CN 115 may facilitate communications with other networks. For example, the CN 115 may include an IP gateway (e.g., an IP Multimedia Subsystem (IMS) server) that acts as an interface between the CN 115 and the PSTN 108 or may communicate with an IP gateway that acts as an interface between the CN 115 and the PSTN 108. In addition, the CN 115 may provide the WTRUs 102a, 102b, 102c with access to other networks 112, which may include other wired networks and / or wireless networks owned and / or operated by other service providers. In one embodiment, the WTRUs 102a, 102b, 102c may be connected to local data networks (DNs) 185a, 185b through the UPF 184a, 184b via an N3 interface to the UPF 184a, 184b and an N6 interface between the UPF 184a, 184b and the DN 185a, 185b.

[0074] Given that Figures 1A to 1D as well as Figures 1A to 1D , one or more or all of the functions described herein for one or more of the following may be performed by one or more simulation devices (not shown): WTRU 102a-102d, base station 114a-114b, evolved Node B 160a-160c, MME 162, SGW 164, PGW 166, gNB 180a-180c, AMF 182a-182ab, UPF 184a-184b, SMF 183a-183b, DN 185a-185b and / or any other device described herein. The simulation device may be one or more devices configured to mimic one or more or all of the functions described herein. For example, the simulation device may be used to test other devices and / or simulate network and / or WTRU functions.

[0075] The simulation device may be designed to implement one or more tests of other devices in a laboratory environment and / or in an operator network environment. For example, one or more simulation devices may perform one or more functions or all functions while being fully or partially implemented and / or deployed as part of a wired and / or wireless communication network in order to test other devices within the communication network. One or more simulation devices may perform one or more functions or all functions while being temporarily implemented / deployed as part of a wired and / or wireless communication network. The simulation device may be directly coupled to another device for testing purposes and / or may use over-the-air wireless communications to perform testing.

[0076] One or more simulation devices can perform one or more (including all) functions without being implemented / deployed as part of a wired and / or wireless communication network. For example, the simulation device can be used in a test scenario in a test laboratory and / or a non-deployed (e.g., testing) wired and / or wireless communication network to implement testing of one or more components. One or more simulation devices can be test equipment. Direct RF coupling and / or wireless communication via an RF circuit system (e.g., which can include one or more antennas) can be used by the simulation device to send and / or receive data.

[0077] As used herein, the term network may include an application management function (AMF), a location management function (LMF), a base station (e.g., gNB), and / or a next generation radio access network (NG-RAN).

[0078] The WTRU may send or receive a physical channel or a reference signal according to (eg, at least one) spatial domain filter.As described herein, the term "beam" may be used to refer to a spatial domain filter.

[0079] The WTRU may use a spatial domain filter to send a physical channel and / or signal. The spatial domain filter may be the same as the spatial domain filter used to receive the RS (e.g., CSI-RS / SSB). The WTRU sending may be referred to as the "target" in this article. The received RS and / or SS block may be referred to as the "reference" and / or "source" in this article. The WTRU may allegedly send a target physical channel and / or signal based on the spatial relationship of the reference RS and / or SS block.

[0080] The WTRU may transmit a first physical channel and / or signal according to the same spatial domain filter as used to transmit a second physical channel and / or signal. The first transmission and the second transmission may be referred to as a "target" and a "reference" (e.g., or "source"), respectively. The WTRU may be said to transmit a first (e.g., target) physical channel and / or signal according to a spatial relationship with reference to a second (e.g., reference) physical channel or signal.

[0081] The spatial relationship may be implicit, configured by RRC, and / or signaled by MAC CE or DCI. For example, the WTRU may implicitly send the PUSCH and / or DM-RS for the PUSCH according to a spatial domain filter that is similar (e.g., the same) as the sounding reference signal (SRS) indicated by the SRI indicated in the DCI or configured by RRC. In certain scenarios, the spatial relationship may be configured by RRC, for example, for an SRS resource indicator (SRI) and / or signaled by a MAC CE for PUCCH. The spatial relationship may be referred to herein as a "beam indication."

[0082] The WTRU may receive a first (e.g., target) downlink channel and / or signal based on a spatial domain filter or spatial reception parameter that is similar (e.g., the same) as a second (e.g., reference) downlink channel and / or signal. For example, there may be an association between a physical channel (e.g., PDCCH and / or PDSCH) and a corresponding DM-RS. For example, if the first signal and the second signal are reference signals, an association may exist when the WTRU is configured with a quasi-co-location (QCL) assumption type D between corresponding antenna ports. The association may be configured as a transmission configuration indicator (TCI) state. The WTRU may be provided with an indication of the association between the CSI-RS and / or SS block and the DM-RS, for example, by an index to a TCI state set (configured by the RRC and / or signaled by the MAC CE). The indication may be referred to as a "beam indication" herein.

[0083] As described herein, a transmit and receive point (TRP) may also refer to one or more of a transmit point (TP), a receive point (RP), a radio remote head (RRH), a distributed antenna (DA), a base station (BS), a sector (e.g., of a BS), and / or a cell (e.g., a geographic cell area served by a BS). As described herein, a multiple TRP may refer to one or more of an MTRP, an M-TRP, and / or multiple TRPs.

[0084] The WTRU may report a subset of channel state information (CSI) components. The CSI components may correspond to one or more of a CSI-RS resource indicator (CRI), an SSB resource indicator (SSBRI), an indication of a panel for reception at the WTRU (e.g., such as a panel identifier or a group identifier), one or more measurements, and / or any other (e.g., suitable) channel state information. The one or more measurements may include one or more of L1-RSRP, L1-SINR (e.g., cri-RSRP, cri-SINR, ssb-Index-RSRP, ssb-Index-SINR) obtained from SSB and / or CSI-RS. Other channel state information may include one or more of a minimum rank indicator (RI), a channel quality indicator (CQI), a precoding matrix indicator (PMI), and / or a layer index (LI), etc.

[0085] The WTRU may receive a synchronization signal / physical broadcast channel (SS / PBCH) block. The SS / PBCH block (SSB) may include one or more of a primary synchronization signal (PSS), a secondary synchronization signal (SSS), and / or a physical broadcast channel (PBCH). The WTRU may monitor, receive, and / or attempt to decode the SSB. For example, the WTRU may monitor, receive, and / or attempt to decode the SSB during one or more of initial access, initial synchronization, radio link monitoring (RLM), cell search, cell handover, etc.

[0086] The WTRU may measure and report channel state information (CSI). One or more of the following may apply. CSI reporting configuration may be performed. CSI reporting quantity may be configured. The CSI reporting quantity may include one or more of a channel quality indicator (CQI), a rank indicator (RI), a precoding matrix indicator (PMI), a CSI-RS resource indicator (CRI), a layer indicator (LI), etc. CSI reporting may be configured. The CSI reporting type may include one or more of aperiodic, semi-persistent, and / or periodic. A CSI reporting codebook may be configured. The CSI reporting codebook may include one or more of type I, type II, type II port selection, etc. The CSI reporting frequency may be configured.

[0087] One or more settings associated with a CSI-RS resource set may be configured. For example, one or more settings associated with a CSI-RS resource may include one or more of a non-zero power (NZP) CSI-RS resource for channel measurement, an NZP-CSI-RS resource for interference measurement, and / or a CSI-IM resource for interference measurement. For example, an NZP CSI-RS resource may include one or more of an NZP CSI-RS resource ID, periodicity and offset, QCL information and TCI state, resource mapping (e.g., number of ports, density, code division multiplexing (CDM) type), etc.

[0088] The WTRU may indicate, determine and / or be configured with one or more reference signals. The WTRU may monitor, receive and / or measure one or more parameters based on (e.g., corresponding) reference signals. For example, the parameters may include one or more of SS-RSRP, CSI-RSRP, SS-SINR, CSI-SINR, RSSI, CLI-RSSI; and / or SRS-RSRP.

[0089] One or more of the following may apply. SS reference signal received power (SS-RSRP) may be measured, for example, based on a synchronization signal (e.g., a demodulation reference signal (DMRS) and / or SSS in a PBCH). For example, SS-RSRP may be defined as a linear average of the power contributions of resource elements (REs) carrying the corresponding synchronization signal. For example, power scaling of the reference signal may be performed when measuring RSRP. For example, if SS-RSRP is used for L1-RSRP, the measurement may be performed based on a CSI reference signal and / or a synchronization signal.

[0090] For example, CSI-RSRP may be measured based on a linear average of power contributions of resource elements (REs) carrying corresponding CSI-RS.CSI-RSRP measurements may be configured within measurement resources of configured CSI-RS opportunities.

[0091] For example, the SS signal-to-noise and interference ratio (SS-SINR) may be measured based on a synchronization signal (e.g., DMRS or SSS in PBCH). For example, the SS-SINR may be defined as the linear average of the power contributions of the resource elements (REs) carrying the corresponding synchronization signal divided by the linear average of the noise and interference power contributions. For example, if the SS-SINR is used for L1-SINR, the noise and interference power measurement may be performed based on resources configured by a higher layer.

[0092] For example, the CSI-SINR may be measured based on a linear average of the power contributions of the resource elements (REs) carrying the corresponding CSI-RS divided by a linear average of the noise and interference power contributions. For example, if the CSI-SINR is used for L1-SINR, the noise and interference power measurement may be performed based on resources configured by a higher layer. For example, if the CSI-SINR is not used for L1-SINR, the noise and interference power may be measured based on the resources carrying the corresponding CSI-RS.

[0093] For example, a received signal strength indicator (RSSI) may be measured based on an average of total power contributions in a configured OFDM symbol and / or bandwidth. Power contributions may be received from different resources. For example, power contributions may be received from one or more of co-channel serving and non-serving cells, adjacent channel interference, thermal noise, etc.

[0094] A cross-layer interference received signal strength indicator (CLI-RSSI) may be measured based on an average of total power contributions in configured OFDM symbols of configured time and frequency resources. Power contributions may be received from different resources (e.g., cross-layer interference, co-channel serving and non-serving cells, adjacent channel interference, thermal noise, etc.).

[0095] For example, the sounding reference signal RSRP (SRS-RSRP) may be measured based on a linear average of the power contributions of the resource elements (REs) carrying the corresponding SRS.

[0096] A CSI report configuration (e.g., CSI-ReportConfigs) may be associated with a BWP (e.g., indicated by a BWP-Id). For example, a CSI report configuration may be used to configure one or more of the following parameters: CSI-RS resources and / or CSI-RS resource sets for channel and interference measurement; CSI-RS report configuration type, including periodic, semi-persistent, and aperiodic; CSI-RS transmission periodicity for periodic and semi-persistent CSI reporting; CSI-RS transmission slot offset for periodic, semi-persistent, and / or aperiodic CSI reporting; CSI-RS transmission slot offset list for semi-persistent and / or aperiodic CSI reporting; time limit for channel and / or interference measurement; reporting band configuration (e.g., wideband / subband CQI, PMI, etc.); thresholds and calculation modes for reporting quantities (e.g., CQI, RSRP, SINR, LI, RI, etc.); codebook configuration; group-based beam reporting; CQI table; subband size; non-PMI port indication; port index, etc.

[0097] A CSI-RS resource set (e.g., NZP-CSI-RS-ResourceSet) may include one or more CSI-RS resources. For example, a CSI-RS resource set may include an NZP-CSI-RS-Resource and a CSI-ResourceConfig. The WTRU may be configured with one or more of the following parameters for a given CSI-RS resource: CSI-RS periodicity and slot offset for periodic and / or semi-persistent CSI-RS resources; CSI-RS resource mapping; bandwidth portion of the CSI-RS to which the configured CSI-RS is allocated; and / or a reference to a TCI-State. The CSI-RS resource mapping may be used to define one or more of the number, density, CDM type, OFDM symbols, and / or subcarrier occupancy of CSI-RS ports. The reference to the TCI state includes the QCL source RS and / or the corresponding QCL type.

[0098] The WTRU may be configured with one or more RS resource sets. For example, the RS resource set configuration may include one or more of the following: an RS resource set ID, one or more RS resources for the RS resource set, repetition (e.g., on or off), aperiodic trigger offset (e.g., one of 0 to 6 time slots); and / or TRS information (e.g., true or false).

[0099] The WTRU may be configured with one or more RS resources. For example, the RS resource configuration may include one or more of the following: RS resource ID, resource mapping (e.g., REs in PRBs), power control offset (e.g., a value from -8, ..., 15), power control offset from SS (e.g., -3dB, 0dB, 3dB, 6db), scrambling ID, periodicity and offset, and / or QCL information (e.g., based on TCI state).

[0100] The attributes of the grant and / or assignment may include one or more of the following: frequency allocation; an aspect of time allocation (e.g., duration); priority; modulation and coding scheme (MCS); transport block size; number of spatial layers; number of transport blocks; TCI state, CRI and / or SRI; number of repetitions; whether the repetition scheme is type A or type B; whether the grant is a configured grant type 1, type 2 and / or a dynamic grant; whether the assignment is a dynamic assignment and / or a semi-persistently scheduled (e.g., configured) assignment; a configured grant index and / or a semi-persistent assignment index; periodicity (e.g., of a configured grant and / or assignment); channel access priority class (CAPC); and / or (e.g., any) parameters provided by MAC and / or by RRC in the DCI for scheduling grants and / or assignments.

[0101] The DCI indication may include one or more of the following: an explicit indication of a DCI field or RNTI (e.g., a CRC for masking a PDCCH) and / or an implicit indication of an attribute. The implicit indication of the attribute may include one or more of the following: a DCI format, a DCI size, a Coreset and / or a search space, an aggregation level, and / or a first resource element of the received DCI (e.g., an index of a first CCE). The mapping between attributes and values ​​may be signaled by RRC and / or MAC.

[0102] RS may refer to one or more of an RS resource, an RS resource set, an RS port, and / or an RS port group. RS may additionally or alternatively refer to one or more of SSB, CSI-RS, SRS, DM-RS, TRS, PRS, and / or PTRS. RS may additionally or alternatively refer to one or more of a sounding reference signal (SRS), a channel state information-reference signal (CSI-RS), a demodulation reference signal (DM-RS), a phase tracking reference signal (PT-RS), and / or a synchronization signal block (SSB). A channel may refer to one or more of a PDCCH, a PDSCH, a physical uplink control channel (PUCCH), a physical uplink shared channel (PUSCH), a physical random access channel (PRACH), and the like. Additionally or alternatively, an RS resource set may refer to an RS resource and / or a beam group. A beam report may refer to one or more of a CSI measurement, a CSI report, and / or a beam measurement. The techniques associated with beam resource prediction described herein may additionally or alternatively be used for beam resources belonging to one or more cells and one or more TRPs.

[0103] Beam management (BM) may be performed. For example, beam management (e.g., in NR) may be based on a hierarchical beam search principle. For example, the WTRU may (e.g., first) select a preferred wide beam from a set of wide beams (e.g., SSBs) and / or feed back the selected SSB to the gNB (e.g., via PRACH in an initial access phase). The gNB may, for example, send one or more (e.g., multiple) refined beams based on the selected SSB as initial spatial information. The refined beams may be sent using a CSI reporting configuration. The CSI reporting configuration may be used to measure a preferred beam (e.g., a refined narrow beam) and feed back the preferred beam (e.g., a refined narrow beam) to the gNB.

[0104] For example, in an NR network, the BM may be classified into a P-1 process, a P-2 process, and / or a P-3 process. The P-1 process may be used to implement WTRU measurements, such as on different Tx beams and / or to support Tx beam / WTRU Rx beam selection. For example, P-1 may be based on a wide beam search (e.g., to determine the optimal receive beam). The WTRU may select a suitable SSB for UL transmission (e.g., based on DL measurements). For example, the P-2 process may be used to implement WTRU measurements on different Tx beams (e.g., changing the inter / intra Tx beam). The P-2 process may be based on a gNB configuring a set of CSI-RS resource sets. For example, each CSI-RS resource set may have one or more (e.g., multiple) resources. The WTRU may use one or more resources to perform measurements for beam refinement and / or select a refined beam. The gNB may determine the refined beams for downlink and uplink reception, for example, based on the WTRU's downlink measurements on one or more Tx refined beams of the gNB. For example, the P-3 process may be used to implement WTRU measurements on the same Tx beam. Measurements on the same Tx beam may be performed to change the WTRU Rx beam (e.g., if the WTRU uses beamforming). The WTRU may determine the WTRU Tx beam to use for uplink transmissions, for example, based on WTRU downlink measurements on one or more Rx beams associated with the WTRU. The gNB may repeat the refined beam for the WTRU to perform measurements.

[0105] L1-RSRP (e.g., CRI-RSRP, SSB-index-RSRP) may be used in a BM process, e.g., in an NR network. L1-RSRP (e.g., CRI-RSRP, SSB-index-RSRP) may be used in a BM process to identify a downlink beam, e.g., the best downlink beam. SSB-Index-RSRP may include RSRP measurements and / or reports, e.g., based on SSB (e.g., using different beams). CRI-RSRP may include RSRP measurements and / or reports, e.g., based on CSI-RS resources (e.g., using different beams).

[0106] The beam selection technique may be based on beam scanning on the gNB side and / or the WTRU side. The beam selection technique may result in beam scanning and / or measurements on multiple (e.g., a large number of) antennas on the gNB and WTRU sides (e.g., particularly for a large number of antennas, high frequencies, etc.). The WTRU may report one or more beams in the beam management process. For example, when selecting the best beam, the WTRU may report up to four beams in the beam management process (e.g., based on RSRP). The WTRU may report one or more beams in the NR network.

[0107] Artificial intelligence / machine learning (AI / ML) technology can be provided. For example, artificial intelligence can be a behavior exhibited by a machine that can, for example, simulate cognitive functions to sense, reason, adapt, and act, etc. For example, machine learning can refer to a type of algorithm based on learning through experience (e.g., data) without programming (e.g., configuring a set of rules). Machine learning can be a subset of AI. For example, different machine learning paradigms can be utilized based on the nature and / or feedback of the data available for learning the algorithm. For example, supervised learning methods can involve learning a function that maps inputs to outputs, for example, based on labeled training examples. Each training example can include a pair including an input and a corresponding output. Unsupervised learning methods can involve detecting patterns in data, for example, without pre-existing labels. Reinforcement learning methods can involve performing a series of actions in an environment, for example, to maximize cumulative rewards. Machine learning algorithms can be applied using a combination and / or interpolation of the methods disclosed herein. For example, a semi-supervised learning method can use a combination of a certain amount of labeled data and a certain amount of unlabeled data during training. Semi-supervised learning can fall between unsupervised learning (e.g., no labeled training data) and supervised learning (e.g., only labeled training data). For example, semi-supervised learning can include a mixture of unsupervised learning (e.g., no labeled training data) and supervised learning (e.g., only labeled training data).

[0108] Deep learning (DL) may refer to a class of machine learning algorithms including artificial neural networks (e.g., in particular DNNs). Artificial neural networks may be based on biological systems. Deep neural networks (DNNs) may include a class of machine learning models based on the human brain. For example, in DNNs, inputs may be linearly transformed and / or passed through a nonlinear activation function once or multiple times (e.g., multiple times). DNNs may include one or more (e.g., multiple) layers, for example, each of which may include a linear transformation and / or a given nonlinear activation function. DNNs may be trained using training data, for example, via a back propagation algorithm. DNNs may be shown in prior art performance in multiple fields (e.g., in one or more of speech, vision, natural language, etc.) and / or for various supervised, unsupervised and / or semi-supervised machine learning settings. In this article, the term AI / ML-based method / processing may refer to implementing behavior and / or learning based on data, such as without the configuration of steps or action sequences (e.g., explicit configuration). Methods (e.g., these methods) may enable learning of complex behaviors.

[0109] Devices and / or functions may, for example, use AI / ML to predict and / or refine beam pairs and associated quality metrics. As used herein, the terms "machine learning" and / or "ML" may refer to a type of algorithm that solves problems based on learning through experience ("data") without being explicitly programmed with a set of configuration rules. Machine learning may be considered a subset of AI. Different machine learning paradigms may be envisioned based on the nature of the data and / or feedback available for learning the algorithm. For example, a supervised learning method may involve learning a function that maps an input to an output based on labeled training examples, where one or more (e.g., each) training examples may be pairs including an input and a corresponding output. An unsupervised learning method may involve detecting patterns in data without pre-existing labels. A reinforcement learning method may involve performing a series of actions in an environment to maximize a cumulative reward. In an example, it is possible to apply one or more machine learning algorithms using a combination and / or interpolation of the methods mentioned herein (e.g., the above methods). For example, a semi-supervised learning method may use a combination of a small amount of labeled data with a large amount of unlabeled data during training. In this regard, semi-supervised learning may fall between unsupervised learning (eg, no labeled training data) and supervised learning (eg, with labeled training data).

[0110] Figure 2A2 is a diagram of an example system environment 201 in which an AI / ML 209 model may be implemented. The AI / ML model 209 may be implemented at a WTRU and / or a network. The AI / ML 209 model may include model data and one or more algorithms and / or functions configured to learn from received input data 207 to train the AI / ML 209 and / or generate output 215. The input data 207 may be input in one or more formats, such as an image format, an audio format (e.g., a spectrogram or other audio format), a tensor format (e.g., including a one-dimensional or multi-dimensional array), and / or another data type capable of being input into the AI / ML 209 algorithm. The input data 207 may be the result of pre-processing 205 that may be performed on the raw data 203, or the input data 207 may include the raw data 203 itself. The raw data 203 may include image data, text data, audio data, or another information sequence (such as a network information sequence associated with a communication network) and / or other types of data. Preprocessing 205 may include format changes or other types of processing to generate input data 207 in a format for input into the AI / ML 209 algorithm. Output 215 may be generated by the AI / ML 209 algorithm in one or more formats, such as a tensor, a text format (e.g., a word, sentence, or other text sequence), a numerical format (e.g., a prediction), an audio format, an image format (e.g., including a video format), another data sequence format, or another output format. As described herein, output 215 may include one or more of the following: predicted beam pairs, predicted measurements for one or more beam pairs based on stored training data samples and a predictive beam refinement configuration, a predictive beam identifier, a codebook identifier, a predicted layer 1 received signal received power (L1-RSRP) value, and / or a predicted angle of arrival (AoA) based on the predicted measurements, for example.

[0111] AI / ML 209 may be implemented using software and / or hardware as described herein. AI / ML 209 may be stored as computer executable instructions on a computer readable medium that may be accessed by one or more processors for execution as described herein. Example AI / ML environments and / or libraries include TENSORFLOW, TORCH, PYTORCH, MATLAB, GOOGLE CLOUD AI and AUTOML, AMAZON SAGEMAKER, AZURE MACHINE LEARNING STUDIO, and / or ORACLE MACHINE LEARNING.

[0112] The AI / ML 209 may include one or more algorithms configured for unsupervised learning. Unsupervised learning may be implemented using an AI / ML 209 algorithm that learns from input data 207 without training toward a specific target output. For example, during unsupervised learning, the AI / ML 209 algorithm may receive unlabeled data as input data 207 and determine patterns or similarities in the input data 207 without additional intervention (e.g., updating parameters and / or hyperparameters). The AI / ML 209 algorithm configured to implement unsupervised learning may include an algorithm configured to identify patterns, groupings, clusters, anomalies, and / or similarities or other associations in the input data 207. For example, the AI / ML may implement a hierarchical clustering algorithm, a k-means clustering algorithm, a K-nearest neighbor (K-NN) algorithm, an anomaly detection algorithm, a principal component analysis algorithm, and / or a priori algorithms. The AI / ML 209 algorithm configured for unsupervised learning may be implemented on a single device or distributed across multiple devices such that the output 215 or portions thereof may be aggregated at one or more devices for further processing and / or implementation in other downstream algorithms or processes, as may be further described herein.

[0113] AI / ML 209 may include one or more algorithms configured for supervised learning. Supervised learning may be implemented using an AI / ML 209 algorithm trained during a training process to determine a predictive model using known results. The AI / ML 209 algorithm may be characterized by parameters and / or hyperparameters that may be trained during a training process. These parameters may include values ​​derived during a training process. These parameters may include weights, coefficients, and / or biases. AI / ML 209 may also include hyperparameters. Hyperparameters may include values ​​for controlling the learning process. Hyperparameters may include learning rates, number of epochs, batch sizes, number of layers, number of nodes in each layer, number of kernels (e.g., CNN), step sizes (e.g., CNN), kernel sizes in pooling layers (e.g., CNN), and / or other hyperparameters. In some cases, certain parameters and hyperparameters may be used interchangeably.

[0114] The AI / ML 209 may be trained during supervised learning by inputting training data into the AI / ML 209 algorithm and adjusting parameters and / or hyperparameters toward a known target output 215 while minimizing the loss or error in the output 215 generated by the AI / ML 209 algorithm. The raw data 203 may include or be divided into training data, validation data, and / or test data for training, validating, and / or testing the AI / ML 209 algorithm during supervised learning, respectively. The training data, validation data, and / or test data may be pre-processed from the raw data 203 for input into the AI / ML 209 algorithm. During supervised learning, the training data may be labeled before being input into the AI / ML 209. The training data may be labeled to teach the AI / ML 209 algorithm to learn from the labeled data and to test the accuracy of the AI / ML 209 implemented on unlabeled input data 207 during production / implementation of the AI / ML 209 algorithm or a similar AI / ML 209 algorithm utilizing similar parameters and / or hyperparameters. The training data may be used to fit the parameters of the AI / ML 209 model using an optimization function, such as a loss function or an error function. Typically, the training data includes pairs of input data 207 and corresponding target outputs 215, and the parameters may be trained to generate these input data and target outputs (e.g., within a threshold loss or error). The trained or fitted AI / ML 209 model may receive validation data as input to evaluate the model fit to the training data set while tuning the hyperparameters of the AI / ML 209 model. The AI / ML 209 model may receive test data to evaluate the final model fit to the training data set and evaluate the performance of the AI / ML 209 model. One or more of training, validation, and / or testing may be performed during supervised learning for different types of AI / ML 209 models.

[0115] Supervised learning can be implemented for various types of AI / ML 209 algorithms, including algorithms that implement linear regression, logistic regression, neural networks (NNs), decision trees, Bayesian logic, random forests, and / or support vector machines (SVMs). NNs and deep NNs (DNNs) are popular examples of algorithms used in AI / ML models that can be trained using supervised learning. However, AI / ML 209 models can implement one or more NN-based and / or non-NN-based algorithms. Various examples of NNs include: perceptrons, multilayer perceptrons (MLPs), feedforward NNs, fully connected NNs, convolutional neural networks (CNNs), recursive NNs (RNNs), long short-term memory (LSTM) NNs, and / or residual NNs (ResNets). A perceptron is a NN that includes a function that multiplies its input by a learned weight coefficient to generate an output value. A feedforward NN is a NN that receives input at one or more nodes of an input layer and moves information in the direction of one or more nodes of an output layer through one or more hidden layers. In a feedforward NN, one or more nodes of a given layer may be connected to one or more nodes of another layer. A fully connected NN is a NN that includes an input layer, one or more hidden layers, and an output layer. In a fully connected NN, each node in one layer is connected to each node in another layer of the NN. MLP is a feed-forward NN of the fully connected class. CNN is a NN with one or more convolutional layers configured to perform convolutions. Various types of NNs may have elements including one or more CNNs or convolutional layers, such as generative adversarial networks (GANs). GANs may include conditional GANs (CGANs), cycle-consistent GANs (CycleGANs), StyleGANs, DiscoGANs, and / or IsGANs. GANs may include a generator submodel and a discriminator submodel. The generator submodel may be configured to receive input data and pass true and independently generated data to the discriminator submodel. The discriminator submodel may be configured to receive true and independently generated data from the generator, discriminate the true and independently generated data, and provide feedback to the generator submodel during training to improve the generator submodel's function in independently generating outputs based on received inputs. GAN is a popular model for generating data types or data sequences such as image data, audio data, and / or text. RNNs are NNs that are recursive in nature, as nodes include feedback connections and internal hidden states (e.g., memory) that allow outputs from nodes in the NN to affect subsequent inputs to the same node. LSTM NNs may be similar to RNNs, as nodes have feedback connections and internal hidden states (e.g., memory). However, LSTM NNs may include additional gates to allow LSTM NNs to learn long-term dependencies between sequences of data. ResNets are NNs that may include skip connections to skip one or more layers of the NN.An autoencoder may be a form of AI / ML 109 that may be implemented for supervised learning such that parameters and / or hyperparameters may be updated during the training process. The parameters and / or hyperparameters may be associated with the encoder portion and / or the decoder portion of the autoencoder. Some NNs include one or more attention layers or functions to enhance or focus on some portions of the input data while reducing or weakening other portions.

[0116] Different types of NNs and / or layers may be implemented to process different types of data and / or produce different types of outputs. For example, a NN may include one or more convolutional layers (e.g., for a CNN or a GAN) that may be commonly used to process image data and / or audio data (e.g., spectrograms). Each convolutional layer may vary according to various convolutional layer parameters or hyperparameters, such as kernel size (e.g., the field of view of the convolution), step size (e.g., the step size of the kernel when traversing the image), padding (e.g., for processing image boundaries), and / or input and output sizes. The processed image may include one or more dimensions (e.g., a row of pixels or a two-dimensional array of pixels). Pixels may be represented according to one or more values ​​(e.g., one or more integer values ​​representing color and / or intensity) that may be received by the convolutional layer. A kernel, which may also be referred to as a convolution matrix or mask, may be a matrix for extracting and / or transforming features from received input data. Kernels may be used for blurring, sharpening, edge detection, and the like. Example kernel sizes may include matrices of 3×3, 5×5, 10×10, etc. (e.g., in pixels of a 2D image). A stride may be a parameter used to identify the amount by which a kernel moves over the image data. An example default stride has a size of 1 or 2 within the matrix (e.g., in pixels of a 2D image). Padding may include the amount of data added to the border of the image data as the image data is processed by the kernel (e.g., in pixels of a 2D image). The kernel may move over the input image data (e.g., according to the stride length) and perform a dot product with an overlapping input region to obtain an activation value for the region. The output of each convolutional layer may be provided to the next layer of the NN, or as the output of the NN itself (e.g., image data, feature maps, etc.) with convolution-based updated features.

[0117] The NN may include similar types of layers (e.g., convolutional layers, feed-forward layers, fully connected layers, etc.) and / or have similar or different configurations for each layer (e.g., size, number of nodes, etc.). The NN may also or alternatively include one or more layers of NNs of different types or subsets that may be interconnected for training and / or implementation as described herein. For example, a NN may include both convolutional layers and feed-forward or fully connected layers.

[0118] Figure 2BAn example of a neural network 209a is illustrated. The purpose of training can be to use input 207a as training data and / or to adjust the Figure 2B One or more weights (e.g., which may be referred to as neuron weights and / or link weights) indicated as w and x in , such that the output 215 from the neural network 209a approaches the desired target value associated with the input 207a values ​​for the training data. In an example, the neural network may include three layers (e.g., Figure 2B ). During training, for a given input, the difference between the output value and the expected value can be calculated and / or the difference can be used to update one or more weights in the neural network. For example, if a significant (e.g., large) difference between the output value and the expected value is observed, then one or more relatively significant (e.g., large) changes in one or more weights can be expected. A small difference (e.g., between the output value and the expected value) can include one or more relatively small changes in one or more weights. For example, for positioning, the input 207a can be a reference signal parameter and / or the output 215 can be an estimated position. The expected value can be positioning information acquired with high accuracy by a global navigation satellite system (GNSS).

[0119] Once the neural network 209a completes its training, the difference between the output 215 value and the expected value may be below a threshold. The neural network 209a may be applied or implemented after training for positioning by feeding the input data 207a and / or by estimating or predicting the output 215 as the expected result of the associated input 207a. The output 215 may be an estimated position and / or location of the WTRU.

[0120] Training the neural network 209a may include identifying one or more of the following information: inputs to the neural network; expected outputs associated with the inputs; and / or actual outputs from the neural network to which the target value is compared.

[0121] In an example, a neural network model may be characterized by one or more parameters and / or hyperparameters, which may include: the number of weights and / or the number of layers in the neural network.

[0122] As used herein, the term "deep learning" may refer to a class of machine learning algorithms that employ artificial neural networks (e.g., deep neural networks (DNNs)) that are loosely inspired from biological systems and / or include at least one hidden layer. DNNs may be a special class of machine learning models inspired by the human brain, in which inputs are linearly transformed and / or passed through a nonlinear activation function one or more times (e.g., multiple times). DNNs may include one or more (e.g., multiple) layers, in which one or more (e.g., each) layer includes a linear transformation and / or a given nonlinear activation function. DNNs may be trained using training data via a back-propagation algorithm. Recently, DNNs have demonstrated state-of-the-art performance in various fields (e.g., speech, vision, natural language, etc.) and / or for various machine learning settings (e.g., supervised, unsupervised, and / or semi-supervised).

[0123] Figure 2C 201a for training and implementing an AI / ML model including a NN 209a. However, other types of AI / ML models (e.g., including NN models and / or non-NN models) may be similarly trained and / or implemented. NN 209a may be trained and / or implemented on one or more devices to determine and / or update parameters and / or hyperparameters 217 of NN 209a. Raw data 203a may be generated from one or more sources. For example, raw data 203a may include image data, text data, audio data, or another information sequence (such as a network information sequence associated with a communication network) and / or other types of data. Raw data 203a may be pre-processed at 205a to generate training data 207a. The pre-processing may include formatting or other types of processing of changes so as to generate training data 207a in a format for input into NN 209a.

[0124] NN 209a may include one or more layers 211. The configuration of NN 209a and / or layer 211 may be based on parameters and / or hyperparameters 217. As described herein, parameters may include weights or coefficients and / or biases for nodes or functions in layer 211. Hyperparameters may include learning rate, number of epochs, batch size, number of layers, number of nodes in each layer, number of kernels (e.g., CNN), step size (e.g., CNN), kernel size in pooling layer (e.g., CNN), and / or other hyperparameters. As described herein, NN 109a may include a feedforward NN, a fully connected NN, a CNN, a GAN, an RNN, a ResNet, and / or one or more other types of NNs. NN 209a may include one or more different types of NNs or different layers for different types of NNs. For example, NN 109a may include one or more separate layers with one or more configurations.

[0125] During the training process, training data 207a may be input into NN 209a and may be used to learn parameters and / or tune hyperparameters 217. Training may be performed by initializing parameters and / or hyperparameters of NN 209a, generating and / or accessing training data 207a, inputting training data 207a into NN 209a, calculating the error or loss from the output of NN 209a to the target output 215a via loss function 213 (e.g., using gradient descent and / or associated back propagation), and / or updating parameters and / or hyperparameters 217.

[0126] The loss function 213 may be implemented using a back-propagation based gradient update and / or gradient descent technique, such as stochastic gradient descent (SGD), synchronous SGD, asynchronous SGD, batch gradient descent, and / or mini-batch gradient descent. Examples of loss or error functions may include functions for determining a squared error loss, a mean squared error (MSE) loss, a mean absolute error loss, a mean absolute percentage error loss, a mean squared logarithmic error loss, a pixel-based loss, a pixel-by-pixel loss, a cross entropy loss, a logarithmic loss, and / or a benchmark-based loss. The loss function may be implemented based on one or more quality metrics such as a signal-to-noise ratio (SNR) metric or another signal or image quality metric.

[0127] An optimizer may be implemented with the loss function 213. An optimizer may be an algorithm or function configured to adjust properties of the NN 209a, such as learning rate and / or weights, to improve the accuracy of the NN 209a and / or reduce loss or error. The optimizer may be implemented to update parameters and / or hyperparameters 217 of the NN 209a.

[0128] The training process may be iterated to update parameters and / or hyperparameters 217 until an end condition is achieved. The end condition may be met when the output of the NN 209a is within a predefined threshold of the target output 215a.

[0129] After the training process is completed, the trained NN 209a or its parts may be stored for implementation by one or more devices. The trained NN 209a or its parts may be implemented in other downstream algorithms or processes, as may be further described herein. The trained NN 209a or its parts may be implemented on the same device that performs the training. The trained NN 209a or its parts may be sent or otherwise provided to another device for implementation. For example, NN 209b, 209c may include one or more parts of the trained NN 209a. NN 209b and NN 209c receive corresponding input data 207b, 207c and generate corresponding outputs 215b, 215c. The outputs 215b, 215c may be generated in one or more formats, such as tensors, text formats (e.g., words, sentences or other text sequences), numerical formats (e.g., predictions), audio formats, image formats (e.g., including video formats), another data sequence format and / or another output format. The outputs 215b, 215c may be aggregated at one or more devices for further processing and / or implementation in other downstream algorithms or processes, as may be further described herein.

[0130] Alternatively or additionally, after the training process is completed, the trained parameters and / or tuned hyperparameters 217 or portions thereof may be stored for implementation by one or more devices. The trained parameters and / or tuned hyperparameters 217 or portions thereof may be implemented in other downstream algorithms or processes, as may be further described herein. The trained parameters and / or tuned hyperparameters 217 or portions thereof may be implemented on the same device that performs the training. The trained parameters and / or tuned hyperparameters 217 or portions thereof may be sent or otherwise provided to another device for implementation. For example, based on the trained parameters and / or tuned hyperparameters 217 being sent or otherwise provided to another one or more devices that can implement NN 209b, 209c. For example, NN 209b, 209c may be constructed at another device based on the trained parameters and / or tuned hyperparameters 217 or portions thereof. NN 209b and NN 209c may be configured according to parameters / hyperparameters 217 or portions thereof to receive respective input data 207b, 207c and generate respective outputs 215b, 215c. Outputs 215b, 215c may be generated in one or more formats, such as tensors, text formats (e.g., words, sentences, or other text sequences), numerical formats (e.g., predictions), audio formats, image formats (e.g., including video formats), another data sequence format, and / or another output format. Outputs 215b, 215c may be aggregated at one or more devices for further processing and / or implementation in other downstream algorithms or processes, as may be further described herein.

[0131] The AI / ML models and / or algorithms described herein may be implemented on one or more devices. For example, the AI / ML 209 may be implemented in whole or in part on one or more devices, such as one or more WTRUs, one or more base stations, and / or one or more other network entities, such as network servers. An exemplary network in which the AI / ML may be distributed may include a federated network. A federated network may include a group of dispersed devices, each device including an AI / ML. Figure 2C As shown, AI / ML 209b and AI / ML 209c may be distributed on separate devices. Figure 2C Two models are shown (e.g., AI / ML 209b and AI / ML 209c), but any number of models may be implemented across any number of devices. AI / ML may be implemented for collaborative learning, where the AI / ML is trained across multiple devices. In another example, the AI / ML may be trained at a centralized location or device and one or more portions of the AI / ML or trained parameters and / or tuned hyperparameters may be distributed to decentralized locations. For example, updated parameters or hyperparameters may be transmitted to one or more devices to update and / or implement the AI / ML thereon.

[0132] Figure 2D 201b is a schematic diagram of an example system environment 201b for training and / or implementing an AI / ML model including an autoencoder. An autoencoder (AE) 209a may include one or more DNNs 211. For example, the AE 209a may include a class of DNNs 211 that arise in the context of an unsupervised machine learning setting. Data (e.g., high-dimensional data) 207b may be transformed (e.g., nonlinearly) into a lower-dimensional latent vector, for example, using a DNN-based encoder. The lower-dimensional latent vector may be used to reproduce the high-dimensional data, for example, using a nonlinear decoder. The encoder may be represented as E(x; W e ), where x can be high-dimensional data, and W e The decoder 219b can be represented as D(z; W d ), where z can be a low-dimensional latent representation, and W d can represent the parameters of the decoder. Using training data 207b{x1,…,x N}, for example, the autoencoder can be trained. For example, the following equation (1) can be used to train the autoencoder:

[0133]

[0134] Equation (1) can be solved (e.g., approximately solved), for example, using a back-propagation algorithm. Trained encoder Can be used to compress high-dimensional data, and the trained decoder Can be used to decompress the latent representation. The autoencoder 209b can be trained and / or implemented on one or more devices to determine and / or update parameters and / or hyperparameters 217 of the NN 209a. The training data 207b may include measurements, such as RS measurements. The target output 215b may include one or more of the following: predicted beam pairs, predicted measurements for one or more beam pairs based on stored training data samples and predictive beam refinement configurations, predictive beam identifiers, codebook identifiers, predictive layer 1 received signal received power (L1-RSRP) values, and / or predicted angle of arrival (AoA) based on predicted measurements, for example. The training data may be associated with a wider beam or beam pair. The output may be associated with a narrower beam or beam pair.

[0135] As described herein, the terms artificial intelligence (AI), machine learning (ML), deep learning (DL), and DNN are used interchangeably. Although the techniques described herein are exemplified based on learning in a wireless communication system, they may additionally or alternatively be used in other systems, including, for example, any type of transmission, communication system, and / or service, etc.

[0136] Artificial intelligence (AI) / machine learning (ML) for air interfaces (e.g., in NR) may be implemented. For example, AI / ML may be used in connection with beam management, including beam prediction in the time and / or spatial domains (e.g., for overhead and latency reduction), beam selection accuracy improvement, etc.

[0137] Figure 3 is a schematic diagram of an example system 300 for training an AI / ML model. At 302, training data can be input. The training data can include historical L-RSRP combination weight vectors. At 304, the AI / ML can predict a next weight vector, for example, based on the training data. At 306, the system can output an error measurement. For example, the training data can include one or more measurements of a reference signal set.

[0138] Figure 4 4 is a schematic diagram of an example system 400 for inferring beam prediction performance. At 402, a beamformed signal may be received. At 404, an AI / ML combined weight vector may be applied. For example, the combined weight vector may be trained offline via historical samples. At 406, the WTRU and / or network entity may check L1-RSRP performance. For example, if L1-RSRP is below a threshold, beam scanning and / or retraining (e.g., further training) may be triggered.

[0139] Figure 5is an example illustration of a process 500 for predictive beam refinement. At 502, the WTRU may receive a configuration for predictive beam refinement, the configuration including at least one threshold associated with information to be used for predicting and selecting a beam pair. At 504, the WTRU may perform measurements on a set of reference signals (RS). At 506, the WTRU may determine to store the measurements as training data samples in a memory. The WTRU may determine to store one or more measurements based on a comparison of the measurements to at least one threshold. At 508, the WTRU may determine to start training or retraining a model for performing predictions associated with one or more beam pairs, for example based on the at least one threshold. At 510, the WTRU may send an indication that the stored training data samples have reached at least one threshold.

[0140] The WTRU may predict measurements, for example, for one or more beam pairs. For example, the WTRU may predict measurements based on stored training data samples and / or a predictive beam refinement configuration. The WTRU may report one or more of a predictive beam identifier, a codebook identifier, a predictive layer 1 received signal received power (L1-RSRP) value, or a predicted angle of arrival (AoA), for example, based on the predicted measurements. The at least one threshold may include at least one of: a buffer size threshold associated with a memory for indicating a maximum number of samples of beam quality measurements or beam combining weight pairs, a threshold associated with a maximum buffer size or a remaining buffer size, a threshold associated with one or more measurements to be performed on one or more RS sets, a threshold number of comparisons of one or more measurements to one or more thresholds, a completed AI / ML model, and / or one or more thresholds for determining to store samples in a memory.

[0141] A determination may be made at the WTRU and / or a network entity to begin predicting measurements for one or more beam pairs. The at least one threshold may be associated with information to be used for predicting and / or selecting one or more beam pairs.

[0142] For example, when one or more of the measurements are greater than or equal to at least one threshold, the WTRU may store the one or more measurements in the memory as training data samples. For example, when one or more of the measurements are less than at least one threshold, the WTRU may repeat the determination to store the one or more measurements in the memory as training data samples.

[0143] The WTRU may calculate a prediction accuracy of measurements for one or more beam pairs, for example based on stored training data samples. The at least one threshold may include a value that increases with the stored training data samples. Predicting measurements for one or more beam pairs may be based on a weight vector.

[0144] The beam inference mechanism may be performed at the gNB and / or the WTRU. The beam inference mechanism at the gNB may be used to reduce the WTRU computational complexity. Additionally or alternatively, the beam inference mechanism at the gNB may enable the WTRU to provide specific feedback (e.g., L1-RSRP) to the gNB. RSRP compression methods may be implemented, for example, to reduce AS feedback overhead. For example, if beam inference is performed at the WTRU (e.g., because the WTRU may predict a preferred beam (e.g., CRI) instead of reporting the measured L1-RSRP), the feedback overhead may be reduced. For example, if beam inference is performed at the WTRU (e.g., due to the computational complexity of beam inference), power consumption may increase.

[0145] Beam inference techniques can be based on spatial domain (SD) beam selection, for example, to predict wide beams or thin beams. Beam management and / or beam prediction based on AI / ML frameworks can address challenges in beam measurement and reporting and / or training and validation of AI / ML models, for example in scenarios where hierarchical spatial relationships and beam associations are in different frequency ranges.

[0146] Beam management (e.g., in NR) may include, for example, selection and / or maintenance of beam directions for unicast transmissions (e.g., including control channels and data channels) between a gNB and a WTRU. Beam management procedures may be categorized into one or more of beam determination, beam measurement and reporting, beam switching, beam indication, and / or beam restoration. For example, in beam determination, a base station (BS) and a WTRU may identify a beam direction for unicast transmissions (e.g., including control and data channels) (e.g., to ensure good radio link quality).

[0147] The WTRU may measure the link quality of one or more (e.g., multiple) transmit (Tx) and / or receive (Rx) beam pairs, for example, after the link is established. The WTRU may report the measurement results to the BS. WTRU mobility, orientation, and channel blocking may affect the radio link quality of the TX and RX beam pairs. For example, if the quality of the current beam pair degrades, the BS and the WTRU may switch to another beam pair (e.g., with better radio link quality). For example, if the quality of the current beam pair drops below a threshold, the BS and the WTRU may switch to another beam pair. For example, in order to switch to another beam pair, the BS and / or the WTRU may monitor the quality of the current beam pair and / or one or more other beam pairs.

[0148] For example, the beam indication procedure may be used when the BS allocates a Tx beam to the WTRU (e.g., via DL control signaling). Beam restoration may include a restoration procedure. For example, the restoration procedure may be used if the link between the BS and the WTRU can no longer be maintained.

[0149] Beam measurement and reporting may be used for higher frequencies (e.g., FR2-1, FR2-2, sub-THz, THz, etc.). Beam selection may include beam scanning at the gNB side and / or the WTRU side. The gNB and / or the WTRU may, for example, scan the beams in a predetermined direction. The WTRU may select, for example, the beam pair with the highest received signal quality (e.g., L1-RSRP, L1-SINR, etc.). Selecting the beam pair with the highest received signal quality may be used to support the transmission and / or reception of a downlink data channel (e.g., PDSCH).

[0150] Beam management may include beam scanning and / or measurements on multiple (e.g., a large number) of antennas on the gNB side and the WTRU side, such as a gNB for L beams and a WTRU for M beams. A total of LM transmit instants may be used and / or beam scanning may be performed on LM combinations, for example to identify the best beam pair. The latency and / or computational complexity associated with conventional beam selection (e.g., particularly involving higher frequency bands such as THz / sub-THz, where latency may be shown to be on the order of seconds) may be reduced by using data-driven inference (e.g., including the use of AI / ML) to implement beam prediction in the time and / or spatial domains (e.g., which may reduce overhead and latency and / or improve beam selection accuracy).

[0151] AI / ML may be used for beam management, including, for example, beam inference in the time and / or spatial domain for overhead, latency reduction, and / or beam selection accuracy. The expected performance of the AI / ML inference model may depend on one or more factors, including, for example: spatial / temporal / frequency variations of the channel (e.g., due to changes in WTRU mobility or environment); beam resolution; and / or the number of scanned beams. Techniques may be provided to implement AI / ML-based inference for predictive beam establishment / refinement. The AI / ML model may be trained based on (e.g., conventional) beam scanning and / or reporting processes. The process may include inference at the WTRU, gNB, and / or include separate and joint processing scenarios.

[0152] There may be challenges associated with the prediction of a refined (e.g., narrow) beam from a subset of measurement beams, such as where the predicted refined / narrow beam may not be included in the first measurement beam (e.g., or reference signal, RS) and / or a subset of the first measurement beam / RS. The gNB may send an inferred / predicted refined / narrow beam, such as a second set of RS / beams. The WTRU may measure the predicted RS / beams and / or may select one or more predicted beams / RS. The WTRU may provide the selected beam to the gNB, such as so that the gNB may determine a transmit configuration indicator (TCI) state to be activated for PDCCH and / or PDSCH transmission.

[0153] The WTRU may perform a beam search, which may, for example, increase computational complexity and / or latency. The WTRU may perform a beam search if, for example, the spatial information of the predicted RS / beam is unknown to the WTRU. The computational complexity and / or latency may further increase with an increasing number of antennas at the gNB and / or WTRU (e.g., the likelihood that the inferred refined / narrow beam is not in the measurement beam and / or a subset of the measurement beams increases). The WTRU may perform (e.g., may frequently perform) additional beam searches, which may, for example, increase beam management latency.

[0154] Techniques are described herein that may reduce measurement and / or reporting overhead for beam scanning, for example, through data driven beam prediction. Techniques associated with configuring predictive beam refinement may be provided. Techniques associated with data driven WTRU specific beam refinement may be provided. Techniques associated with indicating predictive beam operation and / or reporting predicted beams (e.g., beam index and measurement values) between a gNB and a WTRU may be provided. Techniques associated with model performance monitoring and / or retraining triggering may be provided.

[0155] Techniques associated with WTRU predictive beam refinement may be provided herein. The gNB may configure the WTRU with predictive beam refinement / establishment. The WTRU may utilize beam scanning for AI / ML model training. The WTRU may indicate data-driven predictions to the gNB. The WTRU may signal predictive Rx beams to the gNB. The AI / ML model may perform monitoring for retraining.

[0156] Techniques associated with gNB predictive beam refinement may be provided herein. The gNB may predict the Rx beam using WTRU measurements and / or additional WTRU indications. The WTRU measurements and / or additional WTRU indications may be historical. The gNB may signal the predictive receive beam to the WTRU. The AI / ML model may perform monitoring, e.g., for retraining.

[0157] Techniques described herein may reduce measurement and / or reporting overhead for beam scanning, for example, through data-driven beam prediction. One or more of the following may be provided: techniques associated with configuring predictive beam refinement, techniques associated with data-driven WTRU-specific beam refinement, techniques associated with indicating predictive beam operation and / or reporting predicted beams (e.g., beam indices and measurements) between a gNB and a WTRU, and / or techniques associated with model performance monitoring and / or retraining triggering.

[0158] AI / ML may be an example enabler for data-driven predictive row beam refinement, however, other data-driven techniques may additionally or alternatively be used. RS resource sets may refer to RS resources and / or beam groups. i It can represent the beam quality measurement sample i. i w may be a vector for multiple beams or may be a scalar, for example, when a single beam is used (e.g., L1-RSRP). i can represent the beam combining / receiver weight sample i. i This can be a matrix for multiple beams and / or a vector for a single beam.

[0159] Methods for de-coupling and / or co-ordinating WTRU and gNB predictive beam refinement may be provided. Data-driven beam prediction and / or establishment may be performed, for example, via an AI / ML model. For example, configuration information associated with beam scanning may be received by the WTRU. The configuration information may include a buffer size. The buffer size may be associated with the AI / ML model and / or one or more thresholds. The one or more thresholds may include a quality threshold (e.g., a quality metric) and / or a threshold amount of memory containing stored training data samples. For example, beam scanning may be performed across one or more beams based on the configuration information. The AI / ML model may be trained, for example, based on beam scanning performed across one or more beams. For example, the AI / ML model may be implemented at the WTRU and / or a network node (e.g., a gNB). Training the AI / ML model may alternatively or additionally include determining a quality metric. The quality metric may be associated with each of the one or more beams for which beam scanning is performed. Training the AI / ML model may alternatively or additionally include, for example, filling a buffer associated with the AI / ML model based on whether a determined quality metric associated with each of the one or more beams is greater than a threshold (e.g., received via configuration information). For example, using the AI / ML model, a beam of the one or more beams may be identified and / or an indication of the identified beam may be transmitted. After the beam has been identified, the quality metric associated with the identified beam may be periodically measured. For example, if the quality metric associated with the identified beam is below a second threshold, the AI / ML model may be retrained.

[0160] A method for WTRU predictive beam refinement may be provided. The gNB may configure the WTRU with predictive beam operation. The configuration may include one or more of aspects such as buffers, beam quality, number of beams, etc. The WTRU may perform model training using beam scanning based on RS transmissions. The training may be performed online and / or offline. The WTRU may transmit an indication to the gNB that predictive beam refinement / establishment is possible. For example, beam prediction may reduce and / or mitigate conventional beam scanning (e.g., fewer RS ​​transmissions or no RS transmissions). Receiver beamforming may be signaled by the WTRU, for example, explicitly signaled to the gNB. For example, if the predicted refined / narrow beam is not included in the first measurement beam (e.g., or reference signal, RS) and / or a subset of the first measurement beam / RS, the WTRU may signal receiver beamforming.

[0161] The WTRU may be configured by the gNB for predictive beam operation. The WTRU may include data-driven capabilities (e.g., AI / ML capabilities). The gNB may, for example, configure the WTRU with a buffer size and / or indicate the number of past samples of beam quality measurement / beam combining weight pairs. The beam quality measurement / beam combining weight pairs may be used to predict the beam (e.g., in the next transmit opportunity). For example, the buffer size used (e.g., window length N) stores the historical beam scan values ​​{(y0, w0), …, (y N-1 、w N-1 )} to predict the next weight vector, for example (y T ,w T ) may depend on performance (e.g., accuracy) requirements, number of beams, etc. The buffer size indication may be accompanied by one or more other metrics. The metrics may include, for example, one or more of a counter / timer for AI / ML model training, a time unit for beam prediction, number of beams, etc.

[0162] The gNB may configure one or more thresholds (e.g., a set of thresholds) for the WTRU, such as the quality of the beams stored in the buffer. The thresholds may be based on one or more of SNR / ACK / NACK statistics (e.g., measured at the WTRU), an amount of time that has passed (e.g., since a counter started), a beam estimation / prediction error, and / or the number of RSs configured. The beam estimation / prediction error may be the normalized mean square error (NMSE) between other (e.g., conventional) beam scanning-based outputs and / or outputs of a beam predictor (e.g., using the techniques described herein). For example, the number of RSs configured may be greater than the buffer size if the RS / beam scanning measurements do not meet the quality threshold (e.g., additional beams may be used or needed to fill the buffer).

[0163] The WTRU may be configured with an AI / ML model, for example, for beam prediction based on the number of beams. For example, X beams may be associated with an AI / ML model with dimension X. The gNB may indicate the number (e.g., total number) of beams that can be predicted (e.g., including other users with beam possibilities). The gNB may configure the WTRU to predict and / or report X beams. The gNB may identify any X c beams (eg, from among X beams), for example, depending on multi-user interference / requirements.

[0164] The WTRU may receive and / or measure one or more RS resource sets, for example, for beam prediction performance measurements (e.g., model training). The RS resource set may be selected from one or more of periodic, aperiodic, and / or semi-static resource sets. The gNB may send a configuration to the WTRU. The WTRU may be configured with a set of RS resource sets (e.g., via one or more of RRC, MAC CE, and / or DCI). The WTRU may receive / request one or more RS resource sets from the set of RS resource sets, for example, based on the configuration. The WTRU may request one or more RS resource sets from the gNB, for example, based on quality measurements and / or one or more of UL signals and / or channels. For example, if the measured quality is lower than (e.g., or equal to) a threshold value (e.g., block error rate (BLER) (or BER), SINR (or SNR), etc.) for requesting one or more RS resource sets, the WTRU may request one or more RS resource sets from the gNB. The threshold value may be configured / indicated by the gNB (e.g., via one or more of RRC, MAC CE, and DCI) and / or determined by the WTRU. For example, if the WTRU is configured with dedicated UL resources for requesting one or more RS resource sets (e.g., PUCCH, PUSCH, PRACH, SRS), the WTRU may request one or more RS resource sets from the gNB.

[0165] The WTR may send one or more reports to the gNB. The report may include an indication that the stored training data samples have reached at least one threshold. For example, if one or more conditions are met (e.g., based on reports from one or more WTRUs), the gNB may send one or more RS resource sets to the WTRU. The WTRU may measure one or more RS resource sets. One or more conditions may be predefined and / or indicated by the gNB. The one or more WTRU reports may be based on one or more of ACK / NACK, CSI reports, and / or BLER. The WTRU may receive a transmission of one or more RS resource sets. For example, after the WTRU reports +N (e.g., in units of us, ms, symbols, or time slots), the WTRU may receive one or more RS resource sets in the nearest resource of one or more RS resource sets. The threshold may be configured / indicated by the gNB (e.g., via one or more of RRC, MAC CE, and / or DCI) and / or may be determined by the WTRU.

[0166] The WTRU may perform AI / ML model training and / or inference, e.g., for predictive beam refinement / establishment. The AI / ML model may be trained (e.g., using training data). For example, samples of L1-RSRP (e.g., historical samples) and / or a combined weight vector (e.g., {(y0, w0), ... (y T-1 ,w T-1 )} can be used to predict the next weight vector (y T ,w T ). For example, during training, the error measure may include Criteria for populating training samples may be provided, for example where exhaustive beam scanning may be used to identify weights for training.

[0167] Inferences may be calculated / determined. For example, depending on the observed L1-RSRP value (e.g., despite online training), online training (e.g., continuous training of the AI / ML model) may be utilized. For example, if there are changes in the propagation environment (e.g., rapid changes) and / or the performance of the AI / ML is insufficient (e.g., below a threshold), a mechanism (e.g., an additional mechanism) may be used to infer the performance of the beam prediction. A trigger for exhaustive beam scanning may be implemented. The WTRU may calculate the prediction accuracy of the AI / ML model, for example, based on the measurement quality of different beams. The WTRU may create a beam prediction reference for the AI / ML model, for example, based on the measurement quality of different beams. The WTRU may calculate the prediction accuracy, for example, by comparing the output of the AI / ML model (e.g., a predicted beam) using the beam prediction reference. The WTRU may additionally or alternatively be configured to update the beam prediction reference periodically, semi-persistently, and / or via an indication (e.g., via a DCI). The WTRU may update the beam prediction reference based on the RS configuration (eg, based on the CSI-RS configured in CSI-ResourceConfig).

[0168] For example, once buffer storage is performed, the AI / ML model can be trained (e.g., online or offline). The WTRU may signal the gNB to notify the buffer storage performance / model training. The gNB may ease the beam scanning process through predictive beam refinement / model building. The indication sent by the WTRU may be explicit (e.g., through UCI / PUCCH / PDSCH, etc.) or implicit (e.g., by selecting specific UL resources (PUSCH / PUCCH / SRS, etc.)).

[0169] Receiver beamforming may be signaled (e.g., explicitly signaled) by the WTRU to the gNB. For example, if the predicted refined / narrow beam is not in the first measurement beam (e.g., or reference signal RS) and / or a subset of the first measurement beam / RS, the WTRU may signal receiver beamforming. The WTRU may report one or more of a predicted beam ID, a codebook ID (e.g., PMI), and / or a predicted L1-RSRP to the gNB. The WTRU may, for example, report the predicted beam arrival angle (e.g., RSRP) to the gNB.

[0170] CSI-RS transmission may be performed, for example, for quality checking. CSI-RS transmission may be performed periodically, dynamically, and / or based on a threshold (e.g., RSRP or average RSRP is less than or equal to a threshold). The network (e.g., gNB) and / or the WTRU may determine if there is a problem with the channel and / or AI / ML model performance. For example, if the measurement (e.g., RSRP and / or average RSRP) is below a threshold, the network (e.g., gNB) and / or the WTRU may determine if there is a problem with the channel and / or AI / ML model performance.

[0171] The CSI-RS test samples may be used to determine if there are problems with the AI / ML and / or the channel (e.g., a regular channel). The CSI-RS test samples used (e.g., used to determine if there are problems with the AI / ML and / or the channel) may be based on the reported quality. The direction in which the CSI-RS is sent may be based on the reported measured quality. The WTRU may, for example, periodically request CSI-RS transmission to check prediction accuracy. The WTRU may semi-persistently request CSI-RS transmission. For example, if the BER / SNR is less than a threshold, the WTRU may non-periodically request CSI-RS transmission.

[0172] The gNB may transmit a trigger for the WTRU to retrain / perform online training of the beam prediction model, for example if one or more of the following occurs: the WTRU prediction error is above a preconfigured threshold; a counter / timer expires; and / or the gNB determines that the WTRU will retrain / perform online training.

[0173] The gNB may transmit a trigger for the WTRU to retrain / perform online training, for example, if the WTRU prediction error is above a preconfigured threshold. The WTRU may trigger retraining / online training of the beam prediction model, for example, if the error / difference (e.g., expressed herein in terms of NMSE and cosine similarity calculations) exceeds a preconfigured threshold. For example, if the NMSE is greater than a threshold T1 and / or if the cosine is less than a threshold T2, the WTRU may trigger training / retraining of the AI / ML model. The gNB may additionally or alternatively have one or more thresholds corresponding to performance metrics of the WTRU with AI / ML capabilities. For example, if the WTRU reports the calculated NMSE and / or cosine to the gNB, and the error performance metric is not within the corresponding threshold at the gNB, the gNB may transmit an indication to the WTRU (e.g., in a DCI) to trigger training / retraining of the CSI prediction model.

[0174] The gNB may transmit a trigger for the WTRU to retrain / perform online training, for example if a timer / counter expires. For example, each time the WTRU completes training of an AI / ML model, the WTRU may start a counter / timer. Upon expiration of the counter / timer, the WTRU may retrain the model. The length of time set by the timer may be measured / recorded in any of the following units: time slot, symbol duration, SFN, and / or seconds / milliseconds. The counter may be measured in one or more symbols. The length of the timer / counter may be pre-configured by the WTRU and / or indicated to the gNB. The length of the timer / counter may be pre-configured by the gNB and / or indicated to the WTRU.

[0175] The gNB may transmit a trigger for the WTRU to retrain / perform online training, for example if the gNB determines that the WTRU is to retrain / perform online training of the AI / ML model. The gNB may determine that the beam selected by the ML model at the WTRU is not the best beam and / or trigger training / retraining of the ML model at the WTRU. The gNB may make this determination based on low registration throughput (e.g., in the uplink). Additionally or alternatively, the gNB may make this determination through, for example, a CSI report transmitted by the WTRU.

[0176] For example, when the SINR is above a threshold, AI / ML-based predictive beam establishment / refinement may mitigate or degrade the performance of CSI-RSI / beam scanning. The performance of CSI-RSI / beam scanning may be below a threshold, which may, for example, indicate that there is a problem with the AI / ML model (e.g., DNN) and / or channel conditions. The WTRU and / or gNB may trigger AI / ML model retraining, for example, if the AI / ML model performs below a threshold. The WTRU and / or gNB may not perform predictive beam searching (e.g., the predictive beam searching techniques described herein), for example, if channel conditions are causing poor performance.

[0177] The gNB may perform predictive beam refinement. The gNB may use the (e.g., historical) RSRP / beam index values ​​from the WTRU to predict the beamsetter of the WTRU. The gNB may signal the capability of predicting the beamsetter, for example, as an index (e.g., corresponding to a specific codebook and / or lookup table of transmit beams mapped to receive beams).

[0178] For example, if there is no beam correspondence, and / or if there is a suitable Tx or Rx beam (e.g., above a threshold), the gNB may predict the WTRU side beam index. The gNB may request and / or receive additional input from the WTRU to assist in the prediction.

[0179] The gNB may transmit an indication to the WTRU. The WTRU may not perform beam scanning and / or may reduce the beam scanning space, for example if the gNB is able to predict the receiver beam at the WTRU. The gNB may transmit a beam index to the WTRU. The index may indicate the Tx beam and / or the Rx beam to the WTRU.

[0180] An absolute beam indication and / or a relative beam indication may be sent. The WTRU may select and / or report to the gNB an Rx / Tx beam index as its selected best Rx / Tx beam, for example, as part of a beam selection process (e.g., a corresponding CSI report). The report may include one or more of an absolute beam indication and / or a relative beam indication. For example, if the report includes a relative beam indication, the WTRU may determine the relative beam index based on the Rx / Tx beam. For example, if the report includes a relative beam indication, the relative beam index may be determined based on a priority / threshold (e.g., RSRP, LOS, etc.). For example, if the report includes a relative beam indication, beam index #1 may be assigned to the beam with the highest RSRP, beam index #2 may be assigned to the beam with the second highest RSRP, and so on.

[0181] The WTRU may determine a new / updated (e.g., best) Rx / Tx beam, e.g., due to an environmental change or WTRU movement / rotation. The WTRU may determine an operating mode for reporting an updated beam index to the gNB. In a first operating mode, the WTRU may determine not to report an updated beam index if, for example, the new Rx / Tx beam has a similar (e.g., the same) spatial relationship to a previously reported beam index (e.g., a beam index reported due to an environmental change, a WTRU rotation, etc.). In a second operating mode, the WTRU may determine to report an updated beam index if, for example, the new Rx / Tx beam has a different spatial relationship to a previously reported beam index (e.g., a beam index reported due to an environmental change, a WTRU rotation, etc.). The WTRU may report the updated Rx / Tx to the gNB. The WTRU may report the beam index if, for example, the report includes a relative beam.

[0182] The WTRU may be configured to measure and / or monitor a prediction error associated with the AI / ML model. The prediction error may include a difference between a predicted beam applicable at time T (e.g., determined based on the AI / ML model output) and the (e.g., actual) best beam at time T (e.g., determined based on RS measurements). The prediction error may include instances where the predicted beam applicable at time T (e.g., determined based on the AI / ML model output) is not within the top N beams at time T (e.g., determined based on RS measurements).

[0183] The prediction error may include a difference between the RSRP / SNR / SINR associated with a predicted beam applicable at time T (e.g., determined based on an AI / ML model output) and the (e.g., actual) RSRP / SNR / SINR associated with the predicted beam (e.g., determined based on RS measurements) at time T. For example, the prediction error may be determined as an average over a period of time.

[0184] The WTRU may be configured to send a report. The report may include, for example, an indication of beam prediction performance based on a pre-configured trigger. The report may indicate a prediction error of the AI / ML model. For example, if the prediction error is below a threshold (e.g., a pre-configured threshold), the WTRU may send a report. For example, if the prediction error is above a threshold (e.g., a pre-configured threshold), the WTRU may send a report. The WTRU may send reports periodically (e.g., every n milliseconds or time slots).

[0185] The WTRU may be configured to send a report including an indication. The indication may be associated with AI / ML model training. The report may indicate the status of the AI / ML model training. For example, the report may indicate whether the AI / ML model training is completed or in progress (e.g., where the criteria for completing the AI / ML model training may be preconfigured). For example, if the prediction error associated with the AI / ML is greater than a threshold (e.g., a preconfigured threshold), the WTRU may send a report. For example, the WTRU may send a report every n epochs. The value of n may be configured / preconfigured by the network. The epoch may refer to (e.g., one) complete pass of the training data set through the learning algorithm. The WTRU may send a report when the parameters of the model are updated, for example, after the learning algorithm has processed a number of training data samples equal to a given (e.g., preconfigured) batch size. The WTRU may send reports periodically (e.g., every n milliseconds or time slots) while training is ongoing.

[0186] The WTRU may be configured (e.g., pre-configured) with an AI / ML model performance reporting configuration. The reporting configuration may include UL reporting resources. The UL reporting resources may include one or more of periodicity, time offset, content, report size, etc. The WTRU may determine that the configuration is deactivated (e.g., by default). For example, when the WTRU starts training an AI / ML model, the WTRU may be configured to activate the AI / ML model performance reporting configuration. The WTRU may deactivate the AI / ML model performance reporting (e.g., when the AI / ML model training is completed). Additionally or alternatively, the WTRU may be configured with one or more (e.g., two) reporting configurations. When AI / ML model training is in progress, the first reporting configuration may be applicable / active. When the AI / ML model is used for inference, the second reporting configuration may be applicable / active.

[0187] Predictive beam refinement / establishment based on AI / ML may be performed, for example, using WTRU inference. The WTRU may save historical samples of AI / ML inputs to the WTRU using, for example, beam scanning techniques based on CSI-RS transmissions. A beam (e.g., the best beam) for predictive beam operation (e.g., TxOP'T') may be identified based on observed RSRP (e.g., the highest observed RSRP value). The WTRU may continue to identify beams, for example, until the buffer is filled. The WTRU may indicate to the gNB that the buffer is full and / or that the WTRU is ready for data-driven beam refinement. The number of CSI-RS transmissions may be greater than the buffer size. For example, if the beam prediction identifies a beam that is suitable for transmission, the buffer may not be filled based on capacity. The size of the buffer size may be determined based on RSRP requirements. A validation (e.g., monitoring) process may be performed. The CSI-RS may be sent during the validation process, for example, to determine whether the accuracy of the model (e.g., based on AI / ML) is sufficient (e.g., based on the difference between the RSRP observed using the CSI-RS and the RSRP predicted using the model). The WTRU may perform beam refinement with fewer or zero (e.g., no) CSI-RS transmissions. For example, if the predicted refined / narrow beam is not in the first measurement beam (e.g., or RS) and / or a subset of the first measurement beam / RS, the receiver beamforming may be signaled by the WTRU (e.g., explicitly) to the gNB. The WTRU may report one or more of the predicted beam ID, codebook ID (e.g., PMI), and / or predicted L1-RSRP to the gNB. The WTRU may, for example, report the predicted beam arrival angle (e.g., AoA) to the gNB. For example, if the observed RSRP is below a threshold, the WTRU may trigger the gNB to retrain the model (e.g., at the WTRU). The RSRP may be below the threshold, but this may not be due to AI / ML. Techniques may be implemented to determine whether the RSRP is below the threshold due to AI / ML or based on other circumstances (e.g., poor channel conditions). One or more messages / metrics may be exchanged between the WTRU and the gNB. For example, beam recovery may be performed before retraining the AI / ML model. Partial buffer updates and / or CSI-RS retransmission updates (e.g., periodic configuration, etc.) may be performed. The gNB may send CSI-RS for retraining and buffer filling, for example, based on network sent parameters.

[0188] For example, AI / ML based predictive beam refinement / establishment may be performed using gNB inference. The gNB may fill the buffer with (e.g., a specific) beam, for example, based on the WTRU report of RSRP values ​​for each Tx beam. In order to identify the beam (e.g., the best beam), the WTRU may utilize beam scanning techniques. The WTRU may send the observed RSRP value using, for example, PUCCH / PUSCH transmissions / resources. The WTRU may collect RSRP values, for example, for each Tx beam. The WTRU may transmit the observed RSRP value to the gNB (e.g., which may reduce the amount of overhead). For example, using the received RSRP value for each Tx beam, the gNB may fill the buffer. The gNB may transmit a buffer fill indication to the WTRU. The buffer fill indication may alternatively or additionally indicate that the gNB is ready for data transmission. For example, if the observed RSRP is below a threshold, the WTRU may trigger the gNB to retrain the AI / ML (e.g., on the gNB side). The gNB may send, for example, a CSI-RS for retraining. Additionally or alternatively, the gNB may perform buffer filling, for example based on network signaled parameters.

[0189] The gNB may predict the WTRU side beam index. The WTRU side beam index may be based on further input from the WTRU. The gNB prediction may be inaccurate, for example, if the WTRU experiences rotation and / or movement (e.g., severe rotation and fast movement). The WTRU may indicate whether it requests (e.g., requires) a WTRU side prediction by the gNB, for example based on an RSRP value. The gNB may indicate the predicted Rx beam and / or the predicted Tx to the WTRU. The WTRU may determine how to beam and / or on which beam to beam, for example based on the predicted beam. The gNB may indicate the predicted beam using an absolute beam indication and / or a relative beam indication.

[0190] The observed RSRP may be below a threshold. If the RSRP drops below a threshold (e.g., and / or if data collection for AI / ML fine-tuning) the gNB and / or WTRU may trigger a retraining. The WTRU may configure CSI_RS_test samples. For example, if a retraining for an AI / ML model is performed (e.g., based on RSRP dropping below a threshold), the WTRU may configure CSI_RS_test samples. The CSI_RS_test samples may be used to identify which side of the AI / ML is to be retrained.

Claims

1. A wireless transmit / receive unit (WTRU), the wireless transmit / receive unit (WTRU) comprising a processor, the processor configured to: receiving a configuration for predictive beam refinement, wherein the configuration includes at least one threshold associated with information to be used for predicting and selecting one or more beam pairs, wherein the at least one threshold includes a quality threshold or a threshold amount of memory containing stored training data samples; performing measurements on one or more reference signal (RS) sets; determining to store the one or more of the measurements as training data samples in a memory based on a comparison of the one or more of the measurements with the at least one threshold; determining, based on the at least one threshold, to begin training or retraining a model for performing predictions associated with the one or more beam pairs; as well as An indication is sent that the stored training data sample has reached the at least one threshold.

2. The WTRU of claim 1 , wherein the processor is configured to: predicting measurements for the one or more beam pairs based on the stored training data samples and the predictive beam refinement configuration; and One or more of a predicted beam identifier, a codebook identifier, a predicted layer 1 received signal received power (L1-RSRP) value, or a predicted angle of arrival (AoA) are reported based on the predicted measurements.

3. The WTRU of claim 1 or 2, wherein the at least one threshold comprises at least one of: a buffer size threshold associated with the memory for indicating a maximum number of samples of beam quality measurements or beam combining weight pairs, thresholds related to the maximum buffer size or the remaining buffer size, a threshold value associated with one or more measurements to be performed on one or more sets of RSs, A threshold number of comparisons of one or more measurements to one or more thresholds, or one or more thresholds used to determine the storage of samples in said memory.

4. The WTRU of any one of claims 1 to 3, wherein the determination to start training or retraining the model is also at least partially comprised at a network entity.

5. The WTRU of any one of claims 1 to 4, wherein the processor is configured to: Based on receiving the configuration, stored training data samples are retrieved from the memory.

6. The WTRU of any one of claims 1 to 5, wherein the processor is configured to: When the one or more of the measurements are greater than or equal to the at least one threshold, the one or more measurements are stored in the memory as training data samples.

7. The WTRU of any one of claims 1 to 5, wherein the processor is configured to: When the one or more of the measurements are less than the at least one threshold, repeating the determining to store the one or more measurements as training data samples in the memory.

8. The WTRU of any one of claims 1 to 7, wherein the processor is configured to: A prediction accuracy of measurements for the one or more beam pairs is calculated based on the stored training data samples.

9. The WTRU of any one of claims 1 to 8, wherein the at least one threshold comprises a value that increases with the stored training data samples.

10. The WTRU of any one of claims 1 to 9, wherein the processor is configured to predict measurements for the one or more beam pairs based on a weight vector.

11. A method performed by a wireless transmit / receive unit (WTRU), the method comprising: receiving a configuration for predictive beam refinement, wherein the configuration includes at least one threshold associated with information to be used for predicting and selecting one or more beam pairs, wherein the at least one threshold includes a quality threshold or a threshold amount of memory containing stored training data samples; performing measurements on one or more reference signal (RS) sets; determining to store the one or more of the measurements as training data samples in a memory based on a comparison of the one or more of the measurements with the at least one threshold; determining, based on the at least one threshold, to begin training or retraining a model for performing predictions associated with the one or more beam pairs; as well as An indication is sent that the stored training data sample has reached the at least one threshold.

12. The method according to claim 11, comprising: predicting measurements for the one or more beam pairs based on the stored training data samples and the predictive beam refinement configuration; as well as One or more of a predicted beam identifier, a codebook identifier, a predicted layer 1 received signal received power (L1-RSRP) value, or a predicted angle of arrival (AoA) are reported based on the predicted measurements.

13. The method according to claim 11 or 12, wherein the at least one threshold comprises at least one of the following: a buffer size threshold associated with the memory for indicating a maximum number of samples of beam quality measurements or beam combining weight pairs, thresholds related to the maximum buffer size or the remaining buffer size, a threshold value associated with one or more measurements to be performed on one or more sets of RSs, a threshold number of comparisons of one or more measurements to one or more threshold values, or One or more thresholds for determining to store samples in the memory.

14. The method of any one of claims 11 to 13, wherein the determination to begin training or retraining the model is also at least partially comprised at a network entity.

15. The method according to any one of claims 11 to 14, comprising: Based on receiving the configuration, stored training data samples are retrieved from the memory.

16. The method according to any one of claims 11 to 15, comprising: When the one or more of the measurements are greater than or equal to the at least one threshold, the one or more measurements are stored in the memory as training data samples.

17. The method according to any one of claims 11 to 15, comprising: When the one or more of the measurements are less than the at least one threshold, repeating the determining to store the one or more measurements as training data samples in the memory.

18. The method according to any one of claims 11 to 17, comprising: A prediction accuracy of measurements for the one or more beam pairs is calculated based on the stored training data samples.

19. The method of any one of claims 11 to 18, wherein the at least one threshold comprises a value that increases with the stored training data samples.

20. The method according to any one of claims 11 to 19, comprising: Measurements for the one or more beam pairs are predicted based on the weight vectors.

Images