Methods and apparatuses for channel state information compression using a latent representation with multiple parts associated to different time intervals

Abstract

A wireless transmit / receive unit (WTRU) may receive configuration information indicating one or more parameters associated with channel state information (CSI) feedback of time-dependent multi-part (TDMP) latents. The one or more parameters include a set of TDMP feedback transmission modes including a single-part TDMP feedback mode and a multi-part TDMP feedback mode. The WTRU may receive one or more transmissions of one or more CSI reference signals (CSI-RSs). The WTRU may generate, using an artificial intelligence / machine learning (AI / ML) model, a set of TDMP latents based on measurement information associated with the received one or more CSI-RSs. The set of TDMP latents includes a plurality of parts. The WTRU may determine a TDMP feedback transmission mode from the set of TDMP feedback transmission modes. The WTRU may send reporting information indicating at least a part of the set of TDMP latents and the determined TDMP feedback transmission mode.

Description

METHODS AND APPARATUSES FOR CHANNEL STATE INFORMATION COMPRESSION USING A LATENT REPRESENTATION WITH MULTIPLE PARTS ASSOCIATED TO DIFFERENT TIME INTERVALSCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Patent Application No. 63 / 546,324 filed 30-0ct-2023, which is incorporated herein by reference.BACKGROUND

[0002] The present application is related to the fields of communications, software and encoding, including, for example, to methods, architectures, apparatuses, systems directed to reductions of channel state information (CSI) reporting overhead using two-sided artificial intelligence and / or machine learning (AI / ML) models that leverage CSI temporal correlation properties.

[0003] Existing work on reducing CSI overhead relies on spatial-frequency (SF) CSI compression. While current SF CSI compression may provide acceptable reconstruction quality, there remains the opportunity for further improvements which can approach the upper bounds of performance, namely of uncompressed CSI.BRIEF SUMMARY

[0004] Briefly stated, in one embodiment, a wireless transmit / receive unit (WTRU) may operate as an encode-side device and another device (e.g., WTRU or base station) may operate as a decodeside device, or vice versa. In an example embodiment, a WTRU may perform a procedure which includes to receive configuration information associated with reference signal (RS) feedback (e.g., channel state information (CSI) RS (CSI-RS) reporting). The WTRU may measure, at a plurality of times, one or more RSs. The WTRU may generate a plurality of latent structures using an artificial intelligence / machine learning (AI / ML) model that encodes feedback information associated with the measured RSs based on the configuration information. Each latent structure may (e.g., respectively) include a common part and a specific part. The WTRU may send a plurality of transmissions which include the plurality of latent structures. Each transmission may respectively include any of the common part and / or the specific part of one of the latent structures based on one or more conditions.

[0005] In one embodiment, WTRU may receive configuration information indicating one or more parameters associated with CSI feedback of time-dependent multi-part (TDMP) latents. Theone or more parameters may include a set of TDMP feedback transmission modes including a single-part TDMP feedback mode and a multi-part TDMP feedback mode. The WTRU may receive one or more transmissions of one or more CSI-RSs. The WTRU may generate, using an AI / ML model, a set of TDMP latents based on measurement information associated with the received one or more CSI-RSs. The set of TDMP latents may include a plurality of parts. The WTRU may determine a TDMP feedback transmission mode from the set of TDMP feedback transmission modes. The WTRU may send reporting information indicating (i) at least a part of the set of TDMP latents and (ii) the determined TDMP feedback transmission mode.

[0006] In one embodiment, a base station may send, to a WTRU, configuration information indicating one or more parameters associated with CSI feedback of TDMP latents. The one or more parameters may include a set of TDMP feedback transmission modes including a single-part TDMP feedback mode and a multi-part TDMP feedback mode. The base station may send, to the WTRU, one or more transmissions of one or more CSI-RSs. The base station may receive, from the WTRU, reporting information indicating (i) at least a part of a set of TDMP latents and (ii) a WTRU-determined TDMP feedback transmission mode. The set of TDMP latents may be based on measurement information associated with the one or more transmissions. The set of TDMP latents may include a plurality of parts.

[0007] In one embodiment, a WTRU may receive configuration information indicating one or more parameters associated with CSI feedback of TDMP latents. The WTRU may generate a set of TDMP latents based on measurement information associated with received signals. The set of TDMP latents may include a plurality of parts. The WTRU may determine a TDMP feedback transmission mode from the one or more parameters. The WTRU may send reporting information indicating (i) at least a part of the set of TDMP latents and / or (ii) the determined TDMP feedback transmission mode.

[0008] In one embodiment, a base station may send configuration information indicating one or more parameters associated with CSI feedback of TDMP latents. The base station may receive reporting information indicating (i) at least a part of a set of TDMP latents and / or (ii) a WTRU- determined TDMP feedback transmission mode (e.g., according to the configured parameters). The set of TDMP latents may be based on measurement information associated with the one or more transmissions. The set of TDMP latents may include a plurality of parts.BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The following detailed description will be better understood when read in conjunction with the appended drawings, in which there are shown examples of one or more of the multipleembodiments of the present disclosure. It should be understood, however, that the embodiments described herein are not limited to the precise arrangements and instrumentalities shown in the drawings. In the drawings:

[0010] FIG. 1 A is a system diagram illustrating an example communications system, according to one or more embodiments of the present disclosure;

[0011] FIG. IB is a system diagram illustrating an example wireless transmit / receive unit (WTRU) that may be used within the communications system illustrated in FIG. 1 A, according to one or more embodiments of the present disclosure;

[0012] FIG. 1C is a system diagram illustrating an example radio access network (RAN) and an example core network (CN) that may be used within the communications system illustrated in FIG. 1 A, according to one or more embodiments of the present disclosure;

[0013] FIG. ID is a system diagram illustrating a further example RAN and a further example CN that may be used within the communications system illustrated in FIG. 1 A, according to one or more embodiments of the present disclosure;

[0014] FIG. 2 is a report and resource diagram illustrating an example configuration for CSI reporting settings, resource settings, and links, according to one or more embodiments of the present disclosure;

[0015] FIG. 3 is a processing diagram illustrating an example AI / ML autoencoder with a timedependent multipart (TDMP) latent domain that includes a common part and a specific part, according to one or more embodiments of the present disclosure;

[0016] FIG. 4 is a timing diagram illustrating an example timing of uplink (UL) CSI reporting of common and specific TSF parts, according to one or more embodiments of the present disclosure;

[0017] FIG. 5 is a spatial-frequency (SF) compression diagram illustrating an example of a two- sided autoencoder model, according to one or more embodiments of the present disclosure;

[0018] FIG. 6 is an encoding diagram illustrating an example of training of an autoencoder (AE) model, according to one or more embodiments of the present disclosure;

[0019] FIG. 7 is an encoding diagram illustrating an example of blocking the short-term components of an AE model, according to one or more embodiments of the present disclosure;

[0020] FIG. 8 is an encoding diagram illustrating an example of training of an AE model with common and specific parts where only weights associated with the specific part are updated and the common part weights are frozen, according to one or more embodiments of the present disclosure;

[0021] FIG. 9 is another encoding diagram illustrating an example of training an AE model with common encoder / decoder modules, according to one or more embodiments of the present disclosure;

[0022] FIG. 10 is another encoding diagram illustrating an example of training an AE model with residual encoder / decoder modules, according to one or more embodiments of the present disclosure;

[0023] FIG. 11 is a flow diagram illustrating an example procedure for time-dependent multipart (TDMP) feedback transmission, according to one or more embodiments of the present disclosure;

[0024] FIG. 12 is a flow diagram illustrating another example procedure for TDMP feedback transmission, according to one or more embodiments of the present disclosure;

[0025] FIG. 13 is a flow diagram illustrating another example procedure for controlling the validity of a common part used in TDMP feedback transmission, according to one or more embodiments of the present disclosure;

[0026] FIG. 14 is a flow diagram illustrating another example procedure for TDMP feedback transmission, according to one or more embodiments of the present disclosure;

[0027] FIG. 15 is a flow diagram illustrating an example procedure for TDMP feedback reception, according to one or more embodiments of the present disclosure;

[0028] FIG. 16 is a flow diagram illustrating another example procedure for TDMP feedback transmission, according to one or more embodiments of the present disclosure; and

[0029] FIG. 17 is a flow diagram illustrating an example procedure for TDMP feedback reception, according to one or more embodiments of the present disclosure.DETAILED DESCRIPTION

[0030] In describing the various embodiments of the present disclosure, certain terminology is used herein for convenience only and should not be considered as limiting such embodiments. In the drawings, the same reference numerals are employed for designating the same elements throughout the several figures and the present description.

[0031] In the following detailed description, numerous specific details are set forth to provide a thorough understanding of embodiments and / or examples disclosed herein. However, it will be understood that such embodiments and examples may be practiced without some or all of the specific details set forth herein. In other instances, well-known methods, procedures, components and circuits have not been described in detail, so as not to obscure the following description. Further, embodiments and examples not specifically described herein may be practiced in lieu of, or in combination with, the embodiments and other examples described, disclosed or otherwiseprovided explicitly, implicitly and / or inherently (collectively "provided") herein. Although various embodiments are described and / or claimed herein in which an apparatus, system, device, etc. and / or any element thereof carries out an operation, process, algorithm, function, etc. and / or any portion thereof, it is to be understood that any embodiments described and / or claimed herein assume that any apparatus, system, device, etc. and / or any element thereof is configured to carry out any operation, process, algorithm, function, etc. and / or any portion thereof.

[0032] Example Communications System

[0033] The methods, apparatuses and systems provided herein are well-suited for communications involving both wired and wireless networks. An overview of various types of wireless devices and infrastructure is provided with respect to FIGs. 1A-1D, where various elements of the network may utilize, perform, be arranged in accordance with and / or be adapted and / or configured for the methods, apparatuses and systems provided herein.

[0034] FIG. 1A is a system diagram illustrating an example communications system 100 in which one or more disclosed embodiments may be implemented. The communications system 100 may be a multiple access system that provides content, such as voice, data, video, messaging, broadcast, etc., to multiple wireless users. The communications system 100 may enable multiple wireless users to access such content through the sharing of system resources, including wireless bandwidth. For example, the communications systems 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), singlecarrier FDMA (SC-FDMA), zero-tail (ZT) unique-word (UW) discreet Fourier transform (DFT) spread OFDM (ZT UW DTS-s OFDM), unique word OFDM (UW-OFDM), resource block- filtered OFDM, filter bank multicarrier (FBMC), and the like.

[0035] As shown in FIG. 1A, the communications system 100 may include wireless transmit / receive units (WTRUs) 102a, 102b, 102c, 102d, a radio access network (RAN) 104 / 113, a core network (CN) 106 / 115, a public switched telephone network (PSTN) 108, the Internet 110, and other networks 112, though it will be appreciated 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 a "STA", may be configured to transmit and / or receive wireless signals and may include (or be) a user equipment (UE), a mobile station, a fixed or mobile subscriber unit, a subscription-based unit, a pager, a cellular telephone, a personal digital assistant (PDA), a smartphone, a laptop, a netbook, a personal computer, a wireless sensor, a hotspot or Mi-Fi device, an Internet of Things (loT) device, a watch or other wearable, a head-mounted display (HMD), a vehicle, a drone, a medical device and applications (e.g., remote surgery), an industrial device and applications (e.g., a robot and / or other wireless devices operating in an industrial and / or an automated processing chain contexts), a consumer electronics device, a device operating on commercial and / or industrial wireless networks, and the like. Any of the WTRUs 102a, 102b, 102c and 102d may be interchangeably referred to as a UE.

[0036] The communications systems 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 configured to wirelessly interface with at least one of the WTRUs 102a, 102b, 102c, 102d, e.g., to facilitate access to one or more communication networks, such as the CN 106 / 115, the Internet 110, and / or the networks 112. By way of example, the base stations 114a, 114b may be any of a base transceiver station (BTS), a Node-B (NB), an eNode-B (eNB), a Home Node-B (HNB), a Home eNode-B (HeNB), a gNode-B (gNB), a NR Node-B (NR NB), a site controller, an access point (AP), a wireless router, and the like. While the base stations 114a, 114b are each depicted as a single element, it will be appreciated that the base stations 114a, 114b may include any number of interconnected base stations and / or network elements.

[0037] 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), relay nodes, etc. The base station 114a and / or the base station 114b may be configured to transmit and / or receive wireless signals on one or more carrier frequencies, which may be referred to as a cell (not shown). These frequencies may be in licensed spectrum, unlicensed spectrum, or a combination of licensed and unlicensed spectrum. A cell may provide coverage for a wireless service to a specific geographical area that may be relatively fixed or that may change over time. The cell may further be divided into cell sectors. For example, the cell associated with the base station 114a may be divided into three sectors. Thus, in an embodiment, the base station 114a may include three transceivers, i.e., one for each sector of the cell. In an embodiment, the base station 114a may employ multiple-input multiple output (MIMO) technology and may utilize multiple transceivers for each or any sector of the cell. For example, beamforming may be used to transmit and / or receive signals in desired spatial directions.

[0038] 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).

[0039] More specifically, as noted above, the communications 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, and the like. 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 establish the air interface 116 using wideband CDMA (WCDMA). WCDMA may include communication protocols such as High-Speed Packet Access (HSPA) and / or Evolved HSPA (HSPA+). HSPA may include High-Speed Downlink Packet Access (HSDPA) and / or High-Speed Uplink Packet Access (HSUPA).

[0040] In an 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), which may establish the air interface 116 using Long Term Evolution (LTE) and / or LTE- Advanced (LTE-A) and / or LTE-Advanced Pro (LTE-A Pro).

[0041] In an 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).

[0042] In an 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 implement LTE radio access and NR radio access together, for instance using dual connectivity (DC) principles. Thus, the air interface utilized by WTRUs 102a, 102b, 102c may be characterized by multiple types of radio access technologies and / or transmissions sent to / from multiple types of base stations (e.g., an eNB and a gNB).

[0043] In an embodiment, the base station 114a and the WTRUs 102a, 102b, 102c may implement radio technologies such as IEEE 802.11 (i.e., Wireless Fidelity (Wi-Fi), IEEE 802.16 (i.e., Worldwide Interoperability for Microwave Access (WiMAX)), CDMA2000, CDMA2000 IX, 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), and the like.

[0044] The base station 114b in FIG. 1 A may be a wireless router, Home Node-B, Home eNode- B, or access point, for example, and may utilize any suitable RAT for facilitating wireless connectivity in a localized area, such as a place of business, a home, a vehicle, a campus, an industrial facility, an air corridor (e.g., for use by drones), a roadway, and the like. In an 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 an embodiment, thebase 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 an 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 any of a small cell, picocell or femtocell. As shown in FIG. 1 A, the base station 114b may have a direct connection to the Internet 110. Thus, the base station 114b may not be required to access the Internet 110 via the CN 106 / 115.

[0045] 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 varying quality of service (QoS) requirements, such as differing throughput requirements, latency requirements, error tolerance requirements, reliability requirements, data throughput requirements, mobility requirements, and the like. The CN 106 / 115 may provide call control, billing services, mobile location-based services, pre-paid calling, Internet connectivity, video distribution, etc., and / or perform high-level security functions, such as user authentication. Although not shown in FIG. 1 A, 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 be utilizing an NR radio technology, the CN 106 / 115 may also be in communication with another RAN (not shown) employing any of a GSM, UMTS, CDMA 2000, WiMAX, E-UTRA, or Wi-Fi radio technology.

[0046] The CN 106 / 115 may also serve 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 circuit-switched telephone networks that provide 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), user datagram protocol (UDP) and / or the internet protocol (IP) in the TCP / IP internet protocol suite. The networks 112 may include wired and / or wireless communications 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 / 114 or a different RAT.

[0047] Some or all of the WTRUs 102a, 102b, 102c, 102d in the communications 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 over different wireless links). For example, the WTRU 102c shown in FIG. 1A may be configured to communicate withthe 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.

[0048] FIG. IB is a system diagram illustrating an example WTRU 102. As shown in FIG. IB, the WTRU 102 may include 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 elements / peripherals 138, among others. It will be appreciated that the WTRU 102 may include any sub-combination of the foregoing elements while remaining consistent with an embodiment.

[0049] 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 in association with a DSP core, a controller, a microcontroller, Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) circuits, any other type of integrated circuit (IC), a state machine, and the like. The processor 118 may perform signal coding, 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. While FIG. IB depicts the processor 118 and the transceiver 120 as separate components, it will be appreciated that the processor 118 and the transceiver 120 may be integrated together, e.g., in an electronic package or chip.

[0050] The transmit / receive element 122 may be configured to transmit signals to, or receive signals from, a base station (e.g., the base station 114a) over the air interface 116. For example, in an embodiment, the transmit / receive element 122 may be an antenna configured to transmit and / or receive RF signals. In an embodiment, the transmit / receive element 122 may be an emitter / detector configured to transmit and / or receive IR, UV, or visible light signals, for example. In an embodiment, the transmit / receive element 122 may be configured to transmit and / or receive both RF and light signals. It will be appreciated that the transmit / receive element 122 may be configured to transmit and / or receive any combination of wireless signals.

[0051] Although the transmit / receive element 122 is depicted in FIG. IB as a single element, the WTRU 102 may include any number of transmit / receive elements 122. For example, the WTRU 102 may employ MIMO technology. Thus, in an 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.

[0052] The transceiver 120 may be configured to modulate the signals that are to be transmitted by the transmit / receive element 122 and to demodulate the signals that are received by the transmit / receive element 122. As noted above, the WTRU 102 may have multi-mode capabilities. Thus, the transceiver 120 may include multiple transceivers for enabling the WTRU 102 to communicate via multiple RATs, such as NR and IEEE 802.11, for example.

[0053] The processor 118 of the WTRU 102 may be coupled to, and may receive user input data from, the speaker / microphone 124, the keypad 126, and / or the display / touchpad 128 (e.g., a liquid crystal display (LCD) display unit or 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 the non-removable memory 130 and / or the removable memory 132. The non-removable memory 130 may include random-access memory (RAM), readonly 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, memory that is not physically located on the WTRU 102, such as on a server or a home computer (not shown).

[0054] The processor 118 may receive power from the power source 134, and may be configured to distribute and / or control the 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, and the like.

[0055] 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 over the air interface 116 from a base station (e.g., base stations 114a, 114b) and / or determine its location based on the timing of the signals being received from two or more nearby base stations. It will be appreciated that the WTRU 102 may acquire location information by way of any suitable location-determination method while remaining consistent with an embodiment.

[0056] The processor 118 may further be coupled to other elements / peripherals 138, which may include one or more software and / or hardware modules / units that provide additional features, functionality and / or wired or wireless connectivity. For example, the elements / peripherals 138 may include an accelerometer, an e-compass, a satellite transceiver, a digital camera (e.g., forphotographs and / or video), a universal serial bus (USB) port, a vibration device, a television transceiver, a hands free headset, a Bluetooth® module, a frequency modulated (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, and the like. The elements / peripherals 138 may include one or more sensors, the sensors may be one or more of 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.

[0057] The WTRU 102 may include a full duplex radio for which transmission and reception of some or all of the signals (e.g., associated with particular subframes for both the uplink (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 to reduce and or substantially eliminate self-interference via either hardware (e.g., a choke) or signal processing via a processor (e.g., a separate processor (not shown) or via processor 118). In an embodiment, the WTRU 102 may include a half-duplex radio for which transmission and reception of some or all of the signals (e.g., associated with particular subframes for either the uplink (e.g., for transmission) or the downlink (e.g., for reception)).

[0058] FIG. 1C is a system diagram illustrating the RAN 104 and the CN 106 according to an embodiment. As noted above, the RAN 104 may employ an E-UTRA radio technology to communicate with the WTRUs 102a, 102b, and 102c over the air interface 116. The RAN 104 may also be in communication with the CN 106.

[0059] The RAN 104 may include eNode-Bs 160a, 160b, 160c, though it will be appreciated that the RAN 104 may include any number of eNode-Bs while remaining consistent with an embodiment. The eNode-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 an embodiment, the eNode-Bs 160a, 160b, 160c may implement MIMO technology. Thus, the eNode-B 160a, for example, may use multiple antennas to transmit wireless signals to, and receive wireless signals from, the WTRU 102a.

[0060] Each of the eNode-Bs 160a, 160b, and 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 uplink (UL) and / or downlink (DL), and the like. As shown in FIG. 1C, the eNode-Bs 160a, 160b, 160c may communicate with one another over an X2 interface.

[0061] The CN 106 shown in FIG. 1C may include a mobility management entity (MME) 162, a serving gateway (SGW) 164, and a packet data network (PDN) gateway (PGW) 166. While each of the foregoing elements are depicted as part of the CN 106, it will be appreciated that any one of these elements may be owned and / or operated by an entity other than the CN operator.

[0062] The MME 162 may be connected to each of the eNode-Bs 160a, 160b, and 160c in the RAN 104 via an SI 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, and the like. 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.

[0063] The SGW 164 may be connected to each of the eNode-Bs 160a, 160b, 160c in the RAN 104 via the SI 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 user planes during inter-eNode-B handovers, triggering paging when DL data is available for the WTRUs 102a, 102b, 102c, managing and storing contexts of the WTRUs 102a, 102b, 102c, and the like.

[0064] 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.

[0065] 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 communicate with, an IP gateway (e.g., an IP multimedia subsystem (IMS) server) that serves 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 the other networks 112, which may include other wired and / or wireless networks that are owned and / or operated by other service providers.

[0066] Although the WTRU is described in FIGs. 1A-1D as a wireless terminal, it is contemplated that in certain representative embodiments that such a terminal may use (e.g., temporarily or permanently) wired communication interfaces with the communication network.

[0067] In representative embodiments, the other network 112 may be a WLAN.

[0068] 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 an accessor an interface to a distribution system (DS) or another type of wired / wireless network that carries traffic into and / or out of the BSS. Traffic to STAs that originates from outside the BSS may arrive through the AP and may be delivered to the STAs. Traffic originating from STAs to destinations outside the BSS may be sent to the AP to be delivered to respective destinations. Traffic between STAs within the BSS may be sent through the AP, for example, where the source STA may send traffic to the AP and the AP may deliver the traffic to the destination STA. The traffic between STAs within a BSS may be considered and / or referred to as peer-to-peer traffic. The peer-to-peer traffic may be sent between (e.g., directly between) the source and destination STAs with a direct link setup (DLS). In certain representative embodiments, the DLS may use an 802. l ie DLS or an 802.1 Iz tunneled DLS (TDLS). A WLAN using an Independent BSS (IBSS) mode may not have an AP, and the STAs (e.g., all of the STAs) within or using the IBSS may communicate directly with each other. The IBSS mode of communication may sometimes be referred to herein as an "ad-hoc" mode of communication.

[0069] When using the 802.1 lac infrastructure mode of operation or a similar mode of operations, the AP may transmit a beacon on a fixed channel, such as a primary channel. The primary channel may be a fixed width (e.g., 20 MHz wide bandwidth) or a dynamically set width via signaling. The primary channel may be the operating channel of the BSS and may be used by the STAs to establish a connection with the AP. In certain representative embodiments, Carrier sense multiple access with collision avoidance (CSMA / CA) may be implemented, for example in in 802.11 systems. For CSMA / CA, the STAs (e.g., every 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.

[0070] High throughput (HT) STAs may use a 40 MHz wide channel for communication, for example, via a combination of the primary 20 MHz channel with an adjacent or nonadj acent 20 MHz channel to form a 40 MHz wide channel.

[0071] Very high throughput (VHT) STAs may support 20 MHz, 40 MHz, 80 MHz, and / or 160 MHz wide channels. The 40 MHz, and / or 80 MHz, channels may be formed by combining contiguous 20 MHz channels. A 160 MHz channel may be formed by combining 8 contiguous 20 MHz channels, or by combining two non-contiguous 80 MHz channels, which may be referred to as an 80+80 configuration. For the 80+80 configuration, the data, after channel encoding, may be passed through a segment parser that may divide the data into two streams. Inverse fast fourier transform (IFFT) processing, and time domain processing, may be done on each stream separately. The streams may be mapped on to the two 80 MHz channels, and the data may be transmitted bya transmitting STA. At the receiver of the receiving STA, the above-described operation for the 80+80 configuration may be reversed, and the combined data may be sent to a medium access control (MAC) layer, entity, etc.

[0072] Sub 1 GHz modes of operation are supported by 802.1 laf and 802.11 ah. The channel operating bandwidths, and carriers, are reduced in 802.1 laf and 802.1 lah relative to those used in802.1 In, and 802.1 lac. 802.1 laf supports 5 MHz, 10 MHz and 20 MHz bandwidths in the TV white space (TVWS) spectrum, and 802.1 lah supports 1 MHz, 2 MHz, 4 MHz, 8 MHz, and 16 MHz bandwidths using non-TVWS spectrum. According to a representative embodiment,802.1 lah may support meter type control / machine-type communications (MTC), such as MTC devices in a macro coverage area. MTC devices may have certain capabilities, for example, limited capabilities including support for (e.g., only support for) certain and / or limited bandwidths. The MTC devices may include a battery with a battery life above a threshold (e.g., to maintain a very long battery life).

[0073] WLAN systems, which may support multiple channels, and channel bandwidths, such as802.1 In, 802.1 lac, 802.1 laf, and 802.1 lah, include a channel which may be designated as the primary channel. The primary channel may have a bandwidth equal to the largest 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, from among all STAs in operating in a BSS, which supports the smallest bandwidth operating mode. In the example of 802.1 lah, the primary channel may be 1 MHz wide for STAs (e.g., MTC type devices) that support (e.g., only support) a 1 MHz mode, even if the AP, and other STAs in the BSS support 2 MHz, 4 MHz, 8 MHz, 16 MHz, and / or other channel bandwidth operating modes. Carrier sensing and / or network allocation vector (NAV) settings may depend on the status of the primary channel. If the primary channel is busy, for example, due to a STA (which supports only a 1 MHz operating mode), transmitting to the AP, the entire available frequency bands may be considered busy even though a majority of the frequency bands remains idle and may be available.

[0074] In the United States, the available frequency bands, which may be used by 802.1 lah, are from 902 MHz to 928 MHz. In Korea, the available frequency bands are from 917.5 MHz to 923.5 MHz. In Japan, the available frequency bands are from 916.5 MHz to 927.5 MHz. The total bandwidth available for 802.1 lah is 6 MHz to 26 MHz depending on the country code.

[0075] FIG. ID is a system diagram illustrating the RAN 113 and the CN 115 according to an embodiment. As noted above, the RAN 113 may employ an 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.

[0076] 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 an embodiment, the gNBs 180a, 180b, 180c may implement MIMO technology. For example, gNBs 180a, 180b may utilize beamforming to transmit signals to and / or receive signals from the WTRUs 102a, 102b, 102c. Thus, the gNB 180a, for example, may use multiple antennas to transmit wireless signals to, and / or receive wireless signals from, the WTRU 102a. In an 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 unlicensed spectrum while the remaining component carriers may be on licensed spectrum. In an embodiment, the gNBs 180a, 180b, 180c may implement Coordinated Multi-Point (CoMP) technology. For example, WTRU 102a may receive coordinated transmissions from gNB 180a and gNB 180b (and / or gNB 180c).

[0077] The WTRUs 102a, 102b, 102c may communicate with gNBs 180a, 180b, 180c using transmissions associated with a scalable numerology. 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 gNBs 180a, 180b, 180c using subframe or transmission time intervals (TTIs) of various or scalable lengths (e.g., including a varying number of OFDM symbols and / or lasting varying lengths of absolute time).

[0078] 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 the standalone configuration, WTRUs 102a, 102b, 102c may communicate with gNBs 180a, 180b, 180c without also accessing other RANs (e.g., such as eNode-Bs 160a, 160b, 160c). In the standalone configuration, WTRUs 102a, 102b, 102c may utilize one or more of gNBs 180a, 180b, 180c as a mobility anchor point. In the standalone configuration, WTRUs 102a, 102b, 102c may communicate with gNBs 180a, 180b, 180c using signals in an unlicensed band. In a non- standalone configuration WTRUs 102a, 102b, 102c may communicate with / connect to gNBs 180a, 180b, 180c while also communicating with / connecting to another RAN such as eNode-Bs 160a, 160b, 160c. For example, WTRUs 102a, 102b, 102c may implement DC principles to communicate with one or more gNBs 180a, 180b, 180c and one or more eNode-Bs 160a, 160b, 160c substantially simultaneously. In the non- standalone configuration, eNode-Bs 160a, 160b, 160c may serve as amobility anchor for WTRUs 102a, 102b, 102c and gNBs 180a, 180b, 180c may provide additional coverage and / or throughput for servicing WTRUs 102a, 102b, 102c.

[0079] Each of the gNBs 180a, 180b, 180c 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, support of network slicing, dual connectivity, interworking between NR and E-UTRA, routing of user plane data towards user plane functions (UPFs) 184a, 184b, routing of control plane information towards access and mobility management functions (AMFs) 182a, 182b, and the like. As shown in FIG. ID, the gNBs 180a, 180b, 180c may communicate with one another over an Xn interface.

[0080] The CN 115 shown in FIG. ID may include at least one AMF 182a, 182b, at least one UPF 184a, 184b, at least one session management function (SMF) 183a, 183b, and at least one Data Network (DN) 185a, 185b. While each of the foregoing elements are depicted as part of the CN 115, it will be appreciated that any of these elements may be owned and / or operated by an entity other than the CN operator.

[0081] The AMF 182a, 182b may be connected to one or more of the gNBs 180a, 180b, 180c in the RAN 113 via an N2 interface and may serve as a control node. For example, the AMF 182a, 182b may be responsible for authenticating users of the WTRUs 102a, 102b, 102c, support for network slicing (e.g., handling of different protocol data unit (PDU) sessions with different requirements), selecting a particular SMF 183a, 183b, management of the registration area, termination of NAS signaling, mobility management, and the like. Network slicing may be used by the AMF 182a, 182b, e.g., to customize CN support for WTRUs 102a, 102b, 102c based on the types of services being utilized WTRUs 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 massive mobile broadband (eMBB) access, services for MTC access, and / or the like. 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.

[0082] The SMF 183a, 183b may be connected to an AMF 182a, 182b in the CN 115 via an N11 interface. The SMF 183a, 183b may also be connected to a UPF 184a, 184b in the CN 115 via an N4 interface. The SMF 183a, 183b may select and control the UPF 184a, 184b and configure the routing of traffic through the UPF 184a, 184b. The SMF 183a, 183b may perform other functions, such as managing and allocating UE IP address, managing PDU sessions, controlling policyenforcement and QoS, providing downlink data notifications, and the like. A PDU session type may be IP -based, non-IP based, Ethernet-based, and the like.

[0083] The UPF 184a, 184b may be connected to one or more of the gNBs 180a, 180b, 180c in the RAN 113 via an N3 interface, which may provide the WTRUs 102a, 102b, 102c with access to packet-switched networks, such as the Internet 110, e.g., 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 multihomed PDU sessions, handling user plane QoS, buffering downlink packets, providing mobility anchoring, and the like.

[0084] The CN 115 may facilitate communications with other networks. For example, the CN 115 may include, or may communicate with, an IP gateway (e.g., an IP multimedia subsystem (IMS) server) that serves 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 the other networks 112, which may include other wired and / or wireless networks that are owned and / or operated by other service providers. In an embodiment, the WTRUs 102a, 102b, 102c may be connected to a local Data Network (DN) 185a, 185b through the UPF 184a, 184b via the N3 interface to the UPF 184a, 184b and an N6 interface between the UPF 184a, 184b and the DN 185a, 185b.

[0085] In view of FIGs. 1 A-1D, and the corresponding description of FIGs. 1 A-1D, one or more, or all, of the functions described herein with regard to any of: WTRUs 102a-d, base stations 114a- b, eNode-Bs 160a-c, MME 162, SGW 164, PGW 166, gNBs 180a-c, AMFs 182a-b, UPFs 184a- b, SMFs 183a-b, DNs 185a-b, and / or any other element(s) / device(s) described herein, may be performed by one or more emulation elements / devices (not shown). The emulation devices may be one or more devices configured to emulate one or more, or all, of the functions described herein. For example, the emulation devices may be used to test other devices and / or to simulate network and / or WTRU functions.

[0086] The emulation devices may be designed to implement one or more tests of other devices in a lab environment and / or in an operator network environment. For example, the one or more emulation devices may perform the one or more, 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. The one or more emulation devices may perform the one or more, or all, functions while being temporarily implemented / deployed as part of a wired and / or wireless communication network. The emulation device may be directly coupled to another device for purposes of testing and / or may performing testing using over-the-air wireless communications.

[0087] The one or more emulation devices may perform the one or more, including all, functions while not being implemented / deployed as part of a wired and / or wireless communication network. For example, the emulation devices may be utilized in a testing scenario in a testing laboratory and / or a non-deployed (e.g., testing) wired and / or wireless communication network in order to implement testing of one or more components. The one or more emulation devices may be test equipment. Direct RF coupling and / or wireless communications via RF circuitry (e.g., which may include one or more antennas) may be used by the emulation devices to transmit and / or receive data.

[0088] Introduction

[0089] The following abbreviations and acronyms may be used throughout the description:

[0090] NMSE Normalized Mean Squared Error

[0091] SGCS Squared Generalized Cosine Similarity

[0092] AE Autoencoder

[0093] AI / ML Artificial Intelligence / Machine Learning

[0094] RNN Recurrent Neural Networks

[0095] DNN Deep Neural Network

[0096] CNN Convolutional Neural Network

[0097] TSF Temporal-Spatial-Frequency domain

[0098] SF Spatial-Frequency domain

[0099] CR Compression Rate

[0100] ACK Acknowledgement

[0101] BLER Block Error Rate

[0102] BW Bandwidth

[0103] BWP Bandwidth Part

[0104] CAP Channel Access Priority

[0105] CAPC Channel access priority class

[0106] CCA Clear Channel Assessment

[0107] CCE Control Channel Element

[0108] CE Control Element

[0109] CG Configured grant or cell group

[0110] CP Cyclic Prefix

[0111] CP-OFDM Conventional OFDM (relying on cyclic prefix)

[0112] CQI Channel Quality Indicator

[0113] CRC Cyclic Redundancy Check

[0114] CS Cosine Similarity

[0115] CSI Channel State Information

[0116] CW Contention Window

[0117] CWS Contention Window Size

[0118] CO Channel Occupancy

[0119] DAI Downlink Assignment Index

[0120] DCI Downlink Control Information

[0121] DFI Downlink feedback information

[0122] DG Dynamic grant

[0123] DL Downlink

[0124] DM-RS Demodulation Reference Signal

[0125] DRB Data Radio Bearer

[0126] eLAA enhanced Licensed Assisted Access

[0127] EV Eigenvector

[0128] FeLAA Further enhanced Licensed Assisted Access

[0129] HARQ Hybrid Automatic Repeat Request

[0130] KPI Key Performance Indicator

[0131] LAA License Assisted Access

[0132] LBT Listen-Before-Talk

[0133] LTE Long Term Evolution e.g. from 3GPP LTE R8 and up

[0134] NACK Negative ACK

[0135] NMSE Normalized Mean Squared Error

[0136] MCS Modulation and Coding Scheme

[0137] MIMO Multiple Input Multiple Output

[0138] NR New Radio

[0139] OFDM Orthogonal Frequency-Division Multiplexing

[0140] PHY Physical Layer

[0141] PID Process ID

[0142] PO Paging Occasion

[0143] PRACH Physical Random Access Channel

[0144] PSS Primary Synchronization Signal

[0145] RA Random Access (or procedure)

[0146] RACH Random Access Channel

[0147] RAR Random Access Response

[0148] RCU Radio access network Central Unit

[0149] RF Radio Front end

[0150] RLF Radio Link Failure

[0151] RLM Radio Link Monitoring

[0152] RNTI Radio Network Identifier

[0153] RO RACH occasion

[0154] RRC Radio Resource Control

[0155] RRM Radio Resource Management

[0156] RS Reference Signal

[0157] RSRP Reference Signal Received Power

[0158] RSSI Received Signal Strength Indicator

[0159] SDU Service Data Unit

[0160] SGCS Squared Generalized Cosine Similarity

[0161] SRS Sounding Reference Signal

[0162] SF Spatial-frequency

[0163] SS Synchronization Signal

[0164] SSS Secondary Synchronization Signal

[0165] SWG Switching Gap (in a self-contained subframe)

[0166] SPS Semi-persistent scheduling

[0167] SUL Supplemental Uplink

[0168] TB Transport Block

[0169] TBS Transport Block Size

[0170] TRP Transmission / Reception Point

[0171] TSC Time-sensitive communications

[0172] TSF Time-spatial-frequency

[0173] TSN Time-sensitive networking

[0174] UL Uplink

[0175] URLLC Ultra-Reliable and Low Latency Communications

[0176] WBWP Wide Bandwidth Part

[0177] WLAN Wireless Local Area Networks and related technologies (IEEE 8O2.xx domain)

[0178] In certain representative embodiments, methods and procedures are described for a WTRU 102 which may reduce overhead (e.g., for CSI reporting) using two-sided (e.g., AI / ML)models that leverage temporal correlation properties of signal and / or transmission measurements, such as CSI.

[0179] CSI Reporting Framework

[0180] As used herein, Channel State Information (CSI) may be used interchangeably to refer to any of the following: channel quality index (CQI), rank indicator (RI), precoding matrix index (PMI), LI channel measurement (e.g., RSRP such as Ll-RSRP, or SINR), CSLRS resource indicator (CRI), SS / PBCH block resource indicator (SSBRI), layer indicator (LI) and / or any other measurement quantity measured by a WTRU 102 from one or more reference signals.

[0181] As used herein, a reference signal may be used interchangeably with to refer to CSLRS or SS / PBCH block or any other (e.g. configured) reference signal.

[0182] In certain representative embodiments, a WTRU 102 may be configured to report CSI through uplink (UL) transmissions, such as an uplink control channel transmission (e.g., PUCCH) or (e.g., per a gNB 180 request) an uplink shared channel transmission (e.g., PUSCH grant). Depending on the configuration, CSLRS may cover (e.g., be transmitted using) the full bandwidth of a BWP or a fraction of it. Within a CSLRS bandwidth, a CSLRS may be configured in certain resources, such as each PRB or every other PRB. In the time domain, CSLRS resources may be configured either periodic, semi-persistent, or aperiodic. For example, a semi-persistent CSLRS may be similar to a periodic CSLRS, except that the resource may be (de)-activated (e.g., by MAC CE) and the WTRU 102 may report related measurements only when the resource is activated. For an aperiodic CSLRS, the WTRU 102 may be triggered to report the measured CSLRS on PUSCH, such as by a request using DCI. Periodic reports may be carried over the PUCCH, while semi- persistent reports may be carried either on PUCCH or PUSCH. The reported CSI may be used by the scheduler when allocating optimal resource blocks, such as based on the channel’s timefrequency selectivity, determining precoding matrices, beams, transmission mode and / or selecting suitable MCSs. The reliability, accuracy, and timeliness of WTRU 102 CSI reports are important aspects, and as such may be critical to meeting URLLC service requirements.

[0183] In certain representative embodiments, a WTRU 102 may be configured with a CSI measurement setting (e.g., configuration) which may include one or more CSI reporting settings, resource settings, and / or a link between one or more CSI reporting settings and one or more resource settings. FIG. 2 is a report and resource diagram illustrating an example configuration for CSI reporting settings, resource settings, and links.

[0184] As shown in FIG. 2, a plurality of CSI reporting configurations, such as CSI Reporting Setting 0202-a and CSI Reporting Setting 1 202-b, may be linked by a measurement configuration, such as a Measurement Setting 204, with a plurality of resource configurations. In FIG. 2, theresource configurations include a Resource Setting 0206-a (e.g., for non-zero power CSI reference signal(s) (NZP CSI-RS(s))), a Resource Setting 1 206-b (e.g., for ZP CSI-RS(s)), and a Resource Setting 2 206-c (e.g., for NZP CSI-RS(s)). In the example shown in FIG. 2, the CSI Reporting Setting 0 202-a may be linked to the Resource Setting 0 206-a, the Resource Setting 1 206-b, and the Resource Setting 2206-c. The CSI Reporting Setting 1 202-a may (e.g., only) be linked to the Resource Setting 0 206-a.

[0185] Artificial Intelligence and Autoencoding

[0186] For example, Al may be broadly defined as the behavior exhibited by machines. Such behavior may mimic any of cognitive functions to sense, reason, adapt and / or act. The terms Artificial Intelligence (Al), Machine Learning (ML), Deep Learning (DL), DNNs may be used interchangeably herein. Methods and procedures described herein are presented based on learning in wireless communication systems, but are not limited to such scenarios, systems and services and may be applicable to any type of transmissions, communication systems and / or services etc.

[0187] Autoencoders

[0188] For example, autoencoders (AEs) refer to a specific class of deep neural networks (DNNs) that arise in the context of un-supervised machine learning settings where higher-dimensional data is non-linearly transformed to a lower dimensional latent vector using a DNN based encoder. The lower dimensional latent vector may then be used to re-produce the higher-dimensional data using a non-linear decoder. For example, the encoder may be represented as E(x; lVe) where x is the higher-dimensional data and VFerepresents the parameters of the encoder. For example, the decoder may be represented as D(z; Wd~) where z is the low-dimensional latent representation and Wdrepresents the parameters of the decoder. Further, using training data { xlt••• , xN] the autoencoder can be trained by solving the following optimization problem:{Wetr, W = arg

[0189] The above problem may be approximately solved using a backpropagation algorithm. The trained encoder E(x; I / ! / etr) can be used to compress higher-dimensional data and the trained decoder D z; Wdr) can be used to decompress the latent representation.

[0190] Machine learning based approaches (e.g., autoencoders) have the potential to offer a balance between CSI feedback overhead and reconstruction performance. Existing work relies on spatial -frequency (SF) CSI compression using the estimated channel sample at a given time. For example, the estimated SF channel (e.g., a preprocessed version) may be first compressed into a latent representation using an Al encoder model, then transmitted, and decompressed by an Aldecoder model to recover the desired CSI. While SF compression was shown to provide acceptable reconstruction quality, there still remains room for improvement to approach the upper bound of performance for uncompressed CSI.

[0191] One approach to further improve the compression performance (either from an overhead perspective or reconstruction quality) is to leverage the temporal correlation properties of the channel in the compression process. This may be referred to as time-spatial-frequency (TSF) compression. Exploiting the CSI temporal correlation on the top of SF compression has the potential to provide 1) further improvement in the reconstruction performance for a given overhead, 11) further reduction in the overhead for a given performance, iii) further improvement in both performance and overhead relative to the SF compression, and / or iv) flexibility in controlling the CSI feedback overhead over time.

[0192] In certain embodiments, for TSF compression, the latent (e.g., domain) representation may be structured to include two time-dependent parts: one common part and a sample-specific part. For example, the common part may capture the long-term variations or common information in a sequence of correlated samples. For example, the sample-specific part may capture the shortterm variations or the individual information in each sample. For example, the common information may (e.g., only) need to be sent once every number of samples / transmission while the specific information may need to be transmitted every transmission. Enforcing such a structure in the latent domain has the potential to significantly reduce the signaling overhead.

[0193] In certain representative embodiments, methods and procedures described herein may enable seamless and efficient TSF operation through structuring the latent domain into two parts, namely the common and specific. For example,

[0194] In certain representative embodiments, methods and procedures described herein may enable time-dependent multipart generation and transmission thereof.

[0195] In certain representative embodiments, methods and procedures described herein may determine whether a single-part or two-part transmission (e.g., mode) is to be used (e.g., needed).

[0196] In certain representative embodiments, methods and procedures described herein may address error propagation problems associated with multipart transmission.

[0197] In certain representative embodiments, methods and procedures described herein may include monitoring the performance (e.g., validity time) of the common part.

[0198] In certain representative embodiments, methods and procedures described herein may include triggers for sending one or more of the parts.

[0199] In certain representative embodiments, to enable efficient TSF compression, the AE latent space may be structured into a common latent part and a specific latent part. The common latent part may capture the long-term variations in a sequence of temporally correlated CSI samples while the short-term latent may capture the short-term variations in CSI samples (e.g., between two consecutive CSI samples).

[0200] As used herein, the terms long-term latent and / or common latent may be used interchangeably to represent the long-term variations in a sequence of CSI samples.

[0201] As used herein, the terms short-term latent and / or specific latent may be used interchangeably to represent the short-term variations in a sequence of CSI samples.

[0202] FIG. 3 is a processing diagram illustrating an example AI / ML autoencoder with a timedependent multipart (TDMP) latent domain that includes a common part and a specific part.

[0203] As shown in FIG. 3, an AI / ML encoder (e.g., an autoencoder) 302 may use a sequence of samples FL as input to generate a latent structure (e.g., a latent) having a common part 304-a that captures the long-term variations in the sequence and a specific part 304-b that captures the shortterm variations in each sample of the sequence. For example, at a time instance z, the common part 304-a and the specific part 304-b may be provided (e.g., sent) to an AI / ML decoder 306 (e.g., at a gNB 180). The common part 304-a may be stored in a buffer to be used for decoding by the AI / ML decoder 306. At other time instances of 2 < z < Nt, (e.g., subsequent) specific parts 304-b may be provided to the AI / ML decoder 306. The AI / ML decoder 306 may perform decoding using the stored common part 304-a and the specific parts 304-b to recover a sequence of samples FL = / (Zc, z / ).

[0204] In certain representative embodiments, method and procedures are described which address TSF compression. However, such embodiments may also be applied to other frameworks where a latent structure having a common part and a specific part are used.

[0205] Example Procedures for CSI Compression with TDMP Generation and Transmission

[0206] TSF compression utilizes channel temporal correlations to improve the compression performance (e.g., overhead and / or reconstruction relative to space-frequency compression). To enable TSF compression, an AI / ML model can be trained to generate two time-dependent latent parts. Two latent parts may be generated - one long-term / common part that captures the long-term variations in a CSI sequence and one short-term / specific part that captures the short-term variations in each sample in a CSI sequence. Methods and procedures described herein may include performing CSI compression with time-dependent multipart generation, determining (e.g.,selecting) which parts to transmit, and reporting the information associated with the generated / selected parts.

[0207] FIG. 4 is a timing diagram illustrating an example timing of UL CSI reporting of common and specific TSF parts.

[0208] In certain representative embodiments, a WTRU 102 may be configured with one or more parameters to generate CSI feedback (e.g., with an AI / ML model such as an AE). The feedback includes TDMP latents.

[0209] For example, the common part may capture the long-term and / or coarse variations in a given sequence of correlated samples. As shown in FIG. 4, a common part 304-a may be sent once every N CSI feedback transmissions. The common part may be stored by the receiving node and reused (e.g., to generate feedback) for N consecutive measurement and / or feedback instances. The common part may be transmitted semi-persistently, periodically, or aperiodically.

[0210] For example, the specific part may capture the short-term and / or fine variations between (e.g., every) a pair of samples in a given sequence of correlated samples. A respective specific part 304-b (e.g., of a latent structure generated for samples of a measurement instance) may be sent in every CSI feedback transmission. The specific part may be transmitted periodically.

[0211] In certain representative embodiments, a WTRU 102 may receive configuration information for one or more parameters associated with TDMP CSI feedback.

[0212] For example, the configuration information may include any of the following. The WTRU 102 may be configured with a TDMP configuration and parameters thereof which may include any of: (i) parts information (e.g., a number of parts / latents, size of each part, quantization associated with each part); (ii) information indicating TDMP mode activation (e.g., an implicit indication, such as an ID of a specific model with TDMP operation); (iii) TDMP feedback transmission modes (e.g., single-part mode where the WTRU 102 is configured to transmit CSI feedback with only one part, such as the specific part; multi-part mode where the WTRU 102 is configured to transmit CSI feedback with multiple parts, such as common and specific parts); (iv) a TDMP feedback transmission mode selection flag (e.g., to enable the WTRU 102 to select and / or determine and report a preferred TDMP feedback transmission mode); and / or (v) an input domain type (e.g., EV or full CSI). The WTRU 102 may be configured with one or more conditions and / or parameters for TSF parts updating which may include any of: (i) a first performance metric threshold (e.g., a short-term SGCS / NMSE threshold, such as between consecutively generated common latents / parts); (ii) a second performance metric threshold, e.g., a long-term SGCS / NMSE threshold, such as between a first / active common part (where the active common part is one that is stored and used at the gNB 180 for reconstruction), and one or more generated / additionalinactive common parts (where an inactive common part is one that is generated over time but is not yet fed back to the gNB 180 and / or used / stored at the gNB 180). The WTRU 102 may be configured with a reporting configuration (e.g., part-specific and / or part-common reporting resources, MCS and / or quantization associated with each part, and / or UL resources for each part).

[0213] In certain representative embodiments, a WTRU 102 may receive one or more CSI-RS transmissions and may estimate the channel. For example, the WTRU 102 may (e.g., optionally) store the measured CSI (e.g., full channel measurements).

[0214] In certain representative embodiments, a WTRU 102 may apply time-dependent multipart TSF compression by generating one or more parts / latents (e.g., common / long-term part and shortterm-part) associated with the received CSI transmission. The compression may be based on the configured TDMP parameters (e.g., number of parts, size of parts, Model ID) and / or configured reporting (e.g., quantization, MCS, etc.). In some embodiments, the WTRU 102 may (e.g., optionally) store the generated TSF common part in a TSF latent buffer (e.g., based on a TDMP configuration, such as a part monitoring flag).

[0215] In certain representative embodiments, a WTRU 102 may generate time-dependent parts / latents, e.g., common part and specific part associated with the received CSI-RS transmission based on the configured TDMP parameters (e.g., number of parts, size of each, etc.) and / or configured reporting (e.g., quantization, MCS, etc.) In some embodiments, the WTRU 102 may (e.g., optionally) store the generated common part (e.g., based on a TDMP configuration, such as a part monitoring flag).

[0216] In certain representative embodiments, a WTRU 102 may determine and / or select a preferred TDMP feedback transmission mode (e.g., for the next transmission). For example, the determination and / or selection may be based on preconfigured conditions (e.g., TDMP feedback transmission mode selection flag) and / or measurements. For example, the WTRU 102 may determine the TDMP feedback transmission mode based on one or more of the following: measured channel parameters, CSI measurements, PDSCH performance, feedback resources, and / or a network indication. For example, the TDMP feedback transmission mode may be determined based on measured channel parameters (e.g., Doppler spread and / or WTRU speed). For example, a multipart transmission mode may be selected where the estimated Doppler spread exceeds a certain configured threshold. For example, the TDMP feedback transmission mode may be determined based on CSI measurements, such as by comparing the measured long-term and / or short-term SGCS (e.g., full channel or latent) with the configured long / short-term SGCS threshold. For example, the multipart transmission mode may be selected where the measured long-term SGCS value drops below the configured threshold. For example, the TDMP feedback transmissionmode may be determined based on PDSCH performance, such as in terms of a number of NACKs within a time interval. For example, the TDMP feedback transmission mode may be determined based on feedback resources (e.g., time, frequency, BW, and / or duration). For example, the TDMP feedback transmission mode may be determined based on an indication received from a gNB 180.

[0217] In certain representative embodiments, a WTRU 102 may report one or more of the following: (i) compressed CSI based on the configured TDMP feedback transmission mode; and / or (ii) the preferred TDMP feedback transmission mode (e.g., to be used in a next or future reporting instance(s)). In some embodiments, the WTRU 102 may receive information indicating any of a (re)configuration, an activation, and / or a deactivation (e.g., for a TDMP mode) for subsequent CSI reporting.

[0218] Example Procedures for Monitoring, Determining, and Improving the Validity Time of TDMP TSF / CSI Compression

[0219] TSF compression utilizes channel temporal correlation to improve the compression performance (e.g., relative to space-frequency compression). To enable TSF, an AI / ML model can be trained to generate time-dependent latent parts. Two latent parts may be generated - one long- term / common part that captures the long-term variations in a CSI sequence and one short- term / specific part that captures the short-term variations in each sample in a CSI sequence. The common component may be transmitted periodically (e.g., every N slots) and may be reused by the receiving node for the reconstruction process (e.g., every slot). Control aspects are needed to monitor, determine, and / or improve the validity time of the common part for efficient and reliable TSF operation with minimal overhead.

[0220] For example, a WTRU 102 may determine one or more parameters associated with the validity time of the TSF parts (e.g., long-term and short-term parts) as a function of channel conditions and / or configured metrics thresholds.

[0221] In certain representative embodiments, a WTRU 102 may be configured with one or more parameters to generate CSI feedback (e.g., with an AI / ML model such as an AE). The feedback includes TDMP latents.

[0222] For example, the common part may capture the long-term and / or coarse variations in a given sequence of correlated samples. The common part may be sent once every N CSI feedback transmissions. The common part may be stored by the receiving node (e.g., network -side decoder) and reused for reconstruction for N consecutive slots. The common part may also be stored at the transmitting node (e.g., WTRU-side) for monitoring purposes.

[0223] For example, the specific part may capture the short-term and / or fine variations between (e.g., every) a pair of samples in a given sequence of correlated samples. The specific part may be sent in every CSI feedback transmission. The specific part may be transmitted periodically.

[0224] In certain representative embodiments, a WTRU 102 may receive configuration information for one or more parameters associated with monitoring and / or controlling one or more aspects associated with the time-dependent multipart CSI feedback.

[0225] For example, the configuration information may include any of the following. The WTRU 102 may be configured with a TDMP configuration and parameters thereof which may include any of (i) parts information (e.g., number of parts / latents, size of each part, quantization associated w / each part); (ii) an initial and / or max validity time and / or BW of the common part, such as information indicating the initial / maximum number of samples to be used for a common part or time (e.g., in seconds to freeze / reuse the common part); (iii) a monitoring resolution (e.g., full common / all latents monitoring or per-latent monitoring, all latents monitoring); (iv) information indicating a TDMP mode activation (e.g., may be implicit, such as the ID of a specific TDMP model with multi-part operation); and / or (v) an input domain type (e.g., EV or full CSI). The WTRU 102 may be configured with one or more conditions and / or parameters for TSF parts updating and / or monitoring which may include any of (i) a first performance metric threshold (e.g., Short-term SGCS / NMSE threshold: between consecutively generated common latents / parts); (ii) a second performance metric threshold, e.g., Long-term SGCS / NMSE threshold: between first / active common part, such as where the active common part is one that is stored and used at the gNB 180, such as for reconstruction, and generated / additional inactive common parts such as where an inactive common part is one that is generated over time but not fed back to the gNB 180 and not used / stored at the gNB 180); (iii) a Part monitoring flag, such as to indicate to the WTRU 102 to start monitoring one of the parts (e.g., common) performance and / or indicate to the WTRU 102 to start storing the common part / latent in a buffer. The WTRU 102 may be configured with a reporting configuration, such as part-specific and / or part-common reporting aspects (e.g., MCS and / or quantization associated with each part, UL resources for each part).

[0226] In certain representative embodiments, a WTRU 102 may receive one or more CSLRS transmissions and may estimate the channel. In some embodiments, the WTRU 102 may (e.g., optionally) store the measured CSI (e.g., full channel measurements).

[0227] In certain representative embodiments, a WTRU 102 may apply time-dependent multipart TSF compression by generating one or more parts / latents (e.g., common / long-term part and shortterm-part) associated with the received CSI transmission. The compression may be based on the configured TDMP parameters (e.g., number of parts, size of parts, Model ID) and / or configuredreporting (e.g., quantization, MCS, etc.). In some embodiments, the WTRU 102 may (e.g., optionally) store the generated (e.g., TSF) common part, such as in a TSF latent buffer (e.g., based on a TDMP configuration, such as a part monitoring flag).

[0228] In certain representative embodiments, a WTRU 102 may determine one or more parameters associated with the common part, such as a validity time and / or life-cycle (e.g., in seconds or number of samples), a validity BW, and / or an update request (e.g., common part refinement) based on preconfigured conditions and measurements. For example, a WTRU 102 may determine the validity time associated with the common part, (e.g., the validity time is the estimated time for the common part to work properly before the performance drops below a certain level). The validity time may be based on one or more of the following: measured channel parameters, CIS measurements, and / or PDSCH performance. For example, the measured channel parameters may include Doppler spread and / or WTRU speed, such as where the long- term / common part validity time is determined based on the estimated Doppler spread with each Doppler spread (e.g., a value and / or a range) mapping to a specific validity time. For example, the WTRU 102 may compare the measured long-term and / or short-term SGCS with the configured long / short-term SGCS threshold and map the difference to a specific validity time. Such measurements may be either in the full channel domain or in the latent domain associated with the common part. For example, the WTRU 102 may determine the PDSCH performance based on (e.g., in terms of) a number of NACKs within a time interval.

[0229] The WTRU 102 may determine if common part update feedback is needed (e.g., the update may be in one of different formats; replace / reset request, incremental update request by sending the differential common value, model switch for higher resolution common part) based on one or more of the following: based on full CSI measurements, long-term latent measurements, and / or other triggers. For example, common part update feedback may be determined periodically by comparing the measured long-term SGCS with the configured long-term SGCS threshold. As an example, the measured SGCS is below the first configured threshold, the WTRU 102 may indicate a replacement of the common part. In another example, if the measured SGCS is below a second threshold, the WTRU 102 may indicate an incremental update through sending a differential value for the common part (e.g., of one or more latents). For example, common part update feedback may be determined based on long-term latent measurements, such as where there is a discrepancy in the long-term latent. The WTRU 102 may compare the measured long-term SGCS between the stored long-term part at the gNB 180 and the current generated common part with one or more threshold(s) and send an update request accordingly.

[0230] In certain representative embodiments, a WTRU 102 may report one or more of the following: (i) the compressed CSI with one or more time-dependent parts; and / or (ii) parameters associated with the monitoring performance of the common part (e.g., validity time or BW, update type such as replace, incremental update or no update).

[0231] Terminology

[0232] As presented herein, various representative embodiments are described for CSI feedback with TSF compression. However, theses various embodiments are not to be limited to just CSI feedback and TSF. Rather, the various representative embodiments are generally applicable to any transmission, report, measurements, and the likes to which the proposed structure is applicable, namely extracting sample-common information and sample-specific information. In addition, the various embodiments described herein use an AI / ML models for the coding framework, but the embodiments are generally applicable to non-AI / ML encoded transmissions.

[0233] As described herein, SF compression may refer to a compression technique that compresses a current CSI sample (e.g., raw channel or eigenvector) using an encoder model at the WTRU 102. The compressed CSI is used to recover the decompressed CSI using a decoder model at the gNB 180.

[0234] As described herein, TSF compression may refer to a compression technique that utilizes at least one past / historical CSI sample (e.g., raw channel or eigenvector) along with the currently observed CSI sample at the WTRU 102 to generate a current compressed CSI. At least one past / historical CSI measurement along with the current compressed CSI at the gNB 180 may be used to generate and / or recover the current decompressed CSI.

[0235] As described herein, a TDMP structure may refer to an enforced structure on the output of an AI / ML model where the model output is structured to include multiple parts. At least one part captures and / or represents common information in a sequence of samples (e.g., time series data), and may be referred to as sample-common information. At least one other part captures and / or represents specific information in each sample of the sequence, and may be referred to as sample-specific information. The terms TDMP structure, TDMP transmission, and / or TDMP compression may be used interchangeably to indicate an AI / ML model with TDMP.

[0236] As described herein, the term common part may refer to a part carrying latent common information and / or capturing coarse variations in a sequence of correlated data samples. The terms sample-common part, common part, and / or long-term part may be used interchangeably to represent the common information carried by the common part.

[0237] As described herein, the term specific part may refer to a part carrying latent specific information in each sample and / or fine variations between pairs of consecutive samples in asequence of correlated data samples. The terms sample-specific part, specific part, and / or shortterm part may be used interchangeably to represent the specific information carried out by the specific part.

[0238] As described herein, the term TSF latent buffer may refer to a buffer used at the WTRU 102 to store past CSI samples (e.g., raw channel or eigenvector) and / or a buffer used at the gNB 180 to store past decompressed CSI samples.

[0239] Principles and Observations

[0240] In general terms, SF compression operates on a sample-by sample basis, where a WTRU 102 uses an AE model to perform the compression at a time slot n based on an estimated channel HnFIG. 5 is an SF compression diagram illustrating an example of a two-sided autoencoder model. In FIG. 5, the WTRU 102 may first compress Hnusing an encoder model 302 to generate and send back the latent representation zn, (e.g., compressed CSI) and then the gNB 180 may use a decoder model 306 to decompress the received latent znto recover Hn. For example, the difference and / or distance between the estimated channel Hnat the WTRU 102 and the recovered (e.g., decompressed) channel Hnat the gNB 180 may be representative of the compression loss.

[0241] To further improve the SF compression performance, from an overhead reduction and / or performance perspective, the autoencoder model may be trained to represent the compressed information of a given sequence of samples in two parts - one part that carries common information for a sequence of correlated samples and another part that captures specific information in each sample. The two parts information are then fed together to the AI / ML decode model to reconstruct the desired CSI sample. The common information may be sent once every number of compression instances while the specific information may be sent for every compression instance. This model may be referred to as a TDMP model and the compression mode may be referred to as TSF compression wherein the temporal correlation properties of the channel samples are leveraged to generate latent parts with some enforced structure to reduce the feedback overhead over time, as shown in FIG. 3. The trained AI / ML model is provided to generate multiple latent parts, possibly of different sizes, that capture the long-term and short-term correlations of the raw CSI samples which are input thereto. Enforcing such a structure in the latent domain may result in reduction in the signaling overhead relative to SF compression.

[0242] In the example TDMP model of FIG. 3, where in the common part and specific part are generated using a single AI / ML model (e.g., AE). The proposed framework, however, is not limited to the model architecture illustrated in FIG. 3, and can be obtained using different model architectures. For example, the common part may be generated using a first AI / ML (e.g., AE)model while the specific part may be obtained using a second AI / ML model, where in the output of the first and second AI / ML model may be used together to reconstruct / recover the desired information.

[0243] It should be understood that the proposed AI / ML autoencoder model in FIG. 3 may be extended beyond compression of CSI and is generally applicable to other sample types where correlation is present in a sequence of samples.

[0244] TDMP Model Training Procedures

[0245] In certain representative embodiments, an AI / ML model (e.g., Autoencoder) may be trained, such as by a WTRU 102, to generate an encoder output with some enforced structure for TSF compression operation. For instance, given a sequence of N temporally-correlated samples HN}, the AE model may be trained to output a long-term part, zc, and a short-term samplespecific part, z(-, wherein the long-term / common part captures the long-term correlation in the input sequence of sample while the short-term parts capture the short-term correlation between every two consecutive samples in the input sequence. The long-term part size |zc| may be greater than each of the short-term size |z |, for i = 1, . . . , N, and the sum of sizes may be less than or equal to a given max size zmax.

[0246] To train the AE model to generate a latent output with a common structure zc(e.g., a sample-common part) and a short-term part z , i = 1, ... , N (e.g., a sample-specific part) one may use the following procedure. First, the autoencoder model is trained in the traditional fashion with same channel samples fed as input and labels for the model training. Thus, given a sequence of N (e.g., N = 3) temporally-correlated samples {H1, H2, H3}, the model is trained with the following input / output pairs:During the first stage of the training all the weights of the model are updated. The same channel matricesare used as both inputs and as the labels. FIG. 6 is an encoding diagram illustrating an example of training of an AE model.

[0247] Once the training is finished, all the weights of the model except for the ones directly interacting with the latent (e.g., the output layer of the encoder and input layer or the 1stlayer weights of the decoder) are frozen and will not be updated. Then, the training associated with generating the common part zcis undertaken. To generate the long-term / common part zc, the following samples pairs may be constructed (HlzH2), (H , W3), ( / / 2, ^i), (^2- ^3), (^3, ^2), (#3^1)-

[0248] The idea behind the pairing of dis-similar channels is to force the model to learn the common components between different channels. While training this framework, the part of the latent corresponding to the common part zcis allowed to propagate to the decoder in the forwardpass, and the corresponding gradients are allowed to back propagate to the encoder. Whereas the parts of the latent corresponding to the short-term components ztare blocked. The blocking of this part of the latent maybe achieved by forcefully multiplying the latent component zLwith a zero vector. FIG. 7 is an encoding diagram illustrating an example of blocking the short-term components of an AE model.

[0249] During this stage of the training, only the weights directly interacting with zcare being updated, as only the common part of the latent will propagate to the decoder and the gradients associated with only this part will be back propagated.

[0250] Once the training for the common part has been completed, the training for the short-term components {z ,.., zN} may be undertaken. To train the part of the encoder that generates the shortterm components zt, the training is performed with the pair of channels where the same set of channels is used for both input and output, as shown in FIG. 8. For example, H1, H- , (H2, H2~), (H3, H3) wherein the output latent ztshould be used along with the common latent / part zc, for reconstruction. FIG. 8 is an encoding diagram illustrating an example of training of an AE model with common and specific parts where only weights associated with the specific part are updated and the common part weights are frozen.

[0251] In another technique for TDMP model training, different models may be utilized for generating the common part of the latent, zc, 304-a and the short-term part of the latent, z 304- b. In this setup, the model may first be trained to generate the common part using a common encoder / decoder module. FIG. 9 is another encoding diagram illustrating an example of training an AE model with common encoder 302-a and decoder 302-b modules. For training, dis-similar channels may be utilized as input and label pairs to force the model to learn the common components between different channels. Therefore, the training data would have input and label pairs like

[0252] Post training, the common encoder 302-a weights may be frozen, and a second encoderdecoder pair (e.g., residual encoder-decoder) is introduced. To train this model, that would generate the short-term components {Zi,.., zw], the training is performed with the pair of channels where the same set of channels is used both as inputs and labels. For example, (H^ H-^), (H2, H2), H3, H3). FIG. 10 is another encoding diagram illustrating an example of training an AE model with residual encoder 302-b and decoder 306-b modules. In FIG. 10, the weights of the common encoder 302-a and decoder 306-a models are kept fixed and only the residual encoder 302-b and the decoder 302-b are updated using the defined loss function. This residual model effectively learns to compress on top of the common encoder-decoder models.

[0253] As another option, the autoencoder may (e.g., also) be trained to output the latent for each Hi in multiple parts {z^,.., zt M}, such that zi :Lis the latent component associated with the longest temporal window, sayand zi Mis the latent component associated with the smallest temporal window, say KM, where K > K2> ••• > KM.

[0254] Benefits

[0255] The various TSF compression embodiments described herein may offer any of the following benefits. For example, TSF compression leverages CSI temporal correlation properties which may enable high compression capabilities relative to the spatial frequency (SF) and legacy based compression for a given target performance. For example, TSF compression leverages the CSI temporal correlation properties which may (e.g., further) improve the performance of SF and legacy based compression at a given compression rate. For example, TSF compression may provide both performance and overhead reduction gains relative to SF and legacy compression. For example, TSF compression may provide dynamic adaptation and flexible use of the past historical samples to achieve the balance between performance, complexity, overhead and storage.

[0256] TSF / CSI Compression with TDMP Generation and Transmission

[0257] TSF compression utilizes channel temporal correlations to improve the compression performance (e.g., overhead and / or reconstruction) relative to space-frequency (SF) compression. To enable TSF compression, an AI / ML model can be trained to generate time-dependent latent parts, wherein at least two latent parts can be generated. A long-term / common part that captures the long-term variations in a CSI sequence and a short-term / specific part that captures the shortterm variations in each sample in a CSI sequence. Such solutions can enable the WTRU 102 to perform CSI compression with TDMP generation, determine / select which parts to transmit, and report the information associated with the generated / selected parts.

[0258] Configurations for TDMP Generation

[0259] Configuration of AI / ML Model with TDMP Latents

[0260] In certain representative embodiments, a WTRU 102 may be configured to generate one or more quantities associated with CSI feedback reporting using an AI / ML model, such as a configured AE model. The AI / ML encoder model may be trained to generate a latent output with some enforced structure, such as where the structure may be preconfigured and / or predefined for the WTRU 102. For example, the output of the configured AI / ML encoder model may be structured to have two parts, namely a common part and a specific part. The common part captures the common information in a sequence of temporally correlated CSI samples (e.g., N samples) and can be commonly used in reconstructing each of the samples. The specific part captures sample-specific information, which is (e.g., unique or sample-specific) information for each sample in the sequence and can be used along with the common information to reconstruct a desired sample (e.g., of the N samples). As used herein, an AI / ML model with common and specific parts structure may be referred to as a TDMP model, TDMP latent, and / or TDMP compression.

[0261] Configuration of Common and Specific Parts

[0262] In certain representative embodiments, a common part may be configured to be sent once for every N CSI feedback transmissions. For example, the value of N may be preconfigured for a WTRU 102. For example, the value of N may be function of model input dimension, subband size, carrier frequency, and / or other parameters. For example, the value of N may be configured as a function of the TDMP model type, common part size, and / or model complexity. For example, the WTRU 102 may be configured to store the common part to determine one or more parameters associated with the common part. For example, the common part may be configured to be transmitted periodically, aperiodically, and / or semi-persistently.

[0263] In certain representative embodiments, a WTRU 102 may be configured to transmit the specific part in every CSI feedback transmission. As compared to the common part, the specific part may be configured to be transmitted (e.g., only) periodically. For example, the WTRU 102 may be configured to transmit the specific part in a different manner than the common part, such as using a first MCS for the specific part and the common part using a second MCS.

[0264] TDMP Model Configuration

[0265] In certain representative embodiments, different TDMP models may have different part sizes. For example, one TDMP model may have a common part size greater than a specific part size. For example, another TDMP model may have a common part size smaller than or equal to the specific part size. The WTRU 102 may be explicitly configured and / or indicated to use and / or apply TDMP compression. For example, a WTRU 102 may be configured with a TDMP model explicitly, such as through a specific model ID. For example, a WTRU 102 may be configured with a TDMP model implicitly, such as using a function of the configured subbands, carrier frequency range, input dimension, and / or other parameters. The WTRU 102 may be configured with part specific parameters. For example, the WTRU 102 may be configured with a quantization type to use for each of the specific part and / or the common part. The quantization type may be configured to be the same or different for each of the common part and specific part. The WTRU 102 may be configured with an input type of the TDMP model, such as where the input type may be the channel EVs or the full CSI matrix.

[0266] TDMP Feedback Transmission Mode Configuration

[0267] In certain representative embodiments, a WTRU 102 may be configured with different TDMP feedback transmission modes. The terms TMDP transmission modes and TDMP feedback transmission modes may be used interchangeably. For example, a TDMP transmission mode may indicate to the WTRU 102 which part(s) to transmit. The TDMP transmission modes may include (e.g., a TDMP transmission mode for) a single part transmission, where the WTRU 102 is configured to transmit CSI feedback with only one part (e.g., specific part only or common part only). The TDMP transmission modes may include (e.g., a TDMP transmission mode for) a multipart transmission, where the WTRU 102 is configured to transmit CSI feedback with two parts (e.g., specific part and common part). An example of multipart transmission is shown in FIG. 4. For example, a TDMP transmission mode may be configured explicitly, such as with an explicit indication of the transmission mode. For example, a TDMP transmission mode may be configured as a function of a (e.g., associated) UL feedback resource configuration. For example, a first type of UL resource may imply a TDMP mode with single-part transmission while a second type of UL resource may imply a TDMP mode with multipart transmission.

[0268] In certain representative embodiments, a WTRU 102 may be configured and / or indicated to select, determine, and / or report a preferred TDMP transmission mode, such as via any of RRC, UCI, PUSCH resource, and / or MAC CE. The WTRU 102 may be configured with parameters and / or conditions to determine and report the TDMP transmission mode. For example, the WTRU 102 may be explicitly configured by a parameter (e.g., TDMP mode selection flag) to enable the WTRU 102 to select and report the preferred TDMP mode. The WTRU 102 may be configured with one or more assistance information to determine a TDMP transmission mode. For example, the WTRU 102 may be configured with one or more performance metric threshold(s). For example, a performance metric threshold may be either a long-term metric or a short-term metric. The long-term metric may be used to assist the WTRU 102 to determine one or more aspects associated with a common part. The long-term performance metric may include the NMSE or SGCS. The short-term metric may be used to assist the WTRU 102 to determine one or more aspects associated with a specific part. The short-term performance metric may include the NMSE or SGCS.

[0269] Procedures for TDMP Transmission Mode Selection

[0270] In certain representative embodiments, one or more TDMP feedback transmission modes may be used for TSF (e.g., CSI) compression. A TDMP transmission mode may be defined, determined, or identified based on the number of parts associated with the CSI reporting. For example, a first TDMP transmission mode may be associated with CSI reporting of a first number of parts, a second TDMP transmission mode may be associated with a CSI reporting of a secondnumber of parts, and so forth. The parts associated with a TDMP transmission mode may be a common part and / or a specific part.

[0271] For example, when a WTRU 102 is configured with or determined to use a first TDMP transmission mode, the WTRU 102 may report a single part (e.g., specific part) corresponding to the CSI reporting time.

[0272] For example, when a WTRU 102 is configured with or determined to use a second TDMP transmission mode, the WTRU 102 may report more than one part (e.g., a common part and one or more specific parts).

[0273] As described herein, the term TDMP transmission mode may be interchangeably used with CSI reporting mode, CSI reporting configuration, CSI reporting set, CSI reporting contents, and / or CSI reporting subset. For example, a first TDMP transmission mode may use a full set of resource configurations associated with CSI reporting and a second TDMP transmission mode may use a subset of resource configurations assocaited with CSI reporting.

[0274] In certain representative embodiments, a WTRU 102 may determine a TDMP transmission mode for a CSI reporting at time n+T based on one or more of following: (i) channel correlation level, (ii) level of change for a common part, (iii) channel conditions (e.g., Doppler frequency, WTRU 102 speed, channel coherence time), (iv) link performance, and / or (v) intermediate KPI performance.

[0275] For example, a WTRU 102 may report one or more specific parts (e.g., without common part reporting) if a channel correlation level at a time n (e.g., the latest CSI reporting before n+T) and at a time n+T (e.g., the current CSI reporting) is higher than a threshold. For example, if the common part at time n and the common part at time n+T are different (e.g., the level of change is) within a threshold, a WTRU 102 may report one or more specific parts (e.g., without common part reporting). For example, if a WTRU’ s channel conditions meet one or more conditions, the WTRU 102 may determine to report both common part and specific part. Such conditions may include any of: (i) WTRU speed is higher than a threshold, (ii) channel coherence time is less than a threshold, (iii) doppler frequency is higher than a threshold, and / or (iv) CSI reporting periodicity / gap / time offset is larger than a threshold. For example, the threshold may be configured via a higher layer signaling or determined based on channel coherence time. For example, if a WTRU’s channel conditions meet one or more conditions, the WTRU 102 may determine to report the specific part only (e.g., without the common part). Such conditions may include any of: (i) WTRU speed is lower than a threshold, (ii) common part is reported previously, (iii) channel coherence time is larger than a threshold, and / or (iv) doppler frequency is smaller than a threshold. For example, the link performance may be determined for PDCCH and / orPDSCH transmissions. For example, the WTRU 102 may report both common and specific parts (or a WTRU 102 may switch to reporting both common and specific parts if the WTRU 102 reported a specific part only in the latest CSI reporting) when one or more of conditions are met. Such conditions may include any of (i) the WTRU 102 reported N consecutive HARQ-NACKs (or the number of HARQ-NACKs within a time window is larger than a threshold) for PDSCH receptions, (ii) scheduled MCS for a PDSCH has a gap from reported CQI larger than a threshold, (iii) scheduled MCS is lower than a threshold, (iv) reported CQI is lower than a threshold, and / or (v) CCE aggregation level of PDCCH for a received DCI is larger than a threshold. For example, the intermediate KPI performance may be determined in terms of SGCS values and / or NMSE values.

[0276] In certain representative embodiments, a WTRU 102 may report both common and specific parts (or a WTRU 102 may switch to reporting both common and specific parts if the WTRU 102 reported a specific part only in the latest CSI reporting), such as when the measured long-term SGCS value drops below a threshold. The long-term SGCS may be measured between the common part that is generated, transmitted and stored at both the WTRU 102 and gNB 180 at a time T, and the common part that is generated at time T+n but not transmitted to gNB 180.

[0277] In certain representative embodiments, a WTRU 102 may be indicated or configured with a TDMP transmission mode for a CSI reporting based on one or more following. A WTRU 102 may be indicated in a DCI which TDMP transmission mode to use for a CSI reporting. For example, for aperiodic and / or semi-persistent CSI reporting, the triggering DCI for the CSI reporting may indicate which TDMP transmission mode to use for the CSI reporting. When a WTRU 102 received a triggering of aperiodic CSI reporting with a TDMP transmission mode with a specific part only, the WTRU 102 may assume or use the common part which is reported in the latest CSI reporting of common part. For example, a WTRU 102 may be implicitly indicated which TDMP transmission mode to use based on one or more of following: (i) maximum payload size for the CSI reporting, (ii) scheduling parameters (e.g., MCS, number of resources, number of layers, etc.), (iii) reference signal configuration, and / or (iv) model-ID associated with the CSI reporting.

[0278] TDMP Indications

[0279] In certain representative embodiments, a WTRU 102 (e.g., configured) for TDMP compression may feed back (e.g., send) a CSI report. The CSI reporting may be performed based on the configured and / or determined TDMP feedback transmission mode. For example, for a TDMP transmission mode which is a multi -part mode, the WTRU 102 may report more than one part of the encoder output (e.g., common and specific parts of the latent structure). For example,for a TDMP transmission mode which is a single-part mode, the WTRU 102 may report (e.g., only) a specific part of the encoder output (e.g., the specific part of the latent structure). A single part mode CSI report may occupy a lower bit-width size as compared to a multi-part mode CSI report.

[0280] In certain representative embodiments, a WTRU 102 may be configured to determine and report the TDMP mode. For example, such as where the WTRU 102 is configured to report TDMP transmission mode, the WTRU 102 may indicate the selection of the TDMP mode with a flag. As an example, the flag may be a TDMP mode selection flag which may include one or more bits. As an example, the WTRU 102 may transmit TDMP mode selection information, such as a flag or bits, as part of the CSI report together with the compressed CSI feedback (e.g., which is determined based on the TDMP mode). The WTRU 102 may transmit the CSI feedback along with the TDMP mode feedback, such as via UCI, in a PUSCH resource, and / or in MAC CE.

[0281] In certain representative embodiments, a WTRU 102 may transmit TDMP mode selection information and compressed TDMP -based CSI feedback in separate CSI reports. The WTRU 102 may be configured to transmit TDMP mode selection periodically via UCI or semi-persistently via PUSCH or MAC / CE.

[0282] In certain representative embodiments, a WTRU 102 may be configured and / or determine to use (e.g., operate in) a single part TDMP mode. The WTRU 102 may determine that the difference between channel samples (or the latent outputs) is below a threshold, then the WTRU 102 may omit transmitting the CSI feedback corresponding to the single part of the encoder output. The WTRU 102 may feed back the information on the omission of the single part CSI report using a (e.g., single) bit field, such as Single Part Omit. For example, if the Single Part Omit is ‘0’ then the WTRU 102 feeds back the single part report, otherwise the WTRU 102 does not feed back the single part report. The field Single Part Omit may be transmitted along with or included in the CSI report.

[0283] Representative Embodiments for TSF / CSI Compression

[0284] In certain representative embodiments, a WTRU 102 may be configured with one or more parameters to generate CSI feedback (e.g., with an AI / ML model such as an AE). The feedback includes TDMP latents.

[0285] For example, the common part may capture the long-term and / or coarse variations in a given sequence of correlated samples. The common part may be sent once every N CSI feedback transmissions. The common part may be stored by the receiving node and reused (e.g., to generate feedback) for N consecutive measurement and / or feedback instances. The common part may be transmitted semi-persistently, periodically, or aperiodically.

[0286] For example, the specific part may capture the short-term and / or fine variations between (e.g., every) a pair of samples in a given sequence of correlated samples. The specific part may be sent in every CSI feedback transmission. The specific part may be transmitted periodically.

[0287] In certain representative embodiments, a WTRU 102 may receive configuration information for one or more parameters associated with TDMP CSI feedback.

[0288] For example, the configuration information may include any of the following. The WTRU 102 may be configured with a TDMP configuration and parameters thereof which may include any of (i) parts information (e.g., a number of parts / latents, size of each part, quantization associated with each part); (ii) information indicating TDMP mode activation (e.g., an implicit indication, such as an ID of a specific model with TDMP operation); (iii) TDMP feedback transmission modes (e.g., single-part mode where the WTRU 102 is configured to transmit CSI feedback with only one part, such as the specific part; multi-part mode where the WTRU 102 is configured to transmit CSI feedback with multiple parts, such as common and specific parts); (iv) a TDMP feedback transmission mode selection flag (e.g., to enable the WTRU 102 to select and / or determine and report a preferred TDMP feedback transmission mode); and / or (v) an input domain type (e.g., EV or full CSI). The WTRU 102 may be configured with one or more conditions and / or parameters for TSF parts updating which may include any of (i) a first performance metric threshold (e.g., a short-term SGCS threshold, such as between consecutively generated common latents / parts); (ii) a second performance metric threshold (e.g., a long-term SGCS threshold, such as between a first / active common part (where the active common part is one that is stored and used at the gNB 180 for reconstruction), and one or more generated / additional inactive common parts (where an inactive common part is one that is generated over time but is not yet fed back to the gNB 180 and / or used / stored at the gNB 180), NMSE ). The WTRU 102 may be configured with a reporting configuration (e.g., part-specific and / or part-common reporting resources, MCS and / or quantization associated with each part, and / or UL resources for each part).

[0289] The WTRU 102 may receive one or more CSI-RS transmissions and may estimate the channel. For example, the WTRU 102 may (e.g., optionally) store the measured CSI (e.g., full channel measurements).

[0290] The WTRU 102 may apply time-dependent multipart TSF compression by generating one or more parts / latents (e.g., common / long-term part and short-term-part) associated with the received CSI transmission. The compression may be based on the configured TDMP parameters (e.g., number of parts, size of parts, Model ID) and / or configured reporting (e.g., quantization, MCS, etc.). In some embodiments, the WTRU 102 may (e.g., optionally) store the generated TSFcommon part in a TSF latent buffer (e.g., based on a TDMP configuration, such as a part monitoring flag).

[0291] The WTRU 102 may generate time-dependent parts / latents, e.g., common part and specific part, associated with the received CSI-RS transmission based on the configured TDMP parameters (e.g., number of parts, size of each, etc.) and / or configured reporting (e.g., quantization, MCS, etc.) In some embodiments, the WTRU 102 may (e.g., optionally) store the generated common part (e.g., based on a TDMP configuration, such as a part monitoring flag).

[0292] The WTRU 102 may determine and / or select a preferred TDMP feedback transmission mode (e.g., for the next transmission). For example, the determination and / or selection may be based on preconfigured conditions (e.g., TDMP feedback transmission mode selection flag) and / or measurements. For example, the WTRU 102 may determine the TDMP feedback transmission mode based on one or more of the following: measured channel parameters, CSI measurements, PDSCH performance, feedback resources, and / or a network indication. For example, the TDMP feedback transmission mode may be determined based on measured channel parameters (e.g., Doppler spread and / or WTRU speed). For example, a multipart transmission mode may be selected where the estimated Doppler spread exceeds a certain configured threshold. For example, the TDMP feedback transmission mode may be determined based on CSI measurements, such as by comparing the measured long-term and / or short-term SGCS (e.g., full channel or latent) with the configured long / short-term SGCS threshold. For example, the multipart transmission mode may be selected where the measured long-term SGCS value drops below the configured threshold. For example, the TDMP feedback transmission mode may be determined based on PDSCH performance, such as in terms of a number of NACKs within a time interval. For example, the TDMP feedback transmission mode may be determined based on feedback resources (e.g., time, frequency, BW, and / or duration). For example, the TDMP feedback transmission mode may be determined based on an indication received from a gNB 180.

[0293] The WTRU 102 may report one or more of the following: (i) compressed CSI based on the configured TDMP feedback transmission mode; and / or (ii) the preferred TDMP feedback transmission mode (e.g., to be used in a next or future reporting instance(s)). In some embodiments, the WTRU 102 may receive information indicating any of a (re)configuration, an activation, and / or a deactivation (e.g., for a TDMP mode) for subsequent CSI reporting.

[0294] Procedure for Monitoring, Determining, and Improving the Validity Time of TDMP TSF / CSI Compression

[0295] Configurations for TSF Multipart Monitoring

[0296] Configuration of AI / ML Models with TDMP Latents

[0297] In certain representative embodiments, a WTRU 102 may be configured to generate one or more quantities associated with a CSI feedback report using an AI / ML model, such as a (e.g., configured) AE model. The AI / ML encoder model may be trained to generate a latent output with some enforced structure, such as where the structure may be preconfigured and / or predefined for the WTRU 102. For example, the output of the configured AI / ML encoder model may be structured to have two parts, namely one common part and one specific part. A common part may capture the common information in a sequence of temporally correlated (e.g., CSI) samples, such as N samples, and may be commonly used in reconstructing each of the samples. A specific part may capture sample-specific information, which may be (e.g., unique or sample-specific) information for each sample in the sequence and may be used along with the common information to reconstruct the desired sample. As described herein, an AI / ML model with common and specific parts structure may be referred to as a TDMP model, TDMP latent, and / or TDMP compression.

[0298] Configuration of the Generation of CSI Feedback Parts

[0299] In certain representative embodiments, a WTRU 102 may be pre-configured with one or more parameters to generate the CSI feedback parts, such as by using an AI / ML encoder model. The CSI feedback may include TDMP latents. For example, a pre-configuration may include a configuration of (e.g., for generating) the common part, which captures the long-term / coarse variations in a given sequence of correlated samples. In certain representative embodiments, the common part may be sent once every N CSI feedback transmissions. In certain representative embodiments, the common part be stored and reused for consecutive N slots. The configuration may (e.g., also) include parameters of (e.g., for generating) the specific part, which captures the short-term variations between a (e.g., every) pair of samples in a given sequence of correlated samples. For example, the configuration may be sent in every CSI feedback transmission or may not need to be stored.

[0300] Configuration of Monitoring, Controlling, and Updating of TDMP

[0301] In certain representative embodiments, a WTRU 102 may be pre-configured or may receive two sets of configurations (e.g., for monitoring and / or updating the TDMP feedback), one related to the TDMP itself, and another related to the conditions and parameters for monitoring and updating of the TSF parts.

[0302] For example, the configuration of the TDMP (e.g., and parameters thereof) may include TSF parts information, such as any of the number of parts in the latent domain, the size of each part, the quantization associated to each part, a common quantization for all the parts, and / or an indicator for using a different MCS for each part (e.g., for further reliability and protection against the feedback channel noise).

[0303] For example, the configuration may also include information (e.g., parameters) about the initial and / or the maximum validity time of the common part. The configuration may, for example, include an initial and / or maximum number of samples to use in the common part. In another example, the number of samples (e.g., in the buffer) may be static or dynamic. The configuration for the WTRU 102 to reuse, freeze, and / or reinitialize the common part may be based on a timer (e.g., within an interval of time units, such as may be based on a time-domain-resource assignment (TDRA) or seconds). For example, the timer may be either fixed or dynamic.

[0304] In certain representative embodiments, a correlation of the common parts and the validity time of the TSF parts buffer may be dependent on multiple WTRUs’ applicable conditions, such as channel conditions, speed, Doppler spread, and / or coherence time. For example, freezing and / or reinitializing the common part may be based on triggers, such as using a predefined threshold and / or measurements of the WTRU speed, Doppler spread, SINR, and / or coherence time. For example, the WTRU 102 may receive the configuration after assisting the NW by reporting any of these measurements. Further configuration information may include a monitoring resolution (e.g., the full common latent, per latent monitoring, or all latent domain, such as including the common and specific TSF parts of the CSI feedback).

[0305] In certain representative embodiments, configuration information may (e.g., also) include a TDMP mode activation flag (e.g., a new field in DCI, in RRC or MAC CE) which may indicate (e.g., implicitly) the ID of a specific AI / ML model for TDMP with multi-part operating mode, and / or a (e.g., binary) field associated to the model indicating if the model is activated. The configuration received by the WTRU 102 may (e.g., also) include or indicate the input domain type of the TDMP model, such as EV or full CSI input.

[0306] In certain representative embodiments, a WTRU 102 may also be pre-configured and or may receive a configuration (e.g., and parameters thereof) related to conditions and metrics for TSF parts performance monitoring and / or updates, such as based on KPIs. For example, a configuration may include short-term and / or long-term performance metric thresholds, such as one or multiple SGCS and / or NMSE thresholds, according to the resolution configuration. For example, for short-term monitoring, the metrics may be SGCS and / or NMSE differences between consecutive common latents / parts or during a short-term window. For example, monitoring may be based on the difference of performance, such as where the difference is compared to a threshold or comparing the KPI value and a KPI threshold received in the configuration. For long-term monitoring, as an example, the long-term SGCS or NMSE thresholds may be used for comparing the performance of the first and / or active common parts and additional and / or inactive common parts. For example, a combination of short-term and long-term monitoring may enable the WTRU102 to update the TSF buffer, and / or to freeze the buffer based on the resolution and the monitoring result.

[0307] In certain representative embodiments, a WTRU 102 may have to store a common part / latent buffer for the adaptation based on performance monitoring and according to the configuration. The WTRU 102 may receive in the configuration a part-monitoring flag which may indicate to start to monitor a particular part (e.g., common, or specific, or multiple parts). For example, this flag may be a binary sequence indexing the different parts.

[0308] According to the monitoring and resolution configuration information, the WTRU 102 may receive a configuration relative to a reporting aspect (e.g., part-specific and / or part-common). The configuration may, for example, include any of the MCS to use for each part, the quantization function and its parameters associated to each part, UL resources for each part, reporting periods, and / or intervals / delays (e.g., TDRA patterns or time difference, or trigger-based).

[0309] Procedures for Determination of TSF Multipart Monitoring Parameters

[0310] In certain representative embodiments, a WTRU 102 may be configured for TSF-based (e.g., CSI) compression using time dependent multi-parts, and may perform TSF compression. For example, the WTRU 102 may determine one or more parameters for calculating the common (long-term) and specific (short-term) parts of the compressed information (e.g., CSI).

[0311] In certain representative embodiments, the WTRU 102 may estimate the channel response, for example, based on the received CSI-RS. For example, the WTRU 102 may apply time dependent multi-part TSF compression to compress the full (e.g., raw) channel matrix. For example, the WTRU 102 may apply TDMP TSF compression to compress the eigenvectors of the channel response.

[0312] In certain representative embodiments, a WTRU 102 may apply TSF CSI compression to determine the specific (short-term) part of the compressed CSI (e.g., as configured). For example, the WTRU 102 may use a ML encoder model associated to a configured ML Model ID and part size, to determine the specific (short-term) part of the compressed CSI. The WTRU 102 may determine the specific (short-term) part of the compressed CSI for every CSLRS transmission occasion.

[0313] In certain representative embodiments, a WTRU 102 may apply TSF CSI compression to determine the common (long-term) part of the compressed CSI, such as for every N-th CSLRS transmission occasion where N is configured and may be greater to or equal to 1. In certain representative embodiments, a WTRU 102 may determine the common part when the WTRU 102 is triggered to determine the common part.

[0314] In certain representative embodiments, a WTRU 102 may determine one or more parameters for calculating the common (long-term) parts of the compressed CSI, where the parameters may include any of: common part validity time, common part update period, and / or common part update type.

[0315] Common Part Validity Time

[0316] In certain representative embodiments, a common part validity time may represent a duration the common part may be used (e.g., reused) for the CSI reconstruction at the decoder side (e.g., at the NW-side). For example, the validity time may be associated with a time for which the reconstruction performance meets a configured threshold. The common part validity time may be expressed in any units of time, such as a number of seconds, slots, transmission time intervals, CSI reporting samples, and / or CSI-RS transmission occasions.

[0317] In certain representative embodiments, a WTRU 102 may determine the common part validity time as a function of measured channel parameters, such as channel coherence time and / or estimated Doppler spread. For example, the common part validity time may be set to a fraction of the channel coherence time, such as where the fraction of the channel coherence time may be configured (e.g., via RRC configuration) statically or semi-statically. For example, when the WTRU 102 reports a change in the coherence time, such as a larger coherence time, the NW may reconfigure to a larger fraction of the channel coherence time which may reduce the feedback signaling overhead. For example, a lookup table may be used to map the measured values (e.g., estimated Doppler spread values) to a common part validity time. The lookup table may be predefined, and the WTRU 102 may report or otherwise signal an index to the lookup table to indicate the measured or recommended common part validity time.

[0318] In certain representative embodiments, a WTRU 102 may determine the common part validity time as a function of CSI measurements, such as CS, generalized CS (GCS), squared GCS (SGCS), and / or NMSE. For example, the WTRU 102 may be configured with a SGCS threshold measured in the uncompressed channel domain (e.g., raw channel matrix or eigenvectors) and may determine the common part validity time as a function of the difference between the current SGCS and the configured SGCS threshold. The current SGCS may be measured for the current channel and / or a reference channel. For example, the WTRU 102 may store the channel as a reference for SGCS calculation, when (e.g., every time) it updates the common part validity time. For example, the WTRU 102 may be configured with a SGCS threshold for the common part in the latent domain. For example, the WTRU 102 may be configured with a SGCS threshold for the specific part in the latent domain.

[0319] Common Part Update Period

[0320] In certain representative embodiments, a common part update period may be representative of the period the WTRU 102 may (e.g., should) calculate and report the common part, such as for periodic common part reporting. The WTRU 102 may be configured with an initial (e.g., default) common part update period (e.g., via RRC signaling) and may receive from the NW an indication of a new update period, for example, as a result of the WTRU 102 reporting an updated common part validity time.

[0321] Common Part Update Type

[0322] In certain representative embodiments, a common part update type may refer to the updating of the common part. For example, types of common part updates may include any of replacing, resetting, incremental updating, and / or model switching. For example, when the WTRU 102 sends a replace / reset message to the network, the WTRU 102 may indicate that for the network-side reconstruction of the CSI, the network may (e.g., should) replace the previous common part with the latest common part reported by the WTRU 102. For example, when the WTRU 102 sends an incremental update message to the network, the WTRU 102 may include a differential value of the common part (e.g., relative to a reference common part). The network may use the WTRU 102 reported differential value to update the existing network-side common part. For example, the WTRU 102 may transmit a model switch message to the network to request a switch of the AI / ML model (e.g., AE) used for common part compression, such as to achieve a higher resolution for the compressed common part.

[0323] Triggers for Measurements for Common Part Update

[0324] In certain representative embodiments, a WTRU 102 may be triggered to perform configured measurements for a common part update. For example, triggers may be time-based and / or event based.

[0325] For example, time-based triggers for measurements for the common part update may include any of: (i) an expiration of a timer for the common part update period, and / or (ii) expiration of a common part validity timer. As described herein, the timer may refer to a time duration or interval which may be expressed in units such as any of a number of seconds, slots, transmission time intervals, CSI reporting samples, and / or CSI-RS transmission occasions. Expiration of the time may refer to the time duration or interval lapsing.

[0326] For example, event based triggers for measurements for the common part update may include any of: (i) a change in configuration, which may include a change of BWP, change in the number of antenna ports, TRP change, cell activation / deactivation; (ii) a handover event; (iii) a radio link failure (RLF) event; (iv) a beam failure detection event; (v) a beam recovery event; (vi) an update of the beam pair or the DL beam; (vii) performance of the associated data channeldropping below a configured threshold (e.g., monitor the HARQ ACK / NACK statistics for the data channel, and trigger common part update measurements when the HARQ ACK rate is lower than a configured threshold); and / or (viii) a change in the measured channel parameters, such as the channel coherence time and / or the estimated Doppler spread.

[0327] Triggers for Sending Common Part Update Reporting

[0328] In certain representative embodiments, a WTRU 102 may be triggered to perform measurements for the common part update, and the WTRU 102 may measure any of the channel parameters, the full CSI, and / or measure the long-term latent part.

[0329] For example, when the WTRU 102 measures the full CSI, the WTRU 102 may compare the measured long-term SGCS against a configured threshold. For example, when the measured SGCS is smaller than a first SGCS threshold, the WTRU 102 may determine a new common part and send a “replace / resef ’ common part update type message to the network. As another example, when the measured SGCS is smaller than a second SGCS threshold and larger than the first SGCS threshold (e.g., the second SGCS threshold is larger than the first SGCS threshold), the WTRU 102 may determine a differential update of the (e.g., previously reported) common part. Here, the WTRU 102 may send an “incremental update” message to the network and may report the differential update for the common part.

[0330] For example, when the WTRU 102 performs measurements in the latent domain (e.g., in the compressed space), the WTRU 102 may measure the SGCS of the long-term (common) latent part and compare the current measured SGCS of the common latent part to the previously reported SGCS of the common latent part. For example, the WTRU 102 may report the current measured common latent part and send a “replace / resef ’ common part update type message to the network when the difference between the current and the previous SGCS exceeds a third threshold. As another example, the WTRU 102 may report the differential common latent part and send an “incremental update” common part update type message when the difference between the current and the previous SGCS exceeds a fourth threshold and is smaller than the third threshold (e.g., the third threshold is larger than the fourth threshold).

[0331] TDMP Multipart Indications

[0332] In certain representative embodiments, a WTRU 102 may be configured to indicate and / or report one or more parameters associated with (e.g., CSI) feedback operation, such as one or more TDMP (e.g., CSI) feedback transmissions. For example, TDMP CSI feedback may include a common part and a specific part. For example, the common part may be associated with the longterm channel properties and the specific part may be associated with the short-term channel properties. The WTRU 102 may be configured to monitor the performance of the common part.The WTRU 102 may be configured to report one or more parameters associated with common part reporting based on a performance monitoring result. Though the examples herein are described in terms of one common part, the disclosure can be extended for TDMP CSI feedback including more than one common part.

[0333] Reporting Validity of Common Part

[0334] In certain representative embodiments, a WTRU 102 may indicate a validity of the common part. For example, the validity may be expressed in terms of time units, such as a number of seconds (e.g., milliseconds), slots, transmissions, reports, and / or samples. For example, one or more validity values may be predefined to indicate a special or particular meaning. For example, the WTRU 102 may report a validity of 0 ms for the common part to indicate that there is no need for common part reporting, such as where the CSI feedback would consist of only specific part and no common part. For example, a reserved value may indicate that the CSI feedback would consist of only a common part and no specific part. For example, the validity may be expressed in terms of multiples of specific part reporting periodicity. For example, a value of n may indicate that a common part is reported for every n reports of specific parts. For example, a value of 3 may indicate that the common part is valid (from a reconstruction perspective) for the next 3 specific parts. For example, the validity of common part may be expressed as periodicity of common part reporting. For example, the WTRU 102 may be preconfigured with plurality of multiplicity factors with reference to a base periodicity of the common part. For example, the base periodicity may be n ms or n slots. The multiplicity factor may be 1 / 4, 1 / 2, 1, 2, 3, etc. For example, the WTRU 102 may be configured to indicate the validity of the common part by reporting the multiplicity factor. The actual validity may be determined as the multiplicity factor multiplied by the base periodicity of the common part. For example, the validity of common part may be expressed as a percentage. For example, the percentage may be calculated by a ratio of number of common part reports to the total number of CSI reports (e.g., including both common part and specific part) within a time interval.

[0335] Reporting Configuration / Parameterization of Common Part

[0336] In certain representative embodiments, a WTRU 102 may report one or more configurations and / or parameterizations of the common part, such that the common part reporting satisfies a preconfigured criteria. For example, the preconfigured criteria may be associated with performance of the common part. For example, the preconfigured criteria may be associated with the overhead of CSI reporting. For example, the preconfigured criteria may be associated with a payload size of CSI report. For example, the WTRU 102 may report one or more preferred configurations and / or parameterizations of the common part.

[0337] In certain representative embodiments, the configuration and / or parameterization of the common part may indicate a size of the common part. For example, the size of the common part may be determined based on a length of the latent vector at the output of the encoder. For example, a special value of zero as the size of the latent vector may indicate that the common part is not reported.

[0338] In certain representative embodiments, the configuration and / or parameterization of the common part may indicate the quantization applied to the common part. For example, the quantization of the common part may be determined based on a bit width of the latent vector at the output of the encoder.

[0339] Reporting Performance of Common Part

[0340] In certain representative embodiments, a WTRU 102 may report a performance metric associated with common part reporting. For example, the performance metric may be determined based on any of the methods described herein. For example, the WTRU 102 may report a performance metric associated with the currently reported common part. For example, the performance metric of the common part may include one or more of the short-term SGCS and / or NMSE between consecutive common parts, short term SGCS and / or NMSE between a common part and one or more specific parts, etc.

[0341] In certain representative embodiments, a WTRU 102 may report a performance metric associated with (e.g., hypothetical) common part reporting. For example, the performance metric may be determined based on one more methods described herein. For example, the WTRU 102 may report the performance metric associated with common part reporting assuming such common part reporting is configured based on a WTRU-preferred configuration and / or parameterization of the common part. For example, the performance metric of the common part may include one or more of the SGCS and / or NMSE between a hypothetical common part and currently reported common part, a short term SGCS and / or NMSE between the hypothetical common part and one or more specific parts, etc.

[0342] In certain representative embodiments, a WTRU 102 may be configured with periodic (e.g., PUCCH and / or PUSCH) resources for reporting the validity, configuration, parameterization and / or performance of the common part. For example, the WTRU 102 may be configured to report the validity, configuration, parameterization and / or performance of the common part based on an aperiodic request from the gNB 180. For example, the WTRU 102 may be configured to report the validity, configuration, parameterization and / or performance of the common part in a MAC CE.

[0343] Reporting Additional Information Associated With Common Part

[0344] In certain representative embodiments, a WTRU 102 may include additional information in the common part report. The additional information may be associated with or indicate a property of the common part report. For example, the additional information may include an update type. For example, the update type may indicate how the receiver of the CSI feedback may (e.g., should) process the common part, such as in relation to a previously transmitted common part. For example, the update type may indicate that the transmitted common part may (e.g., should) replace any previously reported common part. In this case, the update type may be set to a ‘replace’ indication. For example, the update type may indicate that the transmitted common part is a differential value in relation to a previously reported common part. In this case, the transmitted common part indicates (e.g., only) changes in the channel compared to the previously reported common part. In such cases, the update type may be set to a ‘delta’ or ‘differential’ or ‘incremental’ indication.

[0345] In certain representative embodiments, a WTRU 102 may indicate different levels of granularity for the update type. For example, the update type may correspond to the common part as a whole. For example, different update types may be indicated for different subsets of the common part. For example, the subsets may be determined dynamically by the WTRU 102 based on one or more preconfigured conditions. For example, the subsets may be preconfigured for the WTRU 102.

[0346] In certain representative embodiments, a WTRU 102 may be configured to report additional information along with the CSI report carrying the common part. For example, the WTRU 102 may be configured to report additional information in a part (e.g., part 1) of the CSI report and the common part in another part (e.g., part 2) of the CSI report. For example, the WTRU 102 may be configured with periodic (e.g., PUCCH and / or PUSCH) resources for reporting the additional information associated with a common part report. For example, the WTRU 102 may be configured to report the additional information associated with common part report based on an aperiodic request from the gNB 180.

[0347] Representative Embodiments for Validity Time of TDMP TSF / CSI Compression

[0348] In certain representative embodiments, a WTRU 102 may be configured with one or more parameters to generate CSI feedback (e.g., with an AI / ML model such as an AE). The feedback includes TDMP latents.

[0349] For example, the common part may capture the long-term and / or coarse variations in a given sequence of correlated samples. The common part may be sent once every N CSI feedback transmissions. The common part may be stored by the receiving node (e.g., network -side decoder)and reused for reconstruction for N consecutive slots. The common part may also be stored at the transmitting node (e.g., WTRU-side) for monitoring purposes.

[0350] For example, the specific part may capture the short-term and / or fine variations between (e.g., every) a pair of samples in a given sequence of correlated samples. The specific part may be sent in every CSI feedback transmission. The specific part may be transmitted periodically.

[0351] The WTRU 102 may receive configuration information for one or more parameters associated with monitoring and / or controlling one or more aspects associated with the timedependent multipart CSI feedback.

[0352] For example, the configuration information may include any of the following. The WTRU 102 may be configured with a TDMP configuration and parameters thereof which may include any of: (i) parts information (e.g., number of parts / latents, size of each part, quantization associated w / each part); (ii) an initial and / or max validity time and / or BW of the common part, such as information indicating the initial / maximum number of samples to be used for a common part or time (e.g., in seconds to freeze / reuse the common part); (iii) a monitoring resolution (e.g., full common / all latents monitoring or per-latent monitoring, all latents monitoring); (iv) information indicating a TDMP mode activation (e.g., may be implicit, such as the ID of a specific TDMP model with multi-part operation); and / or (v) an input domain type (e.g., EV or full CSI). The WTRU 102 may be configured with one or more conditions and / or parameters for TSF parts updating and / or monitoring which may include any of: (i) a first performance metric threshold (e.g., Short-term SGCS / NMSE threshold: between consecutively generated common latents / parts); (ii) a second performance metric threshold, e.g., Long-term SGCS / NMSE threshold: between first / active common part, such as where the active common part is one that is stored and used at the gNB 180, such as for reconstruction, and generated / additional inactive common parts such as where an inactive common part is one that is generated over time but not fed back to the gNB 180 and not used / stored at the gNB 180; (iii) a part monitoring flag, such as to indicate to the WTRU 102 to start monitoring one of the parts (e.g., common) performance and / or indicate to the WTRU 102 to start storing the common part / latent in a buffer. The WTRU 102 may be configured with a reporting configuration, such as part-specific and / or part-common reporting aspects (e.g., MCS and / or quantization associated with each part, UL resources for each part).

[0353] The WTRU 102 may receive one or more CSI-RS transmissions and may estimate the channel. In some embodiments, the WTRU 102 may (e.g., optionally) store the measured CSI (e.g., full channel measurements).

[0354] The WTRU 102 may apply time-dependent multipart TSF compression by generating one or more parts / latents (e.g., common / long-term part and short-term-part) associated with thereceived CSI transmission. The compression may be based on the configured TDMP parameters (e.g., number of parts, size of parts, Model ID) and / or configured reporting (e.g., quantization, MCS, etc.). In some embodiments, the WTRU 102 may (e.g., optionally) store the generated (e.g., TSF) common part, such as in a TSF latent buffer (e.g., based on a TDMP configuration, such as a part monitoring flag).

[0355] The WTRU 102 may determine one or more parameters associated with the common part, such as a validity time and / or life-cycle (e.g., in seconds or number of samples), a validity BW, and / or an update request (e.g., common part refinement) based on preconfigured conditions and measurements. For example, a WTRU 102 may determine the validity time associated with the common part, (e.g., the validity time is the estimated time for the common part to work properly before the performance drops below a certain level). The validity time may be based on one or more of the following: measured channel parameters, CIS measurements, and / or PDSCH performance. For example, the measured channel parameters may include Doppler spread and / or WTRU speed, such as where the long-term / common part validity time is determined based on the estimated Doppler spread with each Doppler spread (e.g., a value and / or a range) mapping to a specific validity time. For example, the WTRU 102 may compare the measured long-term and / or short-term SGCS with the configured long / short-term SGCS threshold and map the difference to a specific validity time. Such measurements may be either in the full channel domain or in the latent domain associated with the common part. For example, the WTRU 102 may determine the PDSCH performance based on (e.g., in terms of) a number of NACKs within a time interval.

[0356] The WTRU 102 may determine if common part update feedback is needed (e.g., the update may be in one of different formats; replace / reset request, incremental update request by sending the differential common value, model switch for higher resolution common part) based on one or more of the following: based on full CSI measurements, long-term latent measurements, and / or other triggers. For example, common part update feedback may be determined periodically by comparing the measured long-term SGCS with the configured long-term SGCS threshold. As an example, the measured SGCS is below the first configured threshold, the WTRU 102 may indicate a replacement of the common part. In another example, if the measured SGCS is below a second threshold, the WTRU 102 may indicate an incremental update through sending a differential value for the common part (e.g., of one or more latents). For example, common part update feedback may be determined based on long-term latent measurements, such as where there is a discrepancy in the long-term latent. The WTRU 102 may compare the measured long-term SGCS between the stored long-term part at the gNB 180 and the current generated common part with one or more threshold(s) and send an update request accordingly.

[0357] The WTRU 102 may report one or more of the following: (i) the compressed CSI with one or more time-dependent parts; and / or (ii) parameters associated with the monitoring performance of the common part (e.g., validity time or BW, update type such as replace, incremental update or no update).

[0358] FIG. 11 is a flow diagram illustrating an example procedure for time-dependent multipart feedback transmission. In certain representative embodiments, a WTRU 102 may operate as an encode-side device and another device (e.g., another WTRU 102 or a base station, such as a gNB 180) may operate as a decode-side device, or vice versa. As shown in FIG. 11, a WTRU 102 may receive configuration information associated with RS feedback (e.g., CSI reporting) at 1102. For example, the WTRU may receive any of reporting setting information, measurement setting information, resource setting information, TDMP generation information and the like as described herein. The WTRU 102 may proceed for a reporting period N at 1104 to measure one or more RSs at 1106. At 1108, the WTRU 102 may encode measurement information associated with the measured one or more RSs using an AI / ML model (e.g., an autoencoder) to generate a latent structure having a common part (e.g., zcin FIG. 3) and a specific part (e.g., zi in FIG. 3). At 1110, the WTRU 102 may transmit any of the common part and / or the specific part of the generated latent structure (e.g., for reporting period N) based on the configuration information, such as the configured and / or activated TDMP transmission mode. At 1112, the WTRU 102 may move to processing for a next reporting period by incrementing V+l.

[0359] FIG. 12 is a flow diagram illustrating another example procedure for time-dependent multipart feedback transmission. In certain representative embodiments, a WTRU 102 may operate as an encode-side device and another device (e.g., another WTRU 102 or a base station, such as a gNB 180) may operate as a decode-side device, or vice versa. As shown in FIG. 12, a WTRU 102 may receive configuration information associated with RS feedback (e.g., CSI reporting) at 1202. At 1204, the WTRU 102 may measure, at a plurality of times, one or more RSs. At 1206, the WTRU 102 may generate a plurality of latent structures, each including a common part and a specific part, using an AI / ML model that encodes feedback information associated with the measured RSs based on the configuration information. At 1208, the WTRU may send a plurality of transmissions which include the plurality of latent structures, where each transmission respectively includes any of the common part and / or the specific part of one of the latent structures based on one or more conditions.

[0360] For example, the plurality of latent structures may be generated at 1206 as a sequence (e.g., each latent structure corresponding to a reporting period). At least one of the plurality oftransmissions may include a preferred feedback mode, as described herein, associated with a subsequent transmission of the plurality of transmissions.

[0361] For example, the WTRU 102 may determine the preferred feedback mode using any of the techniques described herein. As an example, the preferred feedback mode may be determined based on any of the measured RSs, communication performance, time and / or frequency resources associated with any of the plurality of transmissions, and / or an indication received from an entity associated with an AI / ML model that decodes the plurality of transmissions.

[0362] For example, the common part of a respective latent structure may include encoded information associated with the measured RSs over the plurality of times.

[0363] For example, the specific part of a respective latent structure may include encoded information associated with the measured RSs at a respective one of the plurality of times.

[0364] For example, the sending at 1208 may include the WTRU 12 sending a first transmission which includes the common part and the specific part of a first one of the latent structures based on any of a periodicity for the RS feedback, reception of an indication from an entity associated with an AI / ML model that decodes the plurality of transmissions, lapsing of a time duration (e.g., expired validity) associated with a previously sent common part, a performance metric associated with the common part included in the first transmission and the previously sent common part, and / or an input format of the AI / ML model.

[0365] FIG. 13 is a flow diagram illustrating another example procedure for controlling the validity of a common part used in time-dependent multipart feedback transmission. In certain representative embodiments, a WTRU 102 may operate as an encode-side device and another device (e.g., another WTRU 102 or a base station, such as a gNB 180) may operate as a decodeside device, or vice versa. As shown in FIG. 13, a WTRU 102 may receive configuration information associated with reference signal (RS) feedback at 1302. At 1304, the WTRU 102 may measure, at a plurality of times, one or more RSs. At 1306, the WTRU may generate a plurality of latent structures, each including a common part and a specific part, using an AI / ML model that encodes feedback information associated with the measured RSs based on the configuration information. At 1308, the WTRU 102 may send a first transmission which includes a first latent structure of the plurality of latent structures. The first transmission may include the common part and the specific part of the first latent structure based on one or more conditions as described herein. At 1310, the WTRU 102 may determine a validity associated with the common part of the first latent structure. At 1312, the WTRU 102 may send a second transmission which includes a second latent structure of the plurality of latent structures. The second transmission may includeinformation associated with the determined validity and the specific part of the second latent structure based on the one or more conditions.

[0366] For example, the determining of the validity at 1310 may use any of the techniques described herein. As example, the validity may be based on any of one or more measured channel parameters, the measured RSs, reception of an indication from an entity associated with an AI / ML model that decodes the first transmission, and / or lapsing of a time duration associated with a previously sent common part.

[0367] For example, the information associated with the determined validity may include the common part of the second latent structure (e.g., to replace, reset, and / or update a stored common part at the decode side).

[0368] For example, the information associated with the determined validity may include a differential between the common part of the first latent structure and the common part of the second latent structure (e.g., to replace, reset, and / or update a stored common part at the decode side).

[0369] For example, the information associated with the determined validity may include an indication of an update type (e.g., to replace, reset, and / or update) associated with the common part of the first latent structure.

[0370] For example, the plurality of latent structures are generated as a sequence. The common part of the first latent structure includes encoded information associated with the measured RSs over the plurality of times. The respective specific parts of the first and second latent structures may include encoded information associated with the measured RSs at different ones of the plurality of times.

[0371] FIG. 14 is a flow diagram illustrating another example procedure for TDMP feedback transmission, according to one or more embodiments of the present disclosure.

[0372] In certain representative embodiments, a WTRU 102 may operate as an encode-side device and another device (e.g., another WTRU 102 or a base station, such as a gNB 180) may operate as a decode-side device, or vice versa. As shown in FIG. 14, a WTRU 102 (e.g., as the encode-side device) may receive (e.g., from the decode-side) configuration information indicating one or more parameters associated with CSI feedback of TDMP latents at 1402. For example, the one or more parameters may include a set of TDMP feedback transmission modes, such as a singlepart TDMP feedback mode and a multi -part TDMP feedback mode. At 1404, the WTRU 102 may receive one or more transmissions of one or more CSI-RSs (or other RSs to be measured). At 1406, the WTRU 102 may generate, using an AI / ML model (e.g., an autoencoder), a set of TDMP latents. For example, the set of TDMP latents may be based on measurement information (e.g., as input to the AI / ML model) associated with the received one or more CSI-RSs. For example, theset of TDMP latents may include a plurality of parts (e.g., common and specific parts). At 1408, the WTRU 102 may determine a TDMP feedback transmission mode (e.g., single-part TDMP feedback mode and multi-part TDMP feedback mode) from the set of TDMP feedback transmission modes. For example, the determination of the TDMP feedback mode may be dependent on one or more conditions as described herein. At 1410, the WTRU 102 may send (e.g., to the decoding-side) reporting information indicating (i) at least a part of the set of TDMP latents and (ii) the determined TDMP feedback transmission mode.

[0373] For example, a common part of at least one TDMP latent of the set of TDMP latents may include information associated with long-term variations between correlated samples included in the measurement information.

[0374] For example, a specific part of at least one (e.g., each) TDMP latent of the set of TDMP latents may include information associated with short-term variations between pairs of correlated samples included in the measurement information.

[0375] For example, the configuration information may include a TDMP configuration (e.g., an indication thereof). The one or more parameters included in the TDMP configuration may include any of (i) a quantity of the parts in the set of TDMP latents, (ii) a quantity of TDMP latents in the set of TDMP latents, (iii) a size of a common part and / or a size of a specific part of the set of TDMP latents, and / or (iv) a quantization level associated with the common part and / or the specific part.

[0376] For example, the WTRU 102 may (e.g., sequentially) generate the set of TDMP latents based on the measurement information and the one or more parameters included in the TDMP configuration.

[0377] For example, the configuration information may include a reporting configuration (e.g., an indication thereof). The one or more parameters included in the reporting configuration may include any of (i) a quantization level associated with common part reporting and / or specific part reporting, (ii) a MCS associated with any of common part reporting and / or specific part reporting, and / or (iii) UL resources associated with any of common part reporting and / or specific part reporting.

[0378] For example, the WTRU 102 may send the reporting information based on the one or more parameters included in the reporting configuration.

[0379] For example, the configuration information may include a TDMP configuration (e.g., an indication thereof). The one or more parameters included in the TDMP configuration may include an indication to activate a TDMP feedback transmission mode of (e.g., from) the set of TDMPfeedback transmission modes. As an example, the WTRU 102 may send the reporting information using the activated TDMP feedback transmission mode.

[0380] For example, the configuration information may include a TDMP configuration (e.g., an indication thereof). For example, the one or more parameters included in the TDMP configuration may include an indication to determine the TDMP feedback transmission mode from the set of TDMP feedback transmission modes (e.g., at 1408). The determined TDMP feedback transmission mode may be associated with a future (e.g., predetermined or preconfigured) reporting instance for (e.g., at least a portion of) the set of TDMP latents.

[0381] For example, the WTRU 102 may determine the TDMP feedback transmission mode from the set of TDMP feedback transmission modes (e.g., at 1408) based on any of (i) the measurement information exceeding a first threshold, (ii) at least the part of the set of TDMP latents exceeding a second threshold, (iii) downlink channel performance, (iv) resources used to send the reporting information, and / or (v) a network indication.

[0382] FIG. 15 is a flow diagram illustrating an example procedure for TDMP feedback reception, according to one or more embodiments of the present disclosure. In certain representative embodiments, a WTRU 102 may operate as an encode-side device and another device (e.g., another WTRU 102 or a base station, such as a gNB 180) may operate as a decodeside device, or vice versa. As shown in FIG. 15, a base station (e.g., gNB 180) may send, to a WTRU 102, configuration information indicating one or more parameters associated with CSI feedback of TDMP latents at 1502. For example, the one or more parameters may include a set of TDMP feedback transmission modes, such as a single-part TDMP feedback mode and a multi-part TDMP feedback mode. At 1504, the base station may send, to the WTRU 102, one or more transmissions of one or more CSI-RSs. For example, the WTRU 102 may determine measurement information based on reception (e.g., sampling) of the CSI-RSs. At 1506, the base station may receive, from the WTRU 102, reporting information indicating (i) at least a part of a set of TDMP latents and (ii) a WTRU-determined TDMP feedback transmission mode (e.g., the single-part TDMP feedback mode or the multi-part TDMP feedback mode). For example, the set of TDMP latents may be based on (e.g., encoded) measurement information associated with the one or more transmissions (e.g., CSI-RSs). For example, the set of TDMP latents may include a plurality of parts (e.g., common and specific parts).

[0383] For example, a common part of at least one TDMP latent of the set of TDMP latents may include information associated with long-term variations between correlated samples included in the measurement information.

[0384] For example, a specific part of at least one TDMP latent of the set of TDMP latents may include information associated with short-term variations between pairs of correlated samples included in the measurement information.

[0385] For example, the configuration information may include a TDMP configuration. The one or more parameters included in the TDMP configuration may include any of (i) a quantity of parts in the set of TDMP latents, (ii) a quantity of TDMP latents in the set of TDMP latents, (iii) a size of a common part and / or a size of a specific part of the set of TDMP latents, and / or (iv) a quantization level associated with the common part and / or the specific part.

[0386] For example, the set of TDMP latents may be generated by the WTRU 102 (e.g., using an AI / ML model) based on the measurement information and the one or more parameters included in the TDMP configuration.

[0387] For example, the configuration information may include a reporting configuration. The one or more parameters included in the reporting configuration may include any of (i) a quantization level associated with common part reporting and / or specific part reporting, (ii) a modulation and coding scheme (MCS) associated with common part reporting and / or specific part reporting, and / or (iii) uplink resources associated with common part reporting and / or specific part reporting.

[0388] For example, the reporting information may be sent by the WTRU 102 based on the one or more parameters included in the reporting configuration.

[0389] For example, the configuration information may include a TDMP configuration. The one or more parameters included in the TDMP configuration may include an indication to activate a TDMP feedback transmission mode of (e.g., from) the set of TDMP feedback transmission modes. For example, the reporting information sent by the WTRU 102 may use the activated TDMP feedback transmission mode.

[0390] For example, the configuration information may include a TDMP configuration. The one or more parameters included in the TDMP configuration may include an indication to determine a TDMP feedback transmission mode from the set of TDMP feedback transmission modes. For example, the WTRU-determined TDMP feedback transmission mode may be associated with a future (e.g., predetermined or preconfigured) reporting instance for the set of TDMP latents.

[0391] For example, the WTRU-determined TDMP feedback transmission mode may be associated with any of the measurement information exceeding a first threshold, at least the part of the set of TDMP latents exceeding a second threshold, downlink channel performance, resources used to send the reporting information, and / or a network indication.

[0392] FIG. 16 is a flow diagram illustrating another example procedure for TDMP feedback transmission, according to one or more embodiments of the present disclosure. In certain representative embodiments, a WTRU 102 may operate as an encode-side device and another device (e.g., another WTRU 102 or a base station, such as a gNB 180) may operate as a decodeside device, or vice versa. As shown in FIG. 16, a WTRU 102 at 1602 may receive configuration information indicating one or more parameters associated with CSI feedback of TDMP latents. The WTRU 102 may generate a set of TDMP latents based on (e.g., encoding) measurement information associated with received signals at 1604. For example, the set of TDMP latents may include a plurality of parts (e.g., common and specific parts). The WTRU 102 may determine a TDMP feedback transmission mode from the one or more parameters (e.g., based on one or more conditions) at 1606. At 1608, the WTRU 102 may send reporting information indicating (i) at least a part of the set of TDMP latents and / or (ii) the determined TDMP feedback transmission mode.

[0393] For example, a common part of at least one TDMP latent of the set of TDMP latents may include information associated with long-term variations between correlated samples included in the measurement information.

[0394] For example, a specific part of at least one TDMP latent of the set of TDMP latents may include information associated with short-term variations between pairs of correlated samples included in the measurement information.

[0395] For example, the configuration information may include a TDMP configuration. The one or more parameters included in the TDMP configuration may include any of (i) a quantity of parts in the set of TDMP latents, (ii) a quantity of TDMP latents in the set of TDMP latents, (iii) a size of a common part and / or a size of a specific part of the set of TDMP latents, and / or (iv) a quantization level associated with the common part and / or the specific part.

[0396] For example, the WTRU 102 may generate the set of TDMP latents based on the measurement information and the one or more parameters included in the TDMP configuration.

[0397] For example, the configuration information may include a reporting configuration. The one or more parameters included in the reporting configuration may include any of (i) a quantization level associated with common part reporting and / or specific part reporting, (ii) a modulation and coding scheme (MCS) associated with common part reporting and / or specific part reporting, and / or (iii) uplink resources associated with common part reporting and / or specific part reporting.

[0398] For example, the WTRU 102 may send the reporting information based on the one or more parameters included in the reporting configuration.

[0399] For example, the configuration information includes a TDMP configuration. The one or more parameters included in the TDMP configuration may include an indication to activate a TDMP feedback transmission mode. The WTRU 102 may send the reporting information (e.g., at 1608) using the activated TDMP feedback transmission mode.

[0400] For example, the configuration information may include a TDMP configuration. The one or more parameters included in the TDMP configuration may include an indication to determine the TDMP feedback transmission mode. The determined TDMP feedback transmission mode my be a associated with a future (e.g., predetermined or preconfigured) reporting instance for the set of TDMP latents.

[0401] For example, the WTRU 102 may determine the TDMP feedback transmission mode from a set of TDMP feedback transmission modes based on any of the measurement information exceeding a first threshold, at least the part of the set of TDMP latents exceeding a second threshold, downlink channel performance, resources used to send the reporting information, and / or a network indication.

[0402] FIG. 17 is a flow diagram illustrating an example procedure for TDMP feedback reception, according to one or more embodiments of the present disclosure. In certain representative embodiments, a WTRU 102 may operate as an encode-side device and another device (e.g., another WTRU 102 or a base station, such as a gNB 180) may operate as a decodeside device, or vice versa. As shown in FIG. 17, a base station (e.g., gNB 180) may send (e.g., to the WTRU 102) configuration information indicating one or more parameters associated with CSI feedback of TDMP latents at 1702. At 1704, the base station may receive (e.g., from the WTRU 102) reporting information indicating (i) at least a part of a set of TDMP latents and / or (ii) a WTRU-determined TDMP feedback transmission mode. For example, the set of TDMP latents may be based on (e.g., generated by an AI / ML model at the WTRU 102 using) measurement information associated with one or more transmissions (e.g., from the base station). The set of TDMP latents may include a plurality of parts (e.g., common and specific parts).

[0403] For example, a common part of at least one TDMP latent of the set of TDMP latents may include information associated with long-term variations between correlated samples included in the measurement information.

[0404] For example, a specific part of at least one TDMP latent of the set of TDMP latents may include information associated with short-term variations between pairs of correlated samples included in the measurement information.

[0405] For example, the configuration information may include a TDMP configuration. The one or more parameters included in the TDMP configuration may include any of (i) a quantity of partsin the set of TDMP latents, (ii) a quantity of TDMP latents in the set of TDMP latents, (iii) a size of a common part and / or a size of a specific part of the set of TDMP latents, and / or (iv) a quantization level associated with the common part and / or the specific part.

[0406] For example, the set of TDMP latents may be generated by the WTRU 102 based on the measurement information and the one or more parameters included in the TDMP configuration.

[0407] For example, the configuration information may include a reporting configuration. The one or more parameters included in the reporting configuration may include any of (i) a quantization level associated with common part reporting and / or specific part reporting, (ii) a modulation and coding scheme (MCS) associated with common part reporting and / or specific part reporting, and / or (iii) uplink resources associated with common part reporting and / or specific part reporting.

[0408] For example, the reporting information may be sent by the WTRU 102 to the base station based on the one or more parameters included in the reporting configuration.

[0409] For example, the configuration information may include a TDMP configuration. The one or more parameters included in the TDMP configuration may include an indication to activate a TDMP feedback transmission mode. The reporting information may be sent to the base station by the WTRU 102 using the activated TDMP feedback transmission mode.

[0410] For example, the configuration information may include a TDMP configuration. The one or more parameters included in the TDMP configuration may include an indication to determine a TDMP feedback transmission mode. The WTRU-determined TDMP feedback transmission mode (e.g., in the reporting information at 1704) may be associated with a future reporting instance for the set of TDMP latents.

[0411] For example, the WTRU-determined TDMP feedback transmission mode may be (e.g., determined as) one of a set of TDMP feedback transmission modes. The WTRU-determined TDMP feedback transmission mode may be based on any of the measurement information exceeding a first threshold, at least the part of the set of TDMP latents exceeding a second threshold, downlink channel performance, resources used to send the reporting information, and / or a network indication.

[0412] In certain representative embodiments, a WTRU 102 may receiving configuration information associated with RS feedback. The WTRU 102 may measure, at a plurality of times, one or more (e.g., received) RSs. The WTRU 102 may generate a plurality (e.g., set of) of latent structures using an AI / ML model (or autoencoder) that encodes feedback information associated with the measured RSs based on the configuration information. For example, a (e.g., each) latent structure may include a common part and a specific part. The WTRU 102 may send a plurality oftransmissions which include (e.g., at least a part of) the plurality of latent structures. For example, a (e.g., each) transmission may respectively include any of the common part and / or the specific part of one of the latent structures based on one or more conditions.

[0413] For example, the plurality of latent structures are generated as a sequence, and wherein at least one of the plurality of transmissions includes a preferred feedback mode associated with a subsequent transmission of the plurality of transmissions.

[0414] For example, the WTRU 102 may determine the preferred feedback mode based on any of the measured RSs, communication performance, time and / or frequency resources associated with any of the plurality of transmissions, and / or an indication received from an entity (e.g. base station or other network entity) associated with an AI / ML model that decodes the plurality of transmissions.

[0415] For example, a common part may include encoded information associated with the measured RSs over the plurality of times.

[0416] For example, a specific part may include encoded information associated with the measured RSs at a respective one of the plurality of times.

[0417] For example, the sending of the plurality of transmissions by the WTRU may include sending a first transmission which includes the common part and the specific part of a first one of the latent structures based on any of a periodicity for the RS feedback, reception of an indication from an entity associated with an AI / ML model that decodes the plurality of transmissions, lapsing of a time duration associated with a previously sent common part, a performance metric associated with the common part included in the first transmission and the previously sent common part, and / or an input format of the AI / ML model.

[0418] One or more embodiments provide a computer program comprising instructions which when executed by one or more processors cause such processors to perform the encoding and / or decoding methods according to any of the embodiments described above. One or more embodiments also provide a computer readable storage medium having stored thereon instructions for encoding or decoding video data according to the methods described above.

[0419] One or more embodiments provide a computer readable storage medium having stored thereon video data generated according to the methods described above. One or more embodiments also provide a method and apparatus for transmitting or receiving video data generated according to the methods described above.

[0420] The embodiments described herein may be implemented in, for example, a method or a process, an apparatus, a software program, a data stream, or a signal. Even if only discussed in the context of a single form of implementation (e.g., as a method), the implementation of suchfeatures may also be implemented in other forms. An apparatus may be implemented in, for example, appropriate hardware, software, and firmware. Corresponding methods may be implemented in, for example, a processor.

[0421] Various numeric values are used in the present application. Such specific values are for example purposes and the embodiments described are not limited to these specific values.

[0422] Various methods are described herein, and such methods comprise one or more steps or actions for achieving the described method. Unless a specific order of steps or actions is required for the proper operation of the method, the order and / or use of specific steps and / or actions may be modified or combined. Additionally, terms such as “first”, “second”, etc. may be used in various embodiments to modify an element, component, step, operation, etc., for example, a “first decoding” and a “second decoding”. Use of such terms does not imply an order to the operations unless specifically required.

[0423] The present disclosure may refer to “determining” various pieces of information. Determining information may include one or more of, for example, estimating, calculating, predicting, or retrieving (e.g., from memory) the information.

[0424] The present disclosure may refer to “accessing” various pieces of information. Accessing information may include one or more of, for example, receiving, retrieving (e.g., from memory), storing, moving, copying, calculating, determining, predicting, or estimating the information. Similarly, the present disclosure may refer to “receiving” various pieces of information. Receiving information may include one or more of, for example, accessing or retrieving (e.g., from memory) the information.

[0425] It is to be understood that use of any of the following “ / ”, “and / or”, and “at least one of’ is intended to encompass all possible selections of listed items, taken either individually or in any combination thereof.

[0426] While specific embodiments have been described in the foregoing description in connection with the accompanying drawings, it should be understood that embodiments described herein are examples only and should not be taken as limiting the scope of the present disclosure or the following claims. Although features and elements are described herein in particular combinations, those of ordinary skill in the art will appreciate that such features or elements may be used alone or in any combination with the other features and elements. It is understood, therefore, that the overall teachings of the present disclosure are not limited to the particular embodiments, implementations, and examples disclosed herein, but are intended to cover variations, modifications, and alternatives as defined by the appended claims and any and all equivalents thereof.

Claims

CLAIMSWhat is claimed is:

1. A wireless transmit / receive unit (WTRU) comprising: a processor, memory, and a transceiver which are configured to: receive configuration information indicating one or more parameters associated with channel state information (CSI) feedback of time-dependent multi-part (TDMP) latents, wherein the one or more parameters includes a set of TDMP feedback transmission modes including a single-part TDMP feedback mode and a multi-part TDMP feedback mode, receive one or more transmissions of one or more CSI reference signals (CSI-RSs), generate, using an artificial intelligence / machine learning (AI / ML) model, a set of TDMP latents based on measurement information associated with the received one or more CSI-RSs, and wherein the set of TDMP latents includes a plurality of parts, determine a TDMP feedback transmission mode from the set of TDMP feedback transmission modes, and send reporting information indicating (i) at least a part of the set of TDMP latents and (ii) the determined TDMP feedback transmission mode.

2. The WTRU of claim 1, wherein a common part of at least one TDMP latent of the set of TDMP latents includes information associated with long-term variations between correlated samples included in the measurement information.

3. The WTRU of any of claims 1-2, wherein a specific part of at least one TDMP latent of the set of TDMP latents includes information associated with short-term variations between pairs of correlated samples included in the measurement information.

4. The WTRU of any of claims 1-3, wherein the configuration information includes a TDMP configuration, and the one or more parameters included in the TDMP configuration include any of (i) a quantity of parts in the set of TDMP latents, (ii) a quantity of TDMP latents in the set of TDMP latents, (iii) a size of a common part and / or a size of a specific part of the set of TDMP latents, and / or (iv) a quantization level associated with the common part and / or the specific part.

5. The WTRU of claim 4, wherein the processor and memory are configured to generate the set of TDMP latents based on the measurement information and the one or more parameters included in the TDMP configuration.

6. The WTRU of any of claims 1-5, wherein the configuration information includes a reporting configuration, and the one or more parameters included in the reporting configuration include any of (i) a quantization level associated with common part reporting and / or specific part reporting, (ii) a modulation and coding scheme (MCS) associated with common part reporting and / or specific part reporting, and / or (iii) uplink resources associated with common part reporting and / or specific part reporting.

7. The WTRU of claim 6, wherein the processor, memory, and the transceiver are configured to send the reporting information based on the one or more parameters included in the reporting configuration.

8. The WTRU of any of claims 1-7, wherein the configuration information includes a TDMP configuration, and the one or more parameters included in the TDMP configuration includes an indication to activate a TDMP feedback transmission mode of the set of TDMP feedback transmission modes, and wherein the processor, memory, and the transceiver are configured to send the reporting information using the activated TDMP feedback transmission mode.

9. The WTRU of any of claims 1-8, wherein the configuration information includes a TDMP configuration, and the one or more parameters included in the TDMP configuration includes an indication to determine the TDMP feedback transmission mode from the set of TDMP feedback transmission modes, and wherein the determined TDMP feedback transmission mode is associated with a future reporting instance for the set of TDMP latents.

10. The WTRU of any one of claims 1-9, wherein the processor and memory are configured to determine the TDMP feedback transmission mode from the set of TDMP feedback transmission modes based on any of the measurement information exceeding a first threshold, at least the partof the set of TDMP latents exceeding a second threshold, downlink channel performance, resources used to send the reporting information, and / or a network indication.

11. A method implemented by a wireless transmit / receive unit (WTRU), the method comprising: receiving configuration information indicating one or more parameters associated with channel state information (CSI) feedback of time-dependent multi-part (TDMP) latents, wherein the one or more parameters includes a set of TDMP feedback transmission modes including a single-part TDMP feedback mode and a multi-part TDMP feedback mode; receiving one or more transmissions of one or more CSI reference signals (CSI-RSs); generating, using an artificial intelligence / machine learning (AI / ML) model, a set of TDMP latents based on measurement information associated with the received one or more CSI-RSs, and wherein the set of TDMP latents includes a plurality of parts; determining a TDMP feedback transmission mode from the set of TDMP feedback transmission modes; and sending reporting information indicating (i) at least a part of the set of TDMP latents and (ii) the determined TDMP feedback transmission mode.

12. The method of claim 11, wherein a common part of at least one TDMP latent of the set of TDMP latents includes information associated with long-term variations between correlated samples included in the measurement information.

13. The method of any of claims 11-12, wherein a specific part of at least one TDMP latent of the set of TDMP latents includes information associated with short-term variations between pairs of correlated samples included in the measurement information.

14. The method of any of claims 11-13, wherein the configuration information includes a TDMP configuration, and the one or more parameters included in the TDMP configuration include any of (i) a quantity of parts in the set of TDMP latents, (ii) a quantity of TDMP latents in the set of TDMP latents, (iii) a size of a common part and / or a size of a specific part of the set of TDMP latents, and / or (iv) a quantization level associated with the common part and / or the specific part.

15. The method of claim 14, wherein the set of TDMP latents are generated based on the measurement information and the one or more parameters included in the TDMP configuration.

16. The method of any of claims 11-15, wherein the configuration information includes a reporting configuration, and the one or more parameters included in the reporting configuration include any of (i) a quantization level associated with common part reporting and / or specific part reporting, (ii) a modulation and coding scheme (MCS) associated with common part reporting and / or specific part reporting, and / or (iii) uplink resources associated with common part reporting and / or specific part reporting.

17. The method of claim 16, further comprising: sending the reporting information based on the one or more parameters included in the reporting configuration.

18. The method of any of claims 11-17, wherein the configuration information includes a TDMP configuration, and the one or more parameters included in the TDMP configuration includes an indication to activate a TDMP feedback transmission mode of the set of TDMP feedback transmission modes, and wherein the reporting information is sent using the activated TDMP feedback transmission mode.

19. The method of any of claims 11-18, wherein the configuration information includes a TDMP configuration, and the one or more parameters included in the TDMP configuration includes an indication to determine the TDMP feedback transmission mode from the set of TDMP feedback transmission modes, and wherein the determined TDMP feedback transmission mode is associated with a future reporting instance for the set of TDMP latents.

20. The method of any one of claims 11-19, wherein the TDMP feedback transmission mode is determined from the set of TDMP feedback transmission modes based on any of the measurement information exceeding a first threshold, at least the part of the set of TDMP latents exceeding a second threshold, downlink channel performance, resources used to send the reporting information, and / or a network indication.

Images