Temporal spatial frequency compression of channel state information associated with multi-resolution multi-part transmission

Abstract

Systems, methods, and instrumentalities are described herein associated with temporal spatial frequency (TSF) compression with multi-resolution multipart transmission. TSF compression may be performed using fixed and / or variable parts size (e.g., common part, specific part). Parameters for long-term and short-term TSF parts may be determined, for example, based on channel conditions, allocated uplink budget, and / or metric thresholds. The determined parameters may be used to compress the channel (e.g., generate the TSF parts). The determined parameters may be reported.

Description

TEMPORAL SPATIAL FREQUENCY COMPRESSION OF CHANNEL STATE INFORMATION ASSOCIATED WITH MULTI-RESOLUTION MULTI-PART TRANSMISSIONCROSS-REFERENCE TO RELATED APPLICATOINS

[0001] The application claims the benefit of U.S. Provisional Application 63 / 552,419, filed February 12, 2024, the contents of which are incorporated by reference in their entirety herein.BACKGROUND

[0002] Mobile communications using wireless communication continue to evolve. A fifth generation of mobile communication radio access technology (RAT) may be referred to as 5G new radio (NR). A previous (legacy) generation of mobile communication RAT may be, for example, fourth generation (4G) long term evolution (LTE).SUMMARY

[0003] Systems, methods, and instrumentalities are described herein associated with temporal spatial frequency (TSF) compression with multi-resolution multipart transmission. TSF compression may be performed using fixed and / or variable parts size (e.g., common part, specific part). Parameters for longterm and short-term TSF parts may be determined, for example, based on channel conditions, allocated uplink budget, and / or metric thresholds. The determined parameters may be used to compress the channel (e.g., generate the TSF parts). The determined parameters may be reported.

[0004] A wireless transmit / receive unit (WTRU) may perform TSF compression with multi-resolution multipart transmission. The WTRU may include a processor configured to perform one or more of the following. The WTRU may receive configuration information associated with TSF compression. The configuration information may include a configuration parameter associated with TSF compression. The configuration parameter associated with TSF compression may include one or more of a number of TSF- compressed CSI channel state information (CSI) parts, a CSI part size (e.g., minimum CSI part size and / or maximum CSI part size), an indication (e.g., a part allocation determination and compression indication), and / or the like. The configuration information may include reporting configuration information. The configuration information may include a condition (e.g., channel condition) associated with TSF part determination. The condition (e.g., channel condition) may be associated with a metric thresholdassociated with TSF part determination. The WTRU may determine a first TSF parameter, for example, based on at least the configuration information and / or a measurement associated with a CSI reference signal (CSI-RS). The first TSF parameter may be determined based on a first measured channel parameter and / or a first look up table. The first TSP parameter may be determined based on the metric threshold. The first TSF parameter may be one or more of the following: a first size, a validity time, a first quantization parameter, etc. The WTRU may determine a second TSF parameter, for example, based on at least the configuration information and / or the measurement associated with the CSI-RS. The second TSF parameter may be determined based on a second measured channel parameter and / or a second look up table. The second TSF parameter may be determined based on the metric threshold. The second TSF parameter may be one or more of the following: a second size, a second quantization parameter, etc. The WTRU may determine a first TSF-compressed CSI part, for example, based on the first TSF parameter. The first TSF- compressed CSI part may be a long-term CSI part (e.g., common CSI part). The first TSF parameter may be associated with the long-term CSI part. The WTRU may determine a second TSF-compressed CSI part, for example, based on the second TSF parameter. The second TSF-compressed CSI part may be a shortterm CSI part (e.g., specific CSI part). The second TSF parameter may be associated with the short-term CSI part. The WTRU may sent at least one of the first TSF-compressed CSI part, the first TSF parameter, the second TSF-compressed CSI part, or the second TSF parameter (e.g., based on reporting configuration information).BRIEF DESCRIPTION OF THE DRAWINGS

[0005] FIG. 1 A is a system diagram illustrating an example communications system in which one or more disclosed embodiments may be implemented.

[0006] FIG. 1 B is a system diagram illustrating an example wireless transmit / receive unit (WTRU) that may be used within the communications system illustrated in FIG. 1A according to an embodiment.

[0007] FIG. 1 C 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 an embodiment.

[0008] FIG. 1 D 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. 1A according to an embodiment.

[0009] FIG. 2 illustrates an example CSI measurement setting.

[0010] FIG. 3 illustrates an example representation of training a variable rate artificial intelligence / machine learning autoencoder model.

[0011] FIG. 4 illustrates an example AI / ML Autoencoder model with multipart latent domain that includes fixed-size common part and specific part.

[0012] FIG. 5 illustrates an example AI / ML Autoencoder model with multipart latent domain that includes variable-size common part and specific part.

[0013] FIG. 6 illustrates an example of SF compression using a two-sided autoencoder model.

[0014] FIG. 7 illustrates an example of traditional training of an AE model.

[0015] FIG. 8 illustrates an example of AE training with common and specific parts with weights associated with the common part being updated and the specific part is frozen.

[0016] FIG. 9 illustrates an example of AE training with common and specific parts, with weights associated with the specific part being updated and the common part weights being frozen.

[0017] FIG. 10 illustrates an example of training an AE model with a common encoder / decoder modules.

[0018] FIG. 11 illustrates an example of AE training with common and specific parts.

[0019] FIG. 12 illustrates an example architecture of TSF compression model with variable-size common and specific parts.

[0020] FIG. 13 illustrates example TSF Types.DETAILED DESCRIPTION

[0021] FIG. 1A is a 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), single-carrier FDMA (SC-FDMA), zero-tail unique-word DFT-Spread OFDM (ZT UW DTS-s OFDM), unique word OFDM (UW-OFDM), resource block-filtered OFDM, filter bank multicarrier (FBMC), and the like.

[0022] As shown in FIG. 1A, the communications system 100 may include wireless transmit / receive units (WTRUs) 102a, 102b, 102c, 102d, a RAN 104 / 113, a ON 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 networkelements. 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 “ST A”, may be configured to transmit and / or receive wireless signals and may include 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.

[0023] 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 to facilitate access to one or more communication networks, such as the CN 106 / 115, the Internet 110, and / or the other networks 112. By way of example, the base stations 114a, 114b may be a base transceiver station (BTS), a Node-B, an eNode B, a Home Node B, a Home eNode B, a gNB, a NR NodeB, 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.

[0024] 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 one 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 sector of the cell. For example, beamforming may be used to transmit and / or receive signals in desired spatial directions.

[0025] 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).

[0026] 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 115 / 116 / 117 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 (DL) Packet Access (HSDPA) and / or High-Speed UL Packet Access (HSUPA).

[0027] 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).

[0028] 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).

[0029] 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., a eNB and a gNB).

[0030] In other embodiments, the base station 114a and the WTRUs 102a, 102b, 102c may implement radio technologies such as IEEE 802.11 (i.e., Wireless Fidelity (WiFi), IEEE 802.16 (i.e., Worldwide Interoperability for Microwave Access (WiMAX)), CDMA2000, CDMA2000 1X, CDMA2000 EV-DO, Interim Standard 2000 (IS-2000), Interim Standard 95 (IS-95), Interim Standard 856 (IS-856), Global System for Mobile communications (GSM), Enhanced Data rates for GSM Evolution (EDGE), GSM EDGE (GERAN), and the like.

[0031] 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 alocalized 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 one embodiment, the base station 114b and the WTRUs 102c, 102d may implement a radio technology such as IEEE 802.11 to establish a wireless local area network (WLAN). In an embodiment, the base station 114b and the WTRUs 102c, 102d may implement a radio technology such as IEEE 802.15 to establish a wireless personal area network (WPAN). In yet another embodiment, the base station 114b and the WTRUs 102c, 102d may utilize a cellular-based RAT (e.g., WCDMA, CDMA2000, GSM, LTE, LTE-A, LTE-A Pro, NR etc.) to establish a 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.

[0032] 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. 1A, 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 a NR radio technology, the CN 106 / 115 may also be in communication with another RAN (not shown) employing a GSM, UMTS, CDMA 2000, WiMAX, E-UTRA, or WiFi radio technology.

[0033] 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 the 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 / 113 or a different RAT.

[0034] 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 WTRU102c shown in FIG. 1A may be configured to communicate with the base station 114a, which may employ a cellular-based radio technology, and with the base station 114b, which may employ an IEEE 802 radio technology.

[0035] FIG. 1 B is a system diagram illustrating an example WTRU 102. As shown in FIG. 1 B, 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 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.

[0036] 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. 1 B 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 in an electronic package or chip.

[0037] 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 one 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 yet another 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.

[0038] Although the transmit / receive element 122 is depicted in FIG. 1 B as a single element, the WTRU 102 may include any number of transmit / receive elements 122. More specifically, the WTRU 102 may employ MIMO technology. Thus, in one embodiment, the WTRU 102 may include two or more transmit / receive elements 122 (e.g., multiple antennas) for transmitting and receiving wireless signals over the air interface 116.

[0039] 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.

[0040] 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), read-only memory (ROM), a hard disk, or any other type of memory storage device. The removable memory 132 may include a subscriber identity module (SIM) card, a memory stick, a secure digital (SD) memory card, and the like. In other embodiments, the processor 118 may access information from, and store data in, memory that is not physically located on the WTRU 102, such as on a server or a home computer (not shown).

[0041] 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.

[0042] 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 locationdetermination method while remaining consistent with an embodiment.

[0043] The processor 118 may further be coupled to other peripherals 138, which may include one or more software and / or hardware modules that provide additional features, functionality and / or wired or wireless connectivity. For example, the peripherals 138 may include an accelerometer, an e-compass, a satellite transceiver, a digital camera (for photographs and / or video), a universal serial bus (USB) port, a vibration device, a television transceiver, a hands free headset, a Bluetooth® module, a frequencymodulated (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 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.

[0044] 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 UL (e.g., for transmission) and downlink (e.g., for reception) may be concurrent and / or simultaneous. The full duplex radio may include an interference management unit 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 WRTU 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 UL (e.g., for transmission) or the downlink (e.g., for reception)).

[0045] FIG. 1 C 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, 102c over the air interface 116. The RAN 104 may also be in communication with the CN 106.

[0046] 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 one 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 / or receive wireless signals from, the WTRU 102a.

[0047] Each of the eNode-Bs 160a, 160b, 160c may be associated with a particular cell (not shown) and may be configured to handle radio resource management decisions, handover decisions, scheduling of users in the UL and / or DL, and the like. As shown in FIG. 1 C, the eNode-Bs 160a, 160b, 160c may communicate with one another over an X2 interface.

[0048] The CN 106 shown in FIG. 1 C may include a mobility management entity (MME) 162, a serving gateway (SGW) 164, and a packet data network (PDN) gateway (or PGW) 166. While each of the foregoing elements are depicted as part of the CN 106, it will be appreciated that any of these elements may be owned and / or operated by an entity other than the CN operator.

[0049] The MME 162 may be connected to each of the eNode-Bs 162a, 162b, 162c in the RAN 104 via an S1 interface and may serve as a control node. For example, the MME 162 may be responsible for authenticating users of the WTRUs 102a, 102b, 102c, bearer activation / deactivation, selecting a particular serving gateway during an initial attach of the WTRUs 102a, 102b, 102c, 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.

[0050] The SGW 164 may be connected to each of the eNode Bs 160a, 160b, 160c in the RAN 104 via the S1 interface. The SGW 164 may generally route and forward user data packets to / from the WTRUs 102a, 102b, 102c. The SGW 164 may perform other functions, such as anchoring 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.

[0051] 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.

[0052] 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.

[0053] Although the WTRU is described in FIGS. 1 A-1 D 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.

[0054] In representative embodiments, the other network 112 may be a WLAN.

[0055] 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 access or an interface to a Distribution System (DS) or another type of wired / wireless network that carries traffic in to 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 thedestination 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 802.11e DLS or an 802.11 z 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.

[0056] When using the 802.11 ac 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.

[0057] 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 nonadjacent 20 MHz channel to form a 40 MHz wide channel.

[0058] Very High Throughput (VHT) STAs may support 20MHz, 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 by a 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 the Medium Access Control (MAC).

[0059] Sub 1 GHz modes of operation are supported by 802.11 af and 802.11 ah. The channel operating bandwidths, and carriers, are reduced in 802.11 af and 802.11 ah relative to those used in 802.11 n, and 802.11 ac. 802.11 af supports 5 MHz, 10 MHz and 20 MHz bandwidths in the TV White Space (TVWS) spectrum, and 802.11ah supports 1 MHz, 2 MHz, 4 MHz, 8 MHz, and 16 MHz bandwidths using non-TVWS spectrum. According to a representative embodiment, 802.11 ah may support Meter Type Control / Machine-Type Communications, 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).

[0060] WLAN systems, which may support multiple channels, and channel bandwidths, such as 802.11 n, 802.11 ac, 802.11 af, and 802.11 ah, 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 802.11 ah, 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.

[0061] In the United States, the available frequency bands, which may be used by 802.11 ah, 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 802.11 ah is 6 MHz to 26 MHz depending on the country code.

[0062] FIG. 1 D 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.

[0063] The RAN 113 may include gNBs 180a, 180b, 180c, though it will be appreciated that the RAN 113 may include any number of gNBs while remaining consistent with an embodiment. The gNBs 180a, 180b, 180c may each include one or more transceivers for communicating with the WTRUs 102a, 102b, 102c over the air interface 116. In one embodiment, the gNBs 180a, 180b, 180c may implement MIMO technology. For example, gNBs 180a, 108b may utilize beamforming to transmit signals to and / or receive signals from the gNBs 180a, 180b, 180c. 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 maytransmit 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).

[0064] The WTRUs 102a, 102b, 102c may communicate with gNBs 180a, 180b, 180c using transmissions associated with a scalable numerology. For example, the 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., containing varying number of OFDM symbols and / or lasting varying lengths of absolute time).

[0065] 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 a mobility anchor for WTRUs 102a, 102b, 102c and gNBs 180a, 180b, 180c may provide additional coverage and / or throughput for servicing WTRUs 102a, 102b, 102c.

[0066] 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 Function (UPF) 184a, 184b, routing of control plane information towards Access and Mobility Management Function (AMF) 182a, 182b and the like. As shown in FIG. 1 D, the gNBs 180a, 180b, 180c may communicate with one another over an Xn interface.

[0067] The CN 115 shown in FIG. 1 D may include at least one AMF 182a, 182b, at least one UPF 184a, 184b, at least one Session Management Function (SMF) 183a, 183b, and possibly a Data Network (DN) 185a, 185b. While each of the foregoing elements are depicted as part of the CN 115, it will beappreciated that any of these elements may be owned and / or operated by an entity other than the CN operator.

[0068] 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 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 in order 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 machine type communication (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.

[0069] 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 policy enforcement and QoS, providing downlink data notifications, and the like. A PDU session type may be IP-based, non-IP based, Ethernetbased, and the like.

[0070] 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, to facilitate communications between the WTRUs 102a, 102b, 102c and IP-enabled devices. The UPF 184, 184b may perform other functions, such as routing and forwarding packets, enforcing user plane policies, supporting multi-homed PDU sessions, handling user plane QoS, buffering downlink packets, providing mobility anchoring, and the like.

[0071] 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 one embodiment, the WTRUs 102a, 102b, 102c may be connected to a local Data Network (DN) 185a, 185b through the UPF184a, 184b via the N3 interface to the UPF 184a, 184b and an N6 interface between the UPF 184a, 184b and the DN 185a, 185b.

[0072] In view of Figures 1A-1 D, and the corresponding description of Figures 1A-1 D, one or more, or all, of the functions described herein with regard to one or more of: WTRU 102a-d, Base Station 114a-b, eNode-B 160a-c, MME 162, SGW 164, PGW 166, gNB 180a-c, AMF 182a-b, UPF 184a-b, SMF 183a-b, DN 185a-b, and / or any other device(s) described herein, may be performed by one or more emulation 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.

[0073] 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.

[0074] 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.

[0075] Systems, methods, and instrumentalities are described herein associated with temporal spatial frequency (TSF) compression with multi-resolution multipart transmission. TSF compression may be performed using fixed and / or variable parts size (e.g., common part, specific part). Parameters for longterm and short-term TSF parts may be determined, for example, based on channel conditions, allocated uplink budget, and / or metric thresholds. The determined parameters may be used to compress the channel (e.g., generate the TSF parts). The determined parameters may be reported.

[0076] A wireless transmit / receive unit (WTRU) may perform TSF compression with multi-resolution multipart transmission. The WTRU may include a processor configured to perform one or more of the following. The WTRU may receive configuration information associated with TSF compression. Theconfiguration information may include a configuration parameter associated with TSF compression. The configuration parameter associated with TSF compression may include one or more of a number of TSF- compressed CSI channel state information (CSI) parts, a CSI part size (e.g., minimum CSI part size and / or maximum CSI part size), an indication (e.g., a part allocation determination and compression indication), and / or the like. The configuration information may include reporting configuration information. The configuration information may include a condition (e.g., channel condition) associated with TSF part determination. The condition (e.g., channel condition) may be associated with a metric threshold associated with TSF part determination. The WTRU may determine a first TSF parameter, for example, based on at least the configuration information and / or a measurement associated with a CSI reference signal (CSI-RS). The first TSF parameter may be determined based on a first measured channel parameter and / or a first look up table. The first TSP parameter may be determined based on the metric threshold. The first TSF parameter may be one or more of the following: a first size, a validity time, a first quantization parameter, etc. The WTRU may determine a second TSF parameter, for example, based on at least the configuration information and / or the measurement associated with the CSI-RS. The second TSF parameter may be determined based on a second measured channel parameter and / or a second look up table. The second TSF parameter may be determined based on the metric threshold. The second TSF parameter may be one or more of the following: a second size, a second quantization parameter, etc. The WTRU may determine a first TSF-compressed CSI part, for example, based on the first TSF parameter. The first TSF- compressed CSI part may be a long-term CSI part (e.g., common CSI part). The first TSF parameter may be associated with the long-term CSI part. The WTRU may determine a second TSF-compressed CSI part, for example, based on the second TSF parameter. The second TSF-compressed CSI part may be a shortterm CSI part (e.g., specific CSI part). The second TSF parameter may be associated with the short-term CSI part. The WTRU may sent at least one of the first TSF-compressed CSI part, the first TSF parameter, the second TSF-compressed CSI part, or the second TSF parameter (e.g., based on reporting configuration information).

[0077] A wireless transmit / receive unit (WTRU) may perform TSF compression with multi-resolution multipart transmission. The WTRU may receive configuration information associated with TSF compression. The configuration information may include a configuration parameter associated with TSF compression (e.g., number of TSF-compressed CSI parts, minimum size of CSI part, maximum size of CSI part, part allocation determination and compression indication). The configuration information may include a condition associated with TSF part determination. The condition may include a metric threshold. The configuration information may include reporting configuration information. The WTRU may receive a channel state information (CSI) reference signal (RS). The WTRU may determine one or more TSF configuration parameters (e.g., a first TSF configuration parameter and a second TSF configurationparameter). The TSF configuration parameters may include one or more of a size, a validity time, a quantization information, etc. The TSF configuration parameters may be associated with respective TSF- compressed CSI parts. The WTRU may determine TSF-compressed CSI parts, for example, based on respective TSF configuration parameters. For example, the first TSF-compressed CSI part may be determined based on the first TSF configuration parameter. The second TSF-compressed CSI part may be determined based on the second TSF configuration parameter. The first TSF compressed CSI part may be a long-term CSI part. The second TSF compressed CSI part may be a short-term CSI part. The WTRU may report at least one TSF-compressed CSI part and its associated TSF configuration parameter.

[0078] The WTRU may reduce the channel state information (CSI) reporting overhead, for example, using a model (e.g., two-sided artificial intelligence (Al) / machine learning (ML)) model) that may leverage the CSI temporal correlation properties (e.g., to improve the compression performance).

[0079] Details associated with channel state information (CSI) reporting may be provided herein.

[0080] Channel state information may include one or more of the following: a channel quality index (CQI), rank indicator (Rl), a precoding matrix index (PMI), an L1 channel measurement (e.g., RSRP such as L1-RSRP, or SINR), a CSI-RS resource indicator (CRI), an SS / PBCH block resource indicator (SSBRI), a layer indicator (LI), and / or a (e.g., any other) measurement quantity measured by the WTRU from the configured reference signals (e.g., CSI-RS or SS / PBCH block or any other reference signal).

[0081] CSI reporting framework details may be provided herein.

[0082] A WTRU may report the CSI through the uplink control channel on a physical uplink control channel (PUCCH) or per the gNBs’ request on an UL PUSCH grant. CSI-RS may cover a bandwidth part (BWP) (e.g., the full bandwidth of a BWP, a fraction of it), for example, depending on the configuration. Within the CSI-RS bandwidth, CSI-RS can be configured in a PRB (e.g., each PRB or every other PRB). In the time domain, CSI-RS resources can be configured either periodic, semi-persistent, or aperiodic. Semi- persistent CSI-RS may be similar to periodic CSI-RS. The resource (e.g., for semi-persistent CSI-RS) may be (de-)activated, for example, via signaling (e.g., by MAC CEs). The WTRU may report related measurements, for example, if (e.g., only when) the resource is activated. For Aperiodic CSI-RS, the WTRU may be triggered to report measured CSI-RS on a PUSCH, for example, by request in a DCI.Periodic reports may be carried over the PUCCH. Semi-persistent reports can be carried (e.g., either) on a PUCCH and / or a PUSCH. The reported CSI may be used by the scheduler, for example, if (e.g., when) allocating optimal resource blocks (e.g., possibly based on channel’s time-frequency selectivity), determining precoding matrices, beams, transmission mode, and / or selecting suitable MCSs. The reliability, accuracy, and timeliness of WTRU CSI reports may meet (e.g., be critical to meeting) URLLC service requirements.

[0083] A WTRU may receive configuration information indicating (e.g., be configured with) a CSI measurement setting. The CSI measurement setting 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 shows an example configuration for CSI reporting settings, resource settings, and link.

[0084] FIG. 2 illustrates an example CSI measurement setting.

[0085] In a CSI measurement setting, one or more of the following configuration parameters may be provided: N>1 CSI reporting settings, M>1 resource settings, and a CSI measurement setting (e.g., which links the N CSI reporting settings with the M resource settings); a CSI reporting setting (e.g., including at least one of a time-domain behavior (e.g., aperiodic or periodic / semi-persistent), frequency-granularity (e.g., at least for PMI and CQI), a CSI reporting type (e.g., PMI, CQI, Rl, CRI, etc.), a PMI Type (Type I or II) and codebook configuration (e.g., if a PMI is reported)); a resource setting (e.g., including at least one of a time-domain behavior (e.g., aperiodic or periodic / semi-persistent), an RS type (e.g., for channel measurement or interference measurement), S>1 resource set(s) (e.g., where each resource set can contain Ks resources); a CSI measurement setting (e.g., including at least one of a (e.g., one) CSI reporting setting, a (e.g., one) resource setting, a reference transmission scheme setting (e.g., for CQI)); a frequency granularity, for example, for CSI reporting for a component carrier (e.g., wideband CSI, partial band CSI, sub band CSI); etc.

[0086] Details associated with using artificial intelligence (Al) may be provided herein.

[0087] Artificial intelligence may include the behavior exhibited by machines. Such behavior may mimic cognitive functions to sense, reason, adapt and act. The terms Artificial Intelligence (Al), Machine Learning (ML), Deep Learning (DL), DNNs may be used interchangeably. Methods described herein are exemplified based on learning in wireless communication systems. The methods may not be limited to such scenarios, systems and services and may be applicable to any type of transmissions, communication systems and / or services etc.

[0088] Details associated with an autoencoder (AE) may be provided herein.

[0089] Auto-encoders (AE) may include a class (e.g., specific class) of deep neural networks (DNNs) that arise in context of un-supervised machine learning setting (e.g., where the high-dimensional data is non-linearly transformed to a lower dimensional latent vector using the DNN based encoder and the lower dimensional latent vector is then used to re-produce the high-dimensional data using a non-linear decoder). The encoder may be represented as E(x; We), for example, where x may include the high-dimensional data and Wemay include the parameters of the encoder. The decoder may be represented as £>(z; Wd~) where z may include the low-dimensional latent representation and Wd may include the parameters of thedecoder. Further, using training data { xlt■■■ , xN} the auto-encoder can be trained by solving the following optimization problem shown in Eq. 1. tr, idtr} = arg min £=1| | xt- D(E(x , We); Wd) | |^ Eq. 1We,Wd

[0090] Eq. 1 may be solved (e.g., approximately solved), for example, using backpropagation (e.g., a backpropagation algorithm). The trained encoder E(x; Wetr) can be used to compress the highdimensional data and trained decoder D z; Wdr) can be used to decompress the latent representation.

[0091] A variable rate AE may be trained.

[0092] The AI / ML Encoder-Decoder model can take advantage of a variable rate compression scheme, for example, by modifying the training process, as shown in FIG. 3. The training process may enable a variable size bottleneck where the top K elements in the latent domain may be used to reconstruct the CSI with a specific reconstruction performance, and an element (e.g., every additional element) starting from the ( + 1) - th up to the K + Kmax - th may be transmitted, for example, to further improve the reconstruction associated with the K top elements (e.g., where Kmax may include the total number of elements in the latent space).

[0093] The training may use (e.g., require) access to encoder and decoder model architectures. The models maybe initialized with some pre-training or may be trained from scratch.

[0094] The training may use (e.g., require) access to training data and a loss function (e.g., required loss function).

[0095] The training may use (e.g., require) access to the set of training masks applied in the latent space and the corresponding probability with a (e.g., each) mask. The training masks may indicate (e.g., define) the granularity at which the compression rate and the feedback can be varied. The (e.g., all) masks may have an element (e.g., at least one element) as 1, for example, allowing information to pass through.

[0096] For example, a (e.g., one) set of training masks for a 4 dimensional latent may include (1 ,0,0,0), (1,1 , 0,0), (1 ,1 , 1,0), (1,1 , 1 ,1), for example, with equal probability to each mask. In these masks, the 1st element may be 1 which may indicate that (e.g., irrespective of the mask being selected) the 1st element of the latent vector z may be the most important and most informative. Similarly, the 2nd element may have the next highest occurrence, thus indicating (e.g., ensuring) it’s the next most informative element, followed by the 3rd and 4th element.

[0097] At a (e.g., each) training iteration (e.g., for each CSI in the training batch), a (e.g., one) mask from the given set of training masks may be selected with the associated probability. The selected maskmay be applied to the latent vector received as the output of the encoder, for example, by element wise multiplication between the latent vector and the mask.

[0098] FIG. 3 illustrates an example representation of training a variable rate artificial intelligence / machine learning autoencoder model.

[0099] CSI enhancements may be applied for use with AI / ML, for example, relating to CSI compression and CSI prediction. CSI compression may be enhanced using AI / ML, for example, taking into considerations the AI / ML complexities, e.g., LCM, model transfer, model generalization, etc. CSI compression may include using the temporal correlation properties of the channel on top of spatial- frequency (SF) compression (e.g., temporal-spatial-frequency compression (TSF)).

[0100] The AI / ML usage for CSI compression may include a model (e.g., a two-sided model, e.g., autoencoder) that may compress the estimated spatial-frequency (SF) channel sample at a given time into a latent representation using an Al encoder model. The compressed channel may be transmitted and / or decompressed by an Al decoder model, for example, to reconstruct the channel. Compression performance (e.g., from an overhead perspective or reconstruction quality or both) may be improved, for example, using TSF through utilizing the temporal correlation of the channel in the compression process. Exploiting the CSI temporal correlation on the top of SF compression may improve reconstruction performance for a given overhead, reduce the overhead for a given performance, improve both performance and overhead relative to the SF compression, and / or add flexibility in controlling the CSI feedback overhead over time.

[0101] TSF compression may be enabled and / or performed. The latent domain / representation can be structured (e.g., to enable TSF compression) to include two or more TSF parts. A TSF part may include a (e.g., one) common part that captures the long-term variations or common information in a sequence of correlated CSI samples. A TSF part may include a specific part that captures the short-term variations or the individual information in a (e.g., each) sample. The common part may be used (e.g., along with the specific part) to reconstruct a (e.g., desired) CSI sample. The specific part may be used (e.g., along with the common part) to reconstruct a (e.g., desired) CSI sample. The common information may be (e.g., need to be) sent periodically (e.g., once every number of CSI samples / reporting occasions). The specific information may be (e.g., need to be) transmitted periodically (e.g., every CSI reporting occasion).Enforcing such a structure in the latent domain may reduce the CSI feedback reporting signaling overhead.

[0102] A flexible TSF operation may include structuring the latent domain into two parts (e.g., a common part and a specific part). For example, the sizes of the common and specific parts may be controlled and / or determined (e.g., based on conditions / measurements).

[0103] TSF multipart generation with controllable part size may be enabled. The different parts sizes for a given uplink overhead budget may be selected / determined. Changing / updating the size of one or more of the parts may be triggered.

[0104] The AE latent space may be structured (e.g., to enable efficient TSF compression) 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. The short-term latent may capture the short-term variations (e.g., between two consecutive CSI samples) in CSI samples. The terms long-term latent and / or common latent and / or common part may be used interchangeably to represent the long-term variations in a sequence of CSI samples. The terms short-term latent and / or specific latent and / or common part may be used interchangeably to represent the short-term variations in a sequence of CSI samples.

[0105] FIG. 4 illustrates an example AI / ML Autoencoder model with multipart latent domain that includes fixed-size common part and specific part.

[0106] FIG. 5 illustrates an example AI / ML Autoencoder model with multipart latent domain that includes variable-size common part and specific part. As shown in FIG. 5, zvmay be a number of elements that may be attached (e.g., fully attached) to the common part (e.g., to increase its size to zc+ zv) or to the specific part (e.g., to increase its size to zs+ zv). In examples, zvmay be split across the common part and the specific part.

[0107] Other implementations may be considered to handle the multi-size multipart transmission. For example, the common part may be generated using a separately trained AE model (e.g., a common AE model) with adaptive bottleneck size. The specific part may be generated using a different AE model that may use the output of the common AE model (e.g., a residual AE model) with adaptive bottleneck size.

[0108] Temporal-Spatial-Frequency (TSF) CSI compression (e.g., with multi-resolution multi-part transmission) may be performed and / or enabled.

[0109] TSF with a fixed size of the common and specific parts may improve the compression performance relative to SF compression. TSF with a fixed size of the common and specific parts may lack flexibility of adapting the parts size based on environment. For example, a small common part and a large specific part may be used (e.g., for high speed scenarios) or a large common part and a small specific part may be used (e.g., for low speed scenarios), for example, because commonality may reduce between samples (e.g., in high-speed scenarios). The latent structure parts sizes (e.g., long-term part size and short-term part size) may be controlled under TSF compression.

[0110] A WTRU may perform TSF based CSI compression. The WTRU may determine a parameter (e.g., the parameters) of the long-term and short-term TSF part(s) (e.g., the size of the long-term and shortterm component in the latent space and the associated quantization with each part), for example, based on(e.g., as a function of) one or more of the following: channel conditions, allocated uplink budget, and / or configured metrics threshold(s). The WTRU may use the determined parameters to compress the channel (e.g., generate the TSF parts). The WTRU may report the determined parameters.

[0111] In examples, the WTRU may receive configuration information (e.g., associated with TSF compression) indicating (e.g., be configured with) one or more parameters, for example, to generate CSI feedback based on temporal-spatial-frequency (TSF) compression (e.g., using two-sided model). The TSF- compressed CSI may include multiple (e.g., two or more) parts. For example, the TSF-compressed CSI may include a long-term / common part. The long-term / common part may capture the long-term variation(s) in a sequence of correlated samples. The long-term / common part may be fed back (e.g., transmitted by the WTRU) after a CSI feedback transmission, for example, once every multiple (e.g., once every N, for example, which may be configured) CSI feedback transmissions (e.g., periodically). The long-term / common part may be stored and / or reused for feedback reports (e.g., consecutive N feedback reports). The long- term / common part may be associated with (e.g., have) controllable / variable size feedback overhead, e.g., size range. For example, the TSF-compressed CSI may include a short-term / specific part. The short- term / specific part may capture the short-term variation(s) between a pair of samples (e.g., every pair of consecutive samples) in a sequence of N correlated samples. The short-term / specific part may be sent in a (e.g., every) CSI feedback transmission. The short-term / specific part may have controllable / variable-size feedback overhead, e.g., size range.

[0112] The WTRU may receive configuration information (e.g., associated with TSF compression), for example, indicating one or more configuration parameters associated with the TSF compression (e.g., TSF configuration parameter). The configuration parameters associated with the TSF compression may include one or more of the following: TSF configuration and parameters thereof (e.g., a number of TSF- compressed CSI parts, a minimum CSI part size, a maximum CSI part size, a part allocation determination and compression indication, etc.); conditions / parameters for TSF parts (e.g., latents) determination (e.g., selection); reporting configuration information; etc.

[0113] TSF configuration and parameters thereof may include one or more of the following: a number of TSF-compressed CSI parts / latents; a size or payload (e.g., minimum size or payload, maximum size or payload) of a (e.g., each) part (e.g., long-term / common part or short-term / specific part); a part allocation determination and compression indication (e.g., flag), for example, where if the indication is set to 1, the WTRU may determine a (e.g., each) part size / payload and use its determination to perform compression; input domain type (e.g., eigenvalue (EV) vs full CSI); etc.

[0114] Conditions and / or parameters for TSF latents / parts determination (e.g., selection) may include one or more of the following: a short-term performance metric threshold (e.g., short term squaredgeneralized cosine similarity (SGCS) threshold, for example, in full CSI domain between consecutive CSI samples); a long-term performance metric threshold (e.g., long-term SGCS threshold, for example, between two CSI samples, such as, a CSI sample and the first CSI sample in a TSF sequence, such as, a TSF sample for which long-term part was fed back by the WTRU, or in the latent domain between the active common part associated with the first CSI sample in a TSF sequence, (e.g., fed back by the WTRU and stored in gNB buffer for CSI reconstruction) and the inactive common parts associated with the rest of CSI samples in a TSF sequence (e.g., determined by the WTRU but not transmitted to gNB)); a combined performance metric threshold (e.g., combined SGCS threshold); etc.

[0115] The reporting configuration information may include information associated with part-specific reporting. For example, a modulation and coding scheme (MCS) and / or quantization (e.g., a quantization parameter), UL resource(s), or Multipart allocation(s) / budget, may be configured as a budget (e.g., max budget) for two or more parts or a budget (e.g., max budget) per part.

[0116] The WTRU may receive a CSI-RS transmission. The channel may be estimated (e.g., measured), for example, based on the received CSI-RS transmission. The measured CSI (e.g., full channel) may be stored.

[0117] The WTRU may determine one or more TSF configuration parameters (e.g., a first TSF parameter and / or second TSF parameter) associated with one or more TSF-compressed CSI parts / latents (e.g., a first TSF parameter and / or a second TSF parameter), for example, based on the received configuration information (e.g., indicating preconfigured condition(s) and / or a metric threshold) and measurements (e.g., a measurement associated with the received CSI-RS transmission). The WTRU may determine the parameter(s) (e.g., size, associated validity time, quantization information (e.g., quantization parameter), etc.) of a first TSF-compressed CSI part (e.g., the long-term (e.g., common) TSF-compressed CSI part), for example, based on one or more of the following conditions / measurements. The WTRU may determine the parameter(s) (e.g., first TSF parameter) of the long-term (e.g., common) TSF-compressed CSI part based on the configured multipart allocation budget and / or configured size (e.g., maximum / minimum size). The WTRU may determine the parameter(s) (e.g., first TSF parameter) of the long-term (e.g., common) TSF-compressed CSI part based on the measured channel parameters (e.g., Doppler, WTRU speed). For example, the long-term latent size may be determined based on the estimated Doppler and comparing with a threshold (e.g., high Doppler indicates small size and vice versa). For example, this may be used to narrow down the search space for the common part size based on a defined mapping between speed ranges and size ranges. The WTRU may determine the parameter(s) (e.g., first TSF parameter) of the long-term (e.g., common) TSF-compressed CSI part based on CSI measurements (e.g., comparing the measured long-term SGCS with the configured long-term SGCS threshold). Forexample, if the measured long-term SGCS is g reater / smaller than a threshold, the common size may be changed (e.g., by ■+ / - x% relative to currently used size). The determination may be based on a predefined mapping between x% and the difference between measured SGCS and threshold. The WTRU may determine the parameter(s) (e.g., first TSF parameter) of the long-term (e.g., common) TSF-compressed CSI part based on look-up table (e.g., a double-index look-up table) mapping the two thresholds (e.g., long and short term) to part sizes.

[0118] The WTRU may determine the parameter(s) (e.g., size and quantization information) of the shortterm (e.g., specific) TSF-compressed CSI latent / part (e.g., size, quantization information (e.g., quantization parameter), etc.), for example, based on one or more of the following. The WTRU may determine the parameter(s) (e.g., second TSF parameter) of the short-term (e.g., specific) TSF-compressed CSI latent / part based on a multi-part allocation budget (e.g., the configured multipart allocation budget), configured minimum / maximum allocated budget, and / or determined budget of the long-term part. The WTRU may determine the parameter(s) (e.g., second TSF parameter) of the short-term (e.g., specific) TSF-compressed CSI latent / part based on measured channel parameters (e.g., Doppler or WTRU speed). For example, a high Doppler may use a large size of the short-term / specific part and vice versa. A size (e.g., maximum size) may be used, for example, if the WTRU speed exceeds a threshold (e.g., certain threshold). The WTRU may determine the parameter(s) (e.g., second TSF parameter) of the short-term (e.g., specific) TSF-compressed CSI latent / part based on CSI measurements (e.g., comparing the measured short-term SGCS with the configured short-term SGCS threshold). For example, the size may be adapted by + / - y% relative to a size (e.g., current specific size) based on the difference between measured and threshold. The WTRU may determine the parameter(s) (e.g., second TSF parameter) of the short-term (e.g., specific) TSF-compressed CSI latent / part based on a look-up table (e.g., double-index look-up table) mapping the two thresholds (e.g., long and short term) to part sizes.

[0119] The WTRU may compress the CSI and / or generate the TSF-compressed CSI parts (e.g., common and specific), for example, based on the determined TSF configuration parameters (e.g., determined allocation for each part). For example, a first TSF-compressed CSI part (e.g., long-term and / or common CSI part) may be determined based on the first TSF parameter. For example, a second TSF- compressed CSI part (e.g., short-term and / or specific CSI part) may be determined based on the second TSF parameter.

[0120] The WTRU may report one or more TSF-compressed CSI parts and / or the associated parts parameters. For example, the WTRU may report the TSF configuration parameters (e.g., number of parts, allocation of each transmitted / reported part). For example, the WTRU may report the TSF-compressed CSI for one or more parts. For example, the WTRU may send one or more of the following: the first TSF-compressed CSI part, the first TSF parameter, the second TSF-compressed CSI part, or the second TSF parameter.

[0121] CSI feedback with time-spatial-frequency (TSF) compression may be enabled and / or used. Details described herein with respect to CSI feedback with TSF compression may not be limited to just CSI feedback and TSF and may be applicable to (e.g., any) transmission, report, measurements, and the likes at which the proposed structure is applicable (e.g., extracting common information and specific information). AI / ML framework may be used as an enablement, and the procedures may apply to non- AI / ML based transmissions.

[0122] Spatial-Frequency (SF) compression may include a compression technique, for example, that compresses the estimated CSI sample (e.g., raw channel or eigenvector) using a model (e.g., an encoder model) at the WTRU and use the compressed CSI to recover the decompressed CSI using a model (e.g., decoder model) at the gNB. SF compression may be employed (e.g., independently) in for a (e.g., each) CSI sample, for example, without using temporal information (e.g., previous CSI estimates). The estimated channel at time t may be used to derive the compressed channel.

[0123] Temporal-Spatial-Frequency (TSF) compression may include a compression technique that utilizes a (e.g., at least one) past / historical CSI sample (e.g., raw channel or eigenvector) along with the currently estimated CSI sample at the WTRU to generate the TSF-compressed CSI, and a (e.g., at least one) past / historical CSI measurement along with the current compressed CSI at the gNB to generate / recover the current decompressed CSI.

[0124] A TSF Multipart structure may include a structure (e.g., an enforced structure) on the output of an AI / ML model to enable TSF compression, for example, where the model output is structured to include multiple parts with at least one part capturing common information in a sequence of samples (e.g., time series data), which may be referred to common information, and other parts capturing specific information in each sample of the sequency, may be referred to as specific information.

[0125] A multiresolution TSF Multipart structure may include a TSF multipart structure with controllable / dynamic common and / or specific part size.

[0126] A TSF Multipart model may include an AI / ML model trained to perform TSF compression, for example, through the generation of common and specific parts.

[0127] A TSF sequence may include a sequence of N correlated CSI samples associated with a specific WTRU. Samples (e.g., all samples) in the TSF sequence may have (e.g., may be assumed to have) a (e.g., one) common part that is reused for samples (e.g., all samples) in the reconstruction / decompression process.

[0128] A common part may include a part carrying latent common information or capturing coarse variations in a sequence of correlated data samples, e.g., TSF sequence. The terms sample-common part, common part, long-term part may be used interchangeably to represent the common information carried by the common part.

[0129] An active common part may include the common part, for example, generated using the first sample in a TSF sequence. This common part may be transmitted, stored, and used (e.g., by gNB) in the reconstruction process of samples (e.g., all samples).

[0130] An inactive common part may include (e.g., refer to) the common part generated from the other N-1 samples in a TSF sequence (e.g., but may not be transmitted to gNB). The inactive common part may be stored (e.g., at the WTRU), for example, to track the discrepancy between the active common part and the inactive common parts.

[0131] A specific part may include a part carrying latent specific information in each sample or fine variations between pairs of consecutive samples in a sequence of correlated data samples. The terms sample-specific part, specific part, short-term part may be used interchangeably to represent the specific information carried out by the specific part.

[0132] A TSF latent buffer may include the buffer used at the WTRU to store the past CSI samples (e.g., raw channel or eigenvector) or the buffer used at the gNB to store the past decompressed CSI samples or the active common part.

[0133] Details associated with spatial frequency (SF) compression may be described herein.

[0134] The spatial frequency (SF) compression may operate on a sample-by sample basis. For example, the WTRU may use an AE model to perform the compression at time slot n based on the estimated channel Hn. The WTRU may (e.g., first) compress Hnusing a model (eg.., an encoder model) to generate and send back the latent representation (e.g., compressed CSI) zn. The gNB may use the model (e.g., decoder model) to decompress the received latent znto recover Hn, for example, where the difference / distance between the estimated channel Hnat the WTRU and the recovered / decompressed channel at the gNB may represent the compression loss. FIG. 6 illustrates an example of SF compression using a two-sided autoencoder model.

[0135] Details associated with TSF compression may be described herein.

[0136] To further improve the SF compression performance (e.g., from overhead reduction and / or performance perspective), the AI / ML encoder model may be trained, for example, to represent the compressed information in two or more parts (e.g., one part that carries a common information for a sequence of correlated samples (e.g., a TSF sequence) and another part that captures a specific information in each sample). The two parts information may be input (e.g., fed together) to the AI / MLdecoder model, for example, to reconstruct the CSI sample. The common information may be sent for the samples (e.g., entire samples) in the TSF sequence. The specific information may be sent along with a sample (e.g., every sample) in the TSF sequence. Such a model may be referred to as TSF multipart model. The compression mode may be referred to as time-spatial-frequency (TSF) compression, for example, where the temporal correlation properties of the channel samples may be leveraged to generate latent parts with some enforced structure to reduce the feedback overhead over time (e.g., as shown in FIG. 4). The AI / ML model may be trained, for example, to generate (e.g., multiple) latent part(s) (e.g., possibly of different sizes as shown in FIG. 5), that capture the long-term and short-term correlations of the raw CSI samples. Enforcing such a structure in the latent domain may reduce (e.g., result in reduction in the) signaling overhead relative to the SF compression.

[0137] TSF with a fixed size of common and specific parts may be enabled, performed, and / or described herein.

[0138] FIG. 4 shows an example of the TSF multipart model, for example, where the common part and specific part may be generated using a single AI / ML model (e.g., AE). The part(s) (e.g., both parts) may have fixed sizes. The proposed framework can be obtained using different model architectures (e.g., may not be limited to the model architecture illustrated in FIG. 4) . For example, the common part may be generated using a first AI / ML (e.g., AE) model. The specific part may be obtained using a second AIML model. The output of the first and second AI / ML model may be combined together, for example, to reconstruct / recover the desired information.

[0139] A TSF with a variable size of common and specific parts may be enabled, performed, and / or described herein.

[0140] The TSF multipart model may be trained to generate controlled-size part(s) (e.g., two or more controlled-size parts), for example, to enable more efficient TSF framework. For example, the model may generate variable size common part and variable size specific parts. This may be used to handle multiple WTRU scenarios (e.g., speed) with a single AI / ML TSF model. For example, high speed scenarios may use (e.g., require) a small common part and large speed parts (e.g., as channel samples tend to be less correlated as the speed increases and hence less commonality in the TSF sequence). Low speed scenarios may use (e.g., require) a large common part for more efficient communication and small specific part to capture the small changes in the channel. FIG. 5 illustrates an example of a TSF model with variable-size common and specific parts.

[0141] Different training procedures for both the variable size and fixed size TSF multipart models may be used, provided, and / or described herein.

[0142] The example structures as shown in FIGs. 4 and 5 may be applicable to a compression problem that involve time-series data, e.g., a sequence of correlated samples. TSF compression may be described (e.g., herein) as an example. Details described herein may be applicable to a (e.g., any) framework where the structures in FIGs. 4 and 5 are applicable.

[0143] TSF training may be performed and / or described herein.

[0144] A WTRU may to train an AI / ML model (e.g., autoencoder) to generate an encoder output with an enforced structure for TSF compression operation. The AE model may be trained to output a long-term part (e.g., Zc) and a short-term sample-specific part (e.g., Zi), for example, given a sequence of N temporally- correlated samples {Hi, . . ., HN} (e.g., TSF sequence with length N). The long-term / common part may capture the long-term correlation in the input sequence of sample. The short-term parts may capture the short-term correlation, for example, between samples (e.g., 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 =The sum of sizes may be less than or equal to a given max size zmax.

[0145] Details associated with example procedures for training a TSF multipart model with fixed size parts may be described herein.

[0146] The AE model may be trained to generate a latent output with a common structure zc(e.g., sample-common part) and a short-term part zt, i =(e.g., sample-specific part), for example, using the following. The autoencoder model may be trained (e.g., in the traditional fashion) with channel sample(s) (e.g., same channel samples) fed as input and labels for the model training. The model may be trained with the following input / output pairs:(H2, W2), (.H3, H3), for example, given a sequence of N (e.g., N=3, temporally-correlated samples {H H2, H3}. The weights (e.g., all the weights) of the model may be updated, for example, during the first stage of the training. The channel matrices Hi may be used as inputs and / or the labels (e.g., both inputs and as the labels), as shown in FIG. 7.

[0147] FIG. 7 illustrates an example of traditional training of an AE model.

[0148] Once the training is finished, the weights (e.g., all the weights) of the model (e.g., except for the ones directly interacting with the latent, for example, the output layer of the encoder and input layer or the 1st layer weights of the decoder) may be frozen and may be refrained from being updated (e.g., not be updated). The training associated with generating the common part zcmay be performed (e.g., undertaken). To generate the long-term / common part zc, one or more of the following samples pairs may be constructed:

[0149] The pairing of dis-similar channels may be used to have the model (e.g., 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 zcmay (e.g., be allowed to) propagate to the decoder in the forwardpass. The corresponding gradients may (e.g., be allowed) to back propagate to the encoder. The parts of the latent corresponding to the short-term components Zi may be blocked. The blocking of this part of the latent may be achieved by multiplying (e.g., forcefully multiplying) the latent component a with a zero vector (e.g., as shown in FIG. 8).

[0150] FIG. 8 illustrates an example of AE training with common and specific parts with weights associated with the common part being updated and the specific part being frozen, e.g., using an all zero vector.

[0151] During this stage of the training, the weights (e.g., only the weights directly interacting with zc) may be updated, for example, because the common part (e.g., only the common part) of the latent may propagate to the decoder and the gradients associated with this part (e.g., only this part) may be back propagated.

[0152] Once the training for the common part has been completed, the short-term componentszN} may be trained. The training may be performed with the pair of channels where a set of channels (e.g., the same set of channels) may be used for both input and output (e.g., as shown in FIG. 9), for example, to train the part of the network that generates the short-term components Zi. For example,(HltH- , (H2, H2), (H3IH3), where the output latent Zi may be used along with the common latent / part Zc for reconstruction.

[0153] FIG. 9 illustrates an example of AE training with common and specific parts, with weights associated with the specific part being updated and the common part weights being frozen.

[0154] Examples associated with training a TSF multipart model with variable size parts may be described herein.

[0155] In examples for TSF model training with variable-size parts, different models may be utilized for generating the variable-size common part of the latent (e.g., zc) and the variable size short-term part of the latent (e.g., a). For example, the model may be trained to generate the common part using a common encoder / decoder module (e.g., as shown in FIG. 10). For training, the dis-similar channels may be used as input. Pairs may be labeled to have (e.g., force) the model to learn the common components between different channels. Therefore, the training data may have input and label pairs such as one or more of the following: (HlzH2), (HlzW3), (H2, H- , (H2, H3),To handle the variable rate common part, the common encoder / decoder modules may be trained using the variable rate training process (e.g., as described herein).

[0156] FIG. 10 illustrates an example of training an AE model with a common encoder / decoder modules.

[0157] In post training, the common encoder weights may be frozen and a second encoder-decoder pair (e.g., residual encoder-decoder) may be used (e.g., introduced). To train this model (e.g., that may generate the short-term components {z1,.., zN]), the training may be performed with the pair of channels where the same set of channels is used both as inputs and labels (e.g., (HltH- , (H2, H2), (H3, H3). Using a training process (e.g., as described herein), a variable rate specific part may be generated. FIG. 11 illustrates an example summary of the training.

[0158] FIG. 11 illustrates an example of AE training with common and specific parts. The weights of the common encoder and decoder models may be fixed (e.g., kept fixed). The residual encoder-decoder may be updated, for example, using the defined loss function.

[0159] The weights of the common encoder and decoder models may be fixed (e.g., kept fixed) and the residual encoder-decoder may be updated using the defined loss function. This residual model may learn (e.g., effectively learn) to compress on top of the common encoder-decoder models.

[0160] An example architecture of the variable size common and specific parts is shown in FIG. 12.

[0161] FIG. 12 illustrates an example architecture of TSF compression model with variable-size common and specific parts.

[0162] The temporal-spatial-frequency compression approach (e.g., as described herein) may enable one or more of the following (e.g., benefits): the TSF may leverage the CSI temporal correlation properties to enable high compression capabilities relative to the spatial frequency (SF) and legacy based compression for a given target performance; the TSF may leverage the CSI temporal correlation properties to further improve the performance of SF and legacy based compression at a given compression rate; the TSF may provide (e.g., both) performance and overhead reduction gains relative to SF and legacy compression; the TSF may provide dynamic adaptation and flexible use of the past historical samples to achieve the balance between performance, complexity, overhead and storage; etc.

[0163] TSF parts may be indicated (e.g., by the WTRU).

[0164] Temporal correlation may be leveraged, for example, via multi-part TSF CSI transmission.

[0165] The WTRU may transmit CSI feedback in parts (e.g., two parts), for example, such as, a first part including common information (e.g., including the long-term channel characteristics / variation / properties) and a second part including specific information (e.g., including short-term channel characteristics / variation / properties). In examples, the WTRU may transmit CSI feedback in parts (e.g., multiple parts). For example, the multiple parts may include a (e.g., one) common part and a (e.g., one or more) specific part(s). In examples, the size of common part and / or specific part may be fixed and / or variable within a preconfigured range. In examples, the WTRU may apply different post-processing (e.g., quantization) for the common part and specific part. The WTRU may receive configuration informationindicating (e.g., be configured with) conditions / rules to determine the size and / or post processing applied to common part and specific part, for example, based on the UL resources configured for feedback transmission. In examples, the WTRU may indicate / report the parameters and / or formats of common and / or specific parts along with the associated CSI feedback. The WTRU may indicate / report the parameters of common and / or specific parts separate from the associated CSI report. The WTRU may indicate the TSF parts or parameters thereof using implicit signaling, explicit signaling, or a combination of both.

[0166] A PUCCH bundle may be used, for example, for multi-part TSF CSI transmission.

[0167] The WTRU may receive configuration information indicating (e.g., be configured with) a PUCCH bundle for multi-part TSF CSI transmission. The PUCCH bundle (e.g., each PUCCH bundle) may be associated with a plurality of PUCCH resources. A (e.g., each) PUCCH resource within the PUCCH bundle may be configured for transmission of a part (e.g., at least one part) of multi-part TSF CSI. For example, the WTRU may transmit a common part in a first PUCCH resource within the PUCCH bundle. The WTRU may transmit specific part in a PUCCH resource (e.g., all the PUCCH resources) within a PUCCH bundle. The WTRU may transmit a specific part in a PUCCH resource (e.g., all the PUCCH resources), for example, except the first PUCCH resource within the PUCCH bundle. The WTRU may receive configuration information indicating (e.g., be configured) conditions / rules for handling PUCCH resources within a PUCCH bundle. For example, the WTRU may transmit a (e.g., one) common part within the PUCCH bundle. The common part transmission may be performed before a (e.g., any) specific part transmission. The specific parts (e.g., all the specific parts) within the PUCCH bundle may be associated with a common part transmitted within the PUCCH bundle. The WTRU may receive configuration information indicating (e.g., be configured with) parameters (e.g., different parameters) for PUCCH bundle, for example, including but not limited to a number of PUCCH resources within a PUCCH bundle, time / frequency domain pattern / periodicity of PUCCH resources within a PUCCH bundle, periodicity of PUCCH bundle, offset of first PUCCH resource within PUCCH bundle, etc.

[0168] A resource type may be a function of a common and / or specific part.

[0169] The WTRU may transmit a common part using a first resource type (e.g., in a first resource type) and / or transmit the specific part using a second resource type (e.g., in a second resource type). The WTRU may transmit a common part using a PUSCH resource (e.g., in a PUSCH resource). For example, the WTRU may transmit specific part using a PUCCH resource. The WTRU may receive configuration information indicating (e.g., be configured with) a PUSCH resource (e.g., semi-persistent PUSCH resource) for common part transmission and a PUCCH resource for specific part transmission. The WTRU may receive configuration information indicating (e.g., be configured with) a link between the PUSCH resourcecarrying the common part and the PUCCH resource carrying the specific part. The WTRU may receive configuration information indicating (e.g., be configured for) different behavior, for example, based on the status of the PUSCH resource (e.g., semi-persistent PUSCH resource). For example, the WTRU may fallback to legacy CSI reporting, for example, if (e.g., when) the PUSCH resource (e.g., semi-persistent PUSCH resource) associated with common part is deactivated. For example, the WTRU may transmit (e.g., multi-part) TSF CSI feedback (e.g., common part using a semi-persistent PUSCH resource and specific part using a PUCCH resource), for example, if (e.g., when) the PUSCH resource (e.g., semi- persistent PUSCH resource) associated with common part is activated. The WTRU may transmit a common part in uplink control information (UCI) and / or a specific part in the same or different UCI. The WTRU may transmit a common part via signaling (e.g., in a MAC CE signaling) and specific part in UCI.

[0170] Resource selection / indication may be a function of a common part and / or specific part.

[0171] The WTRU may transmit a common part and specific part using a PUCCH resource (e.g., same or different PUCCH resource). The WTRU may receive configuration information indicating (e.g., be configured with) a PUCCH resource (e.g., plurality of PUCCH resources) for transmission of multi-part TSF CSI feedback (e.g., where an association may be preconfigured between PUCCH resources carrying common part feedback and PUCCH resources carrying specific part feedback). Such association may be defined, for example, in time domain and / or frequency domain. Such association may be used to (e.g., help) determine / indicate the common part that may be applicable for a specific part (e.g., so that the full CSI can be reconstructed at the gNB).

[0172] The WTRU may determine the mapping between a part within multi-part TSF and the PUCCH resource, for example, based on a condition and / or rule (e.g., preconfigured / predefined conditions / rules). For example, the WTRU may map a common part to a PUCCH resource whose payload is configured to be greater than or equal to the size of common part. For example, the WTRU may map a common part to a PUCCH resource whose periodicity matches to the periodicity preconfigured for common part. For example, the WTRU may map a common part to a PUCCH resource whose report format (e.g., quantization, quantization parameter) matches to the quantization (e.g., quantization parameter) (pre)configured for common part. Similar rules may be configured for specific part. The WTRU may indicate in the feedback whether the PUCCH carries the common part and / or specific part. For example, such indication may be included (e.g., explicitly included) in the CSI feedback, for example, via an indicator bit. Such indication may be indicated (e.g., implicitly indicated) by means of PUCCH resource selection among multiple configured PUCCH resources each associated with part type (e.g., common, or specific). For example, such indication may be implicitly indicated by means order of inclusion in the report content. For example, the common part may be included first followed by the specific part. The WTRU may transmitboth the common part and specific part using a single PUCCH resource, for example, where the WTRU may be transmit common part in a first CSI part (e.g., CSI part 1) and specific part in a second CSI part (e.g., CSI part 2) using (e.g., within) the same PUCCH resource.

[0173] Temporal-Spatial-Frequency CSI compression (e.g., with multi-resolution multi-part transmission) may be performed and / or enabled.

[0174] WTRU configurations (e.g., configuration information) for TSF multiresolution multipart structure may be received, indicated, and / or determined.

[0175] A WTRU may be capable of performing TSF based CSI compression. The WTRU (e.g., capable of performing TSF based CSI compression) may receive configuration information for the TSF operation, for example, which may include parameters for the TSF operation, configuration of the TSF CSI report (e.g., reporting parameters), parameters for determining the common part size (e.g., TSF-compressed CSI part parameter(s). The configuration or reconfiguration may be performed via signaling (e.g., RRC signaling).

[0176] Parameters for the TSF operation may be determined and / or received. Parameters for the TSF operation may be described as follows. The parameters for the TSF operation may include one or more of the following: ML Model(s) to be used for the TSF CSI compression; a number of parts in the latent domain; part sizes (e.g., these parameters may indicate the size of the common part and of the (e.g., each of the) specific part(s), for example, if (e.g., when) the WTRU is configured for multi-part operation (e.g., two or more parts)); quantization information (e.g., a quantization parameter) for a (e.g., each) part (e.g., common and one or more specific parts), for example, where the quantization information may include a quantization type (e.g., scalar of vector quantization, uniform of non-uniform quantization) or number of bits per latent element; a multiresolution indicator (e.g., whether the sizes of the parts (e.g., common and / or specific) are fixed or may be controlled dynamically); a validity time (e.g., associated with the common part size, for example, which may indicate a duration (e.g., how long) the WTRU may use the configured common part size (e.g., before it may be triggered or configured to determine an updated part size)), e.g., where the validity time may be expressed in time units (e.g., ms, or in number of slots, TTIs, frames, or as the number of CSI reports to use the configured common part size); etc.

[0177] Parameters for the TSF CSI report configuration may be determined and / or received. Parameters for the TSF CSI report configuration may be described as follows. The parameters for the TSF CSI report configuration may include one or more of the following: a common part reporting period (e.g., period between two consecutive CSI feedback reports that include the common part); a report size (e.g., the WTRU may be configured for separate reporting of the common part vs. the specific part(s), for example, where the WTRU may receive a separate report size for the common part reporting, and a report size forthe specific part(s) reporting; the WTRU may be configured for reporting multiple (e.g., all) parts in the same CSI feedback report, for example, where the report size may indicate the total size available for the common part and specific part / parts); report resources, e.g., for periodic and semi-persistent reporting; etc.

[0178] Parameters for determining the common part size may be determined and / or received. Parameters for determining the common part size may be described as follows.

[0179] A WTRU may determine a size of the common part (e.g., be enabled to use different procedures to determine the size of the common part). The WTRU may receive an indication from the NW for the method to use (e.g., channel based methods or methods based on the temporal variability of the common part), for example, for WTRUs supporting one or more methods to determine the size of the common part.

[0180] Channel-based methods may use one or more of the WTRU speed, the Doppler frequency, or the channel coherence time, for example, to determine the common part size. The WTRU may be configured with tables (e.g., one or more tables) mapping the configured channel metric (WTRU speed, Doppler frequency, channel coherence time) to the common part size.

[0181] The WTRU may be configured with a performance metric (e.g., with one or more thresholds corresponding to the performance metric), for example, for methods based on the temporal variability of the common part. For example, the performance metric may be associated with determining the amount of variability between the active common part and the inactive common part. The WTRU may use the metric to determine whether to update the common part or the part parameters, for example, as a function of the distance between the inactive common part and the active common part. Examples of metric may include one or more of the following: SGCS, NMSE, or Euclidean distance. The WTRU may be indicated whether to determine the metric in the latent domain, or in the full (i.e., input) domain.

[0182] For example, a performance threshold may be used. The performance threshold may indicate the threshold to be used for the configured performance metric, for example, for the WTRU to determine whether trigger conditions are met for parts parameter update. The WTRU may trigger a parts parameter update if (e.g., when) the measured SGCS is smaller than a configured (SGCS) threshold), for example, if SGCS is used as a performance metric. The WTRU may trigger a parts parameter update if (e.g., when) the measured NMSE is larger than a configured (NMSE) threshold, for example, if NMSE is used as performance metric.

[0183] TSF multipart parameters may be determined (e.g., by the WTRU).

[0184] A WTRU may be configured for TSF multipart compression using a model (e.g., two-sided AI / ML model). The TSF multipart compression may involve generation of parts (e.g., two or more parts) in the latent domain of the two-sided model, e.g., AE. For example, TSF compression with two parts may include a (e.g., one) common part that may represent a latent common representation in a sequence of CSIsamples (e.g., TSF sequence) and a specific part that captures the specific information for a sample (e.g., specific information in each sample). Herein, the generated CSI information associated with the two parts in the latent domain may be referred to as TSF-compressed CSI.

[0185] The WTRU may (e.g., be configured to) estimate the channel response, for example based on the received CSI-RS. The WTRU may (pre)process the estimated CSI, e.g., by deriving the eigenvectors of the channel. The WTRU may determine one or more parameters associated with the one or more parts, e.g., the common (e.g., long-term) part and the specific (e.g., short-term) part. The WTRU may perform TSF multipart compression, for example, based on the determined parameters. The WTRU may (e.g., be configured to) store the measured CSI or a preprocessed version (e.g., EV), for example, for the WTRU to determine (e.g., to assist the WTRU in determining) one or more of the parts parameters.

[0186] Parts parameters may be used and / or described herein.

[0187] The TSF parts parameters may include the common part parameter(s) and the specific part parameter(s). The common part parameter(s) may include one or more of the following: the size of the common part, e.g., expressed in number of latent elements; the quantization information (e.g., quantization parameter), e.g., expressed in number of bits per latent and type of quantization (e.g., where different elements may use different number of quantization bits); the validity time associated with the common part size (e.g., which may define how long the determined / indicated common part size may be used before the WTRU may be reconfigured to determine the updated size); a performance metric, e.g., measured longterm cosine similarity between the active common part and inactive common part; etc.

[0188] The specific part parameter(s) may include one or more of the following: the size of the specific part, e.g., expressed in number of latent elements; the quantization information (e.g., quantization parameter), e.g., expressed in number of bits per latent and type of quantization (e.g., where different elements may use different number of quantization bits); a performance metric, e.g., measured short-term cosine similarity between the current CSI at time t and time t-1; a common part parameters determination; etc.

[0189] Common part size determination may be performed and / or enabled.

[0190] The WTRU may (e.g., be configured to) determine and indicate the size of the common part based on one or more of the following: resource / reporting budget; channel parameters conditions; link performance; intermediate KPI (e.g., based on CSI measurements); etc.

[0191] For example, the WTRU may determine and indicate the size of the common part based on a resource / reporting budget. The WTRU may determine the common part size (e.g., \zc|) based on the maximum allocated budget for TSF reporting or the configured minimum and maximum allowed size for the common part,

[0192] For example, the WTRU may determine and indicate the size of the common part based on channel parameters conditions. The channel parameters may include Doppler frequency, WTRU speed and / or channel coherence time. The WTRU may determine the common part size based on one or more of the following: a preconfigured mapping between the WTRU speed and common part size; a preconfigured mapping between the Doppler frequency and common part size; a preconfigured mapping between the channel coherence time and common part size; etc. In any of the options (e.g., as described herein), the WTRU may measure a (e.g., any) channel parameter (e.g., speed) and select the corresponding size from a mapping table or a predefined relationship that relates the channel parameter (e.g., speed) with the size (e.g., |zc| = (v), where v may include the measured WTRU speed and f(.) may include the preconfigured function). For example, the size of the common size and the speed may be defined through an inverse relationship where increasing the WTRU speed implies reducing the common part size.

[0193] For example, the WTRU may determine and indicate the size of the common part based on link performance (e.g., a PDSCH).

[0194] In examples, the WTRU may determine the common part size based on one or more of the following: function of the PDSCH performance; measured CQI; etc.

[0195] The WTRU may determine the common part size based on a function of the PDSCH performance, e.g., BLER. The WTRU may reduce the common part size by x% (e.g., where x may be configured or determined by the WTRU), for example, if the number of consecutive NACKs exceeds a threshold (e.g., certain threshold). The WTRU may increase the common part size by x%, for example, if the number of consecutive ACKs exceeds one or more thresholds,.

[0196] The WTRU may determine the common part size based on a measured CQI. The WTRU may reduce the common part size, for example, if the measured CQI is lower than one or more thresholds. The WTRU may increase or fix the size, for example, otherwise.

[0197] For example, the WTRU may determine and indicate the size of the common part based on intermediate KPI (e.g., based on CSI measurements).

[0198] In examples, the WTRU may determine the common part size based on CSI measurements.... HNmay denote the CSI samples in a TSF sequence and z*, ... , z ' may denote the associated common parts generated for each sample in the TSF sequence. The common part (e.g., only the common part) associated with the first TSF sample (e.g., active common part) may be transmitted and / or used by a gNB for reconstruction of samples (e.g., all samples) in the TSF sequence. The other common parts (e.g., all the other common parts) may be generated at the WTRU side to determine the discrepancy between z and z , for i = 2, .. , N . In examples (e.g., an ideal scenario), zcl= zcfor i = 1, . . , N. There may be some errors depending on how well the model is trained. The WTRU may determine the common part sizeby performing one or more of the following measurements: measured long-term metric in the latent domain for the ith sample in the TSF sequence p^; measured long-term metric in the full domain for the ith sample in the TSF sequence p^p, etc.

[0199] The WTRU may determine the common part size by performing a measured long-term metric, (e.g., SGCS), in the latent domain for the ith sample in the TSF sequence p^, for example, which may be expressed as Eq. 2.

[0200] For the i-th compression instance, the WTRU may compare the measured long-term SGCS pmlcwith the configured long-term threshold. The WTUR may determine the common part size based on the difference between p^ and the configured threshold pcth. The WTRU may increase the common size by x% relative to the current size, for example, if p^ > pcth. The WTRU may decrease the common size by x% relative to the current size, for example otherwise. In examples, the WTRU may select the percentage increase / decrease based on the gap between pmlcand pcth.

[0201] The WTRU may determine the common part size by performing a measured long-term metric, (e.g., SGCS), in the full domain for the ith sample in the TSF sequence p^, for example, which may be expressed as Eq. 3.where utmay include the eigenvector associated with the i-th CSI sample in the TSF sequence. The WTRU may determine the common part size (e.g., increase / decrease with x% relative to the current size), for example, based on the gap between p^ and pcth.

[0202] In examples, the WTRU may determine the common part size based on a predefined table that maps the long-term metric to the common part size as a percentage of the maximum latent size, as shown in Table 1.Table 1 . Mapping between the long-term metric (e.g., SGCS) and the common part size expressed as a percentage of the maximum latent size

[0203] Common part quantization information (e.g., common part quantization parameter) may be determined.

[0204] The WTRU may determine the quantization information (e.g., quantization parameter, e.g., quantization type and / or number of bits) associated with the common part. The quantization types may include a uniform quantization, for example, where the common part elements / latents (e.g., all common part elements / latents) may be assigned the same number of bits. The quantization types may include a non-uniform quantization, for example, where different latents may be quantized using different number of bits. The non-uniform quantization may be used (e.g., beneficial) in the case of variable rate common part, for example, where the top latents may be allocated higher number of bits compared to the bottom latents. The WTRU may determine the quantization type based on the rate of change of each latent element in the common part.

[0205] A common part size validity time may be used, determined, and / or enabled.

[0206] The WTRU may be configured to determine the validity time associated with the determined common part size. This may indicate how long the determined and indicated common part size should be used by gNB, for example, until a size update is performed (e.g., required). In examples, the size update may occur as a supplement (e.g., incremental) on top of the common size used at the NW. In that case, the WTRU may transmit the common information associated with the incremental / delta latent elements. In examples, the size update may occur as a replacement of the total size used by the NW. In that case, the WTRU may determine and indicate the size (e.g., new size) and the associated common part CSI information.

[0207] In examples, the WTRU may determine the validity time of the common size based on the rate of change of the common part. For example, by measuring the difference betweenand z^ and comparing it with a preconfigured threshold. A smaller difference (e.g., compared to a larger difference) between the active common part derived from the first sample in the TSF sequence and the inactive common part associated with the second sample in a TSF sequence may be associated with the higher the validity timeof the determined size. The WTRU may determine the validity time based on a mapping (e.g., predefined mapping) between the measured quantity z - zc2and the validity time of the common part size. The validity time may be expressed in terms of the number of samples or in milli-seconds unit. The number of samples may be less than or equal to the number of samples in a TSF sequence.

[0208] Specific part parameters may be determined.

[0209] In examples, the WTRU may determine and indicate the size of the specific part based on one or more of the following: a resource / reporting budget, channel parameters conditions, intermediate KPI (e.g., based on CSI measurements), etc.

[0210] The WTRU may determine and indicate the size of the specific part based on a resource / reporting budget. The WTRU may determine the specific part size (e.g., |zs|) based on the allocated budget (e.g., maximum allocated budget) for TSF reporting (e.g., zmax), the allowed size (e.g., configured minimum and maximum allowed size) for the specific part (e.g., z™in< |zs| < z™ax), or the allocated budget (e.g., maximum allocated budget) along with the determined common part size (e.g., |zs| = zmax- |zc|).

[0211] The WTRU may determine and indicate the size of the specific part based on channel parameters condition(s). The channel parameters may include one or more of the following: a Doppler frequency, a WTRU speed, and / or a channel coherence time. The WTRU may determine the specific part size based on one or more of the following: a preconfigured mapping between the WTRU speed and specific part size; a preconfigured mapping between the Doppler frequency and specific part size; a preconfigured mapping between the channel coherence time and specific part size; etc. In any of the options (e.g., as described herein), the WTRU may measure a (e.g., any) channel parameter (e.g., speed) and select the corresponding size from a mapping table or a predefined relationship that relates the channel parameter (e.g., speed) with the specific part size (e.g., \zs| = gO), where v may include the measured WTRU speed and g(.) may include the preconfigured function). For example, the size of the specific size and the speed may be associated with (e.g., defined through) a linear or non-linear relationship, for example, where increasing the WTRU speed is associated with increasing the specific part size.

[0212] The WTRU may determine and / or indicate the size of the specific part, for example, based on intermediate KPI (e.g., based on CSI measurements). In examples, the WTRU may determine the specific part size based on CSI measurements.... HNmay denote the CSI samples in a TSF sequence and zj, ... , z^ may denote the associated specific parts generated for each sample in the TSF sequence. Different from the common parts, the specific parts associated with the different samples in the TSFsequence may be difference (e.g., entirely different). The WTRU may determine the specific part size by performing one or more of the following measurements.

[0213] The WTRU may determine the specific part size by performing a measured short-term metric (e.g., SGCS) in the full CSI domain for the i-th sample in the TSF sequence, p^ , which may be expressed as Eq. 4.The eigenvector associated with the i-th CSI sample in the TSF sequence may include itz. The WTRU may determine the specific part size (e.g., increase / decrease with x% relative to the current size), for example, based on the gap between pmlsand psth. In examples, the WTRU may determine the specific part size based on a predefined mapping between the measured correlation and the specific part size, as shown in Table 2. The higher the correlation between the consecutive samples may be associated with the smaller the specific part size, and vice versa.Table 2 - Mapping between the short-term metric (e.g., e.g., SGCS) and the specific part size expressed as a percentage of the maximum latent size.

[0214] Determination of the parts size may be performed, for example, based on the common part first determination or the specific part first determination.

[0215] In examples, the WTRU may determine (e.g., first determine) the common part size and then use its determination along with the maximum allocated uplink budget to compute the specific part size as |zs| =zmax- . In examples, the WTRU may determine the specific part size while considering (e.g., taking into consideration) that |zs| may (e.g., should) satisfy the condition c. In examples, the WTRU may determine (e.g., first determine) the specific part size. The WTRU may use (e.g., then use) its determination along with the allocated budget (e.g., maximum allocated budget) to compute the common part size as |zc| = zmax- |zs|. The WTRU may consider computing the common part size, for example, considering (e.g., taking into consideration or based on) the condition |zc| < zmax- |zs|. This may be beneficial in the case where the allocated budget (e.g., maximum allocated budget) for both parts needs to be reduced. In such a case, the WTRU may recommend the allocated uplink budget (e.g., maximum allocated uplink budget).

[0216] TSF parts parameters may be reported and / or indicated (e.g., by the WTRU).

[0217] The WTRU may be configured to i nd icate / report one or more parameters associated with the TSF multipart CSI feedback. In examples, the TSF multipart CSI feedback may include the common part parameter(s) and specific part parameter(s). The WTRU may be configured to indicate / report one or more of the common part parameter(s) and / or one or more of the specific part parameter(s). Herein, the common part parameter(s) may be used interchangeably with common parameters. The specific part parameter(s) may be used interchangeable with the specific parameters. The terms TSF multipart feedback report, TSF reporting, TSF CSI reporting may be used interchangeably, for example, to describe the CSI report carrying the TSF multipart parameters. The TSF multipart reporting may include (e.g., be divided into) common part reporting and / or specific part reporting, for example, where the common part reporting may include the common part CSI along with the common part parameters and the specific part reporting may include the specific part CSI along with the specific part parameters.

[0218] The common part parameters may be reported and / or indicated (e.g., by the WTRU).

[0219] The WTRU may report the common part parameters as part of the TSF multipart feedback report. The common part parameter(s) may include at least one of the following: a common part size, a common part size validity time, a common part quantization information (e.g., quantization parameter, such as, for example, quantization types and / or number of bits), a common part performance metric, an AI / ML model identification, an AI / ML model input (e.g., EV or full CSI), etc.

[0220] In examples, the WTRU may report the common part size expressed as the number of elements / latents. In examples, the WTRU may report the common part size as a fraction of the total number of elements / latents in the AI / ML encoder (e.g., 0.25). In examples, the WTRU may report the common part size as an update of the of the current used size (e.g., x% increase or decrease relative tothe current used size). The WTRU may report the quantization information associated with the common part size. The quantization information may include a quantization parameter, such as, for example, the quantization type (e.g., uniform or non-uniform) and / or the number of quantization bits. The number of quantization bits may be indicated to be the same for common latents (e.g., all common latents) or different number of bits per (e.g., each) latent.

[0221] In examples, the WTRU may report the validity time of the common part size. In examples, the validity time may be expressed in terms of integer number of sample, integer number of slots, or in time units (e.g., milliseconds). The validity time of the common part size may indicate how long the indicated common part size may be used. In examples, the common part size validity time may be indicated in different formats, for example, where a (e.g., each) format corresponds to different meaning. For example, format 1 may indicate that the reported common part size and the associated CSI information may be valid for n slots and after that a different common part size (e.g., new common part size) and CSI information may be transmitted. For example, format 2 may indicate that the reported common part size and the associated CSI information may be valid for a maximum of n2 slots but incremental / supplementary common size and associated CSI information, used on top of what is available at NW for improved reconstruction, may be signaled after m slots, where m < n2.

[0222] In examples, the WTRU may report a performance metric associated with the common part reporting. The performance metric may be determined based on one or more long-term metrics (e.g., as described herein with respect to common part parameters determination). For example, the performance metric may include one or more of the SGCS / NMSE between the active common part and inactive common part(s). In examples, the performance metric may include one or more of the SGCS / NMSE between the first CSI sample or the associated EV, in a TSF sequence, and the other CSI sample(s) or the associated EV in the same TSF sequence.

[0223] In examples, the WTRU may receive configuration information indicating (e.g., be configured with) periodic PUCCH / PUSCH resources for reporting the common part parameter(s) and / or specific part parameter(s). In examples, the WTRU may report a first set of common part parameter(s) (e.g., common part size and associated validity time) with periodic PUCCH / PUSCH transmission(s) and a second set of common part parameter(s) (e.g., performance metric) based on an aperiodic request from the gNB. In examples, the WTRU may report the one or more of the common part parameter(s) via signaling (e.g., in a MAC CE).

[0224] The specific part parameter(s) may be reported and / or indicated (e.g., by the WTRU).

[0225] The WTRU may report the specific part parameter(s) as part of the TSF multipart feedback report. The specific part parameter(s) may include one or more of the following: a specific part size, aspecific part quantization information (e.g., quantization parameter, such as, for example, quantization types and number of bits), a specific part performance metric, an AI / ML model identification, an AI / ML model input (e.g., EV or full CSI), etc.

[0226] In examples, the WTRU may report the specific part size expressed as the number of elements / latents. In examples, the WTRU may report the specific part size as a fraction of the total number of elements / latents in the AI / ML encoder (e.g., 0.25). In examples, the WTRU may report the specific part size as an update of the of the current used size, e.g., x% increase or decrease relative to the current specific part size. The WTRU may report the quantization information (e.g., quantization parameter) associated with the specific part size. The quantization information may include a quantization parameter, such as, for example, the quantization type (e.g., uniform or non-uniform) and / or the number of quantization bits.

[0227] In examples, the WTRU may report a performance metric associated with the specific part reporting. The performance metric may be determined based on one or more short-term metrics (e.g., as described herein with respect to specific part parameter(s) determination). For example, the performance metric may include one or more of the SGCS / NMSE between the i-th CSI sample, or (e.g., possibly) the associated EV, and the (i-1 )-th sample, or possibly the associated EV, in a TSF sequence.

[0228] In examples, the WTRU may receive configuration information indicating (e.g., be configured with) periodic PUCCH / PUSCH resources for reporting specific part parameter(s). In examples, the WTRU may report a first set of specific part parameter(s) (e.g., specific part size) with periodic PUCCH / PUSCH transmission(s) and a second set of specific part parameter(s) (e.g., performance metric) based on an aperiodic request from the gNB. In examples, the WTRU may report the one or more of the specific part parameter(s) via signaling (e.g., in a MAC CE).

[0229] The WTRU may send the TSF feedback report with parts (e.g., two parts). For example, a first part may be sent for the parameter(s). For example, the second part may be sent for TSF CSI information.

[0230] In examples, the WTRU may report TSF multipart CSI feedback with parts (e.g., two separate parts). The WTRU may report the determined common and specific parts parameter(s) (e.g., the size and quantization information) in a first part of the TSF CSI feedback report and the common part and specific parts CSI (e.g., described herein with respect to WTRU indication of TSF parts) in a second part of the TSF CSI feedback report. In examples, the WTRU may receive configuration information indicating (e.g., be configured with) periodic PUCCH / PUSCH resources for reporting the first part of the TSF feedback report. In examples, the WTRU may receive configuration information (e.g., be configured) for reporting the first part of the TSF feedback report based on an aperiodic request from the gNB.

[0231] A parts parameter(s) update may be triggered (e.g., for the WTRU to update parts parameters).

[0232] One or more of the following may constitute triggers for the WTRU to update the TSF parts parameters, e.g., number of TSF compressed CSI parts, size and / or size range of each part, number of symbols in each part, size ratio of common part to specific part, allocation of each transmitted / reported part, etc.

[0233] A trigger for a WTRU to update TSF parts parameters may include variations in the common latent part larger than a preconfigured threshold (e.g., measured in terms of number of data samples / symbols that are uncorrelated or measured in terms of number of data samples that are uncorrelated relative to the total number of data samples / symbols in the common part).

[0234] A trigger for a WTRU to update TSF parts parameter(s) may include variations in the specific latent part (e.g., if the variations between pairs of consecutive samples in a sequence of correlated data samples are larger than a preconfigured threshold or measured in terms of data samples and / or symbols).

[0235] A trigger for a WTRU to update TSF parts parameters may include a duration / length of variations in the latent part, for example, a number of consecutive samples where variations larger than a threshold are measured exceeding a preconfigured threshold.

[0236] A trigger for a WTRU to update TSF parts parameters may include a duration / length of variations in the common part.

[0237] A trigger for a WTRU to update TSF parts parameters may include a reception of an indication from the network. The indication may include one or more of the following: an update indication (e.g., simple update indication); an indication to update one or more parameters (e.g., size of each part); an indication of a decoder (e.g., new decoder) at the gNB; an indication of an update in the TSF configuration from the network; an indication from the gNB to switch to a different encoder, activate a different encoder, (re)train the encoder (e.g., WTRU may reassess the TSF parts and associated parameters).

[0238] A simple update indication may be received by the WTRU. The WTRU may have multiple (e.g., two) configurations for TSF compression (e.g., e.g., tsf_config1 , tsf_config2), such that a simple switch / update indication from the network may indicate the WTRU to switch from the one currently being applied. A configuration (e.g., tsf_config) may be associated with specific parameter(s) for the number of part, size of each part, allocation of each transmitted / reported part, etc.

[0239] An indication of a different decoder at the gNB may be received by the WTRU. The WTRU may refrain from performing (e.g., may not perform) TSF compression for a time (e.g., until it receives information on the tsf_config to apply and / or an update of parameters of the currently activated tsf_config), for example, if the WTRU receives an indication (e.g., a simple indication) of a different (e.g., new) decoder activated at the gNB. The indication of a different (e.g., new) decoder at the gNB may be transmitted with the associated update in parameters, e.g., size (e.g., new size) of common latent part.

[0240] An indication of an update in the TSF configuration from the network may be received by the WTRU. For example, the WTRU may receive an update in the TSF configuration with different (e.g., new) parameter(s) for number of parts, size of each part, quantization information, etc. (e.g., new tsf_config3, new tsf_config4 and / or an update of tsf_config1 / tsf_config2 where the initial configuration may have already been received previously in RRC (re)configuration).

[0241] A trigger for a WTRU to update TSF parts parameters may include a WTRU measurement of channel conditions. The WTRU may measure channel conditions (e.g., cell RSRP, L1-RSRP, Doppler, channel coherence time, channel coherence bandwidth, CSI parameters e.g., CQI, PMI, Rl, etc.). A change (e.g., sudden change) in the channel quality may trigger the WTRU to switch to a different configuration or to update some parameters within a configuration, e.g., a degradation in channel within a short time period may trigger the WTRU to use a larger size for the specific latent part.

[0242] A trigger for a WTRU to update TSF parts parameters may include a WTRU measurement of WTRU speed / movement / location / positioning. At high speeds, the WTRU may refrain from using (e.g., may not use) TSF compression (e.g., any TSF compression) or may adapt the parameters of the TSF configuration accordingly. At a different (e.g., new) location, the WTRU may determine to update the TSF parts and the one or more associated parameters (e.g., size, quantization info)

[0243] A trigger for a WTRU to update TSF parts parameters may include an WTRU activation of a different AI / ML encoder (e.g., new AI / ML encoder).

[0244] A trigger for a WTRU to update TSF parts parameters may include a WTRU switching to a different AI / ML encoder.

[0245] TSF type determination may be performed. Multipart transmission with controllable part-specific overhead WTRU configurations for time-dependent multipart generation may be performed.

[0246] The latent structure may be categorized (e.g., into cases / types), for example, as TSF types. Fixed / variable may be used to indicate whether the size of the part is fixed or variable. Common / specific may be used to indicate whether the information associated with the part is shared / reusable / common across multiple CSI samples or specific for a (e.g., each) sample.

[0247] Common-fixed specific-fixed may include where the latent domain is structured to include a common part of fixed size and a specific part of fixed size. Common-fixed specific-variable may include where the latent domain may include (e.g., may be structured to include) a common part (e.g., common part of fixed size) and a specific part (e.g., specific part of variable size). Common-variable specific-fixed may include where the latent domain is structured to include a common part of variable size and a specific part of fixed size. Common-variable specific-variable may include where the latent domain may include (e.g., may be structured to include) a common part of variable size and a specific part of variable size.

[0248] TSF types and parameters may be configured.

[0249] A TSF CSI reporting may be defined or used. A CSI reporting may be determined or defined as TSF CSI, for example, if one or more following condition is met: a CSI is generated, derived, or estimated (e.g., based on previously reported CSI); a portion of CSI was reported at a time (e.g., in an earlier time) and the rest of the part of CSI is reported;, a receiver uses (e.g., needs to use) reported CSIs over multiple time instances, for example, to reconstruct CSI; a CSI is constructed with multiple parts, for example, where a first part may be considered as common part (e.g., long-term part) and a second part may be considered as specific part (e.g., short-term part), and where the common part may be used or valid over multiple CSI reporting instances and the specific part may be used or valid at the time instance where the specific part is reported (e.g., a common part may be fixed size or variable size and / or a specific part may be fixed size or variable size); a CSI is compressed with an AI / ML model (e.g., two-sided model); etc.

[0250] One or more TSF types may be used, defined, or configured for a CSI reporting. A TSF type may be determined or defined based on one or more of following: a number of parts used, determined, or defined to construct a CSI; whether fixed or variable size is used for each part configured or used; maximum and / or minimum size of each part configured or used; uplink resource type used for CSI reporting (e.g., PUSCH, PUCCH, etc.); etc.

[0251] A WTRU may be configured with one or more of the following for TSF CSI reporting.

[0252] A WTRU may be configured (e.g., for TSF CSI reporting) with a (e.g., at least one) common part and a (e.g., one or more) specific part.

[0253] A WTRU may be configured (e.g., for TSF CSI reporting) with a size of a common part and / or a size of one or more specific parts, for example, if the common part size or the specific part size is fixed. A range of the common part size or the specific part may be indicated, configured, predefined, or used, for example, otherwise,. The WTRU may determine the size of the common part or the specific part within the range based on one or more of following: WTRU speed, channel correlation in time domain; time domain channel property (TDCP) value associated with the reporting; etc. In examples, a TDCP value may be reported together with the common part. The common part size may be determined based on the TDCP value. The common part size may be relatively large, for example, if (e.g., when) a TDCP value is high,; otherwise, the common part size may be relatively small (e.g., or vice-versa).

[0254] A WTRU may be configured (e.g., for TSF CSI reporting) that a range of the common part size or the specific part size (e.g., that is indicated, configured, or used) may be considered as variable size.

[0255] A WTRU may be configured (e.g., for TSF CSI reporting) with an uplink resource to use for the reporting of common part and / or specific part(s). The uplink resource (e.g., PUCCH, PUSCH, PRACH, etc.) may be different for common part and specific part. The common part may be reported with a more reliableuplink resource or lower coding rate as compared to the specific part (e.g., considering that an erroneous transmission / reception of a common part may result in consecutive error on CSI reporting).

[0256] A WTRU may be configured (e.g., for TSF CSI reporting) with uplink reporting characteristic(s) (e.g., periodic, aperiodic, semi-persistent) for one or more common parts and specific parts. A common part may be reported periodically while one or more specific parts may be reported semi-persistently or aperiodically.

[0257] A WTRU may be configured (e.g., for TSF CSI reporting) such that a cyclic redundancy check (CRC) may be attached to a common part while no CRC may be attached to a specific part. Therefore, a receiver may detect an error for the common part reception, for example, while the specific part may be used without checking an error.

[0258] A WTRU may be configured (e.g., for TSF CSI reporting) with a reporting duration / cycle for one or more of common parts and specific parts. The reporting duration / cycle may be different across common parts and / or specific parts. A common part reporting cycle may be longer than a specific part. Duration, cycle, period, or gap may be used interchangeably.

[0259] A WTRU may be configured (e.g., for TSF CSI reporting) with a prioritization of parts. The WTRU may be configured with priorities across one or more common and specific part, for example, if (e.g., when) a WTRU reports (e.g., needs to report) multiple parts. Prioritization may (e.g., alternatively) be predefined or predetermined. One or more of following may apply: a common part may be a higher priority than a specific part; a common part which is the same as the latest reported common part may be lower priority than a specific part (e.g., in this case, a WTRU may indicate that the reported common part can be reused); if (e.g., when) more than one specific part is used, or reported, the specific part with a lower part index may be considered as a higher priority; a WTRU may drop a lower priority part when the uplink resource allocated or granted is not enough to carry the parts determined for the reporting at the instance; etc.

[0260] A WTRU may fall back to SF CSI from TSF CSI, for example, if (e.g., when) one or more of following condition is met (e.g., where SF CSI may be referred to as a CSI reporting which is independent from the previously reported CSI): if (e.g., when) a WTRU received consecutive NACKs (e.g., a number of consecutive NACKs) for the reported CSI (e.g., common parts); if a channel time correlation is lower than a threshold (e.g., a TDCP value is lower than a threshold); if (e.g., when) a WTRU received a fallback indication; if a common part change is larger than a threshold within a time window; if a channel reconstruction error is higher than a threshold; if a TSF compression error is higher than a threshold; etc.

[0261] TSF type and parameters may be determined (e.g., by a WTRU).

[0262] Details for a WTRU capable of TSF based compression may be described herein.

[0263] A WTRU with TSF based compression capability may receive a CSI-RS transmission and / or estimate the CSI. The WTRU may generate an input for the TSF based compression model, for example depending on the configured input domain type. The input domain type may be a channel representation such as channel matrix, channel eigenvectors, or processed channel matrix, etc. The WTRU may store the input to the TSF based compression (e.g., channel matrix, eigenvector), for example, if configured.

[0264] The WTRU may activate and use a TSF compression model, for example, based on the configured default TSF type and parameters. In examples, different types (e.g., four different types) of TSF based compression may be used and / or provided (e.g., as shown in FIG. 13). The WTRU may be configured with one of the types of TSF based compression as described herein.

[0265] The WTRU may be configured with common-fixed and specific fixed TSF based compression (e.g., TSF Type 1). The latent structure of the encoder may include a common part of fixed size |zc| and a specific part of fixed size |zz|, for example, in TSF Type 1.

[0266] The WTRU may be configured with common-fixed and specific-variable TSF based compression (e.g., TSF Type 2). The latent structure of the encoder may include a common part of fixed size |zc| and a specific part of variable size between zfand (|zz| + \ziv|), for example, in TSF Type-2.

[0267] The WTRU may be configured with common-variable and specific-fixed TSF based compression (e.g., TSF Type 3). The latent structure of the encoder may include a common part of variable size between |zc| and (|zc| + |zcv|) and a specific part of fixed size |z , for example, in TSF Type-3.

[0268] The WTRU may be configured with common-variable and specific-variable TSF based compression (e.g., TSF Type 4). The latent structure of the encoder may include a common part of variable size between |zc| and (|zc| + |zv|) and a specific part of variable size betweenfor example, in TSF Type-3.

[0269] The WTRU may generate the TSF-compressed CSI parts (e.g., common and specific), for example, based on a configured first TSF type and associated parameters. For example, the associated parameters may be the size common and specific parts of the TSF compression. The WTRU may store one or more parts of the TSF-compressed CSI.

[0270] The WTRU may determine a second type of a TSF compression and associated parameter(s). The second type of TSF compression may be, for example, one of Type-1 , Type-2, Type-3 or Type-4. The associated parameters may be for example sizes of common part and specific part and / or quantization information (e.g., quantization parameter(s)) associated with a (e.g., each) part. The WTRU may determine a second type of TSF compression and associated parameter(s), for example, based on the preconfigured conditions and measurements.

[0271] The WTRU may determine the second TSF type, for example based on one or more of the following.

[0272] The WTRU may determine the second TSF type, for example, based on channel parameter(s) (e.g., Doppler, WTRU speed). For example, the WTRU may calculate an estimation of the Doppler using channel measurement and compare the estimated Doppler with a configured one or more thresholds to determine the second TSF type (e.g., each Doppler range maps to a specific type). The WTRU may determine TSF Type 1 , for example, if the WTRU is within a stable environment (e.g., no change in the WTRU speed or WTRU speed / Doppler below a configured threshold). The WTRU may determine TSF Type 4, for example, if the WTRU is with a highly dynamic environment (e.g., range of WTRU speed / Doppler above a configured threshold).

[0273] The WTRU may determine the second TSF type, for example, based on CSI measurements. For example, the WTRU may calculate long / short-term SGCS for the common and specific parts. The WTRU may compare the calculated long / short-term SGCS with the configured long / short-term SGCS threshold, for example, to determine the identity of the common and specific parts (e.g., fixed or variable). The WTRU may determine the identity of the common part as fixed, for example, if the long-term SGCS is higher than the configured long-term SGCS. Otherwise, the WTRU may determine the identity of the common part as variable. The WTRU may determine the identity of the specific part as fixed, for example, if the short-term SGCS is higher than the configured short-term SGCS. Otherwise, the WTRU may determine the identity of the specific part as variable.

[0274] The WTRU may determine the second TSF type, for example, based on the first TSF type. For example, the WTRU may determine the second TSF type relative to the first TSF type. The WTRU may determine a TSF type with a higher index (e.g., if the first TSF type is Type-2, then the WTRU may select Type-3 for the second TSF type), for example, if the WTRU measures an increase in the WTRU speed / Doppler. The WTRU may determine a TSF type with a lower index (e.g., if the first TSF type is 2, then the WTRU may select Type-1 for the second TSF type), for example, if the WTRU measures a decrease in the WTRU speed / Doppler.

[0275] The WTRU may determine the associated parameters for the determined TSF type. For example, the associated parameters may include part size and quantization information (e.g., quantization parameter(s)) per part. The WTRU may determine the associated parameter(s), for example based on one or more of the following.

[0276] The WTRU may determine the associated parameter(s), for example, such as part size, based on the WTRU speed / Doppler. The WTRU may determine the size of specific part as a function of the WTRU speed (e.g., such that the size increases with increasing WTRU speed), for example, if thedetermined TSF type is Type-2. The WTRU may determine the size of common part as a function of the WTRU speed (e.g., such that the size decreases with decreasing WTRU speed), for example, if the determined TSF type is Type-3. The WTRU may determine the part sizes as a function of the WTRU speed (e.g., such that common part size decreases and specific part size increases with increasing WTRU speed), for example, if the determined TSF type is Type-4.

[0277] The WTRU may determine the associated parameter(s), for example, such as quantization level, based on the WTRU speed / doppler. For example, the WTRU may determine the quantization level of common part and specific part as a function of the WTRU speed (e.g., such that the quantization level increases with increasing WTRU speed).

[0278] The WTRU may determine the associated parameter(s), for example, such as part size, based on CSI measurements. The WTRU may determine the size of specific part as function of the short-term SGCS (e.g., such that the size increases with a decrease in the short-term SGCS), for example, if the determined TSF type is Type-2. The WTRU may determine the size of common part as function of the long-term SGCS (e.g., such that the size increases with a decrease in the long-term SGCS), for example, if the determined TSF type is Type-3. The WTRU may determine the size of common part and specific part as function of the short-term and long-term SGCS (e.g., such that the common part size increases with a decrease in the long-term SGCS and the specific part size increases with a decrease in the short-term SGCS), for example, if the determined TSF type is Type-4.

[0279] The WTRU may determine the associated parameter(s), for example, such as quantization level, based on short / long-term SGCS. The WTRU may determine the quantization level of common part as a function of long-term SGCS, for example, such that the quantization level of common part increases with a decrease in the long-term SGCS. The WTRU may determine the quantization level of specific part as a function of short-term SGCS, for example, such that the quantization level of specific part increases with a decrease in the short-term SGCS.

[0280] The WTRU may determine the associated parameter(s), for example, such as part size, based on latent measurements. For example, if the determined TSF type is Type-2 or Type-4, where the common part is variable, the WTRU may decrease the size of the common part (e.g., if the average difference between the multiple consecutive latent measurements exceeds a threshold).

[0281] The WTRU may determine the associated parameter(s), for example, such as quantization level, based on latent measurements. The WTRU may determine the quantization level of common part as a function of latent measurements, for example, such that the quantization level increases with an increase in the average difference of the multiple consecutive latent measurements.

[0282] The WTRU may represent the part size parameters, for example, based on one more of the following: the actual size of the common and variable parts; the percentage of the common and variable parts with respect the to the full latent dimension; the incremental size of the variable parts of the latent compared to the previously transmitted latent; etc.

[0283] The TSF type and parameters may be indicated (e.g., by the WTRU).

[0284] The WTRU may report the selected and applied TSF type. The WTRU may indicate a supported types (e.g., Type 1, Type 2, Type 3 and Type 4). The WTRU may indicate which AI / ML model for TSF compression is applied (e.g., including which TSF type is selected among the available types in the activated AI / ML model). The WTRU may report the values of the parameters associated with the TSF type (e.g., the number of parts, the size of the fixed common-part and specific-parts, and the size of the variable common-part and specific-parts). The WTRU may report the sizes of common-part and specific-part as a (e.g., one) percentage, e.g., percentage of the size of the common part with respect to the latent dimension, or percentage of one or more specific-parts with respect to the latent dimension.

[0285] The WTRU may report the selected compression ratio for both common-part and specific-part, for example, if the TSF AIML model has a variable / adaptive compression ratio.

[0286] The WTRU may indicate the quantization level applied for a (e.g., each) part of the TSF CSI feedback. The WTRU may indicate the quantization function (e.g., uniform or non-uniform) and the quantization level (e.g., quantization bits).

[0287] The WTRU may report TSF multipart CSI feedback with parts (e.g., two separate parts). The WTRU may report the common part and specific parts CSI (e.g., described herein) in a first part (e.g., part 1) of the TSF CSI feedback report. The generated parts may be based on the configured TSF parameters (e.g., TSF type and associated parameters). The preferred TSF type and associated parameters may be reported in a second part (e.g., in part 2) of the TSF CSI feedback report.

[0288] TSF type determination and multipart transmission with controllable part-specific overhead may be performed.

[0289] FIG. 13 illustrates example TSF Types.

[0290] In some scenarios (e.g., static environment), a TSF model with fixed size parts trained on a specific speed may be used (e.g., may be efficient) for a WTRU (e.g., particular WTRU) moving at that speed. In other scenarios (e.g., a highly dynamic scenario where the WTRU speed may be changing over a wide range), another TSF model with flexibility in controlling the parts sizes may be used (e.g., required) to cope with the dynamic nature of the environment. Different TSF types (e.g., four different TSF types) may be used to handle different scenarios, fixed size parts, variable size parts, one fixed, and one variable. TheTSF type and the associated parameters may be selected / determined and / or reported (e.g., as shown with respect to FIG. 6).

[0291] A WTRU performing TSF based CSI compression may determine the TSF type and associated parameters (e.g., the size of the long-term and short-term component in the latent space and the associated quantization with each part), for example, as a function of channel conditions and / or configured metrics thresholds and may report the determined TSF type and parameters.

[0292] The WTRU may receive configuration information indicating (e.g., configured with) one or more parameters to generate CSI feedback, for example, based on temporal-spatial-frequency (TSF) compression (e.g., using a two-sided model). The TSF-compressed CSI may include multiple parts. The TSF-compressed CSI may include a long-term / common part, for example, which may capture the longterm variations in a given sequence of correlated samples. The long-term / common part may be sent in periodic CSI feedback transmissions (e.g., once every N, possibly configured, CSI feedback transmissions). The long-term / common part may be stored and reused for a subsequent slot (e.g., consecutive N slots). The long-term / common part may have controllable / variable size feedback overhead, e.g., size range. The TSF-compressed CSI may include a short-term / specific part. The short-term / specific part may capture the short-term variations between every pair of consecutive samples in a given sequence of N correlated samples. The short-term / specific part may be sent in a (e.g., every) CSI feedback transmission. The short-term / specific part may have controllable / variable-size feedback overhead, e.g., size range.

[0293] The WTRU may receive configuration information indicating one or more configuration parameters associated with the TSF compression, which may include one or more of the following: TSF configuration and parameters thereof; conditions / parameters for TSF latents / parts determination or selection; reporting configuration information; etc.

[0294] The configuration parameters associated with TSF compression may include TSF configuration information and parameters thereof. The configuration parameters may include TSF types, a default type (e.g., TSF Type 1), type-specific information (e.g., number of parts, size range of each part, quantization info for each part associated with each type), or input domain type (e.g., EV or full CSI). TSF types (e.g., based on the size of each part, for example, fixed or variable) may include one or more of the following: common-fixed and specific-fixed (e.g., TSF Type 1); common-fixed and specific-variable (e.g., TSF Type 2); common-variable and specific-fixed (e.g., TSF Type 3); common-variable and specific-variable (e.g., TSF Type 4).

[0295] The configuration parameters associated with TSF compression may include conditions / parameters for TSF latents / parts determination / selection. For example, theconditions / parameters may include a short-term performance metric threshold (e.g., short term SGCS / NMSE threshold: between every two consecutive CSI samples). For example, the conditions / parameters may include a long-term performance metric threshold (e.g., long-term SGCS / NMSE threshold: between current CSI sample and first CSI sample in a TSF sequence).

[0296] The configuration parameters associated with TSF compression may include reporting configuration information (e.g., part-specific reporting, e.g., MCS and / or quantization, UL resources)

[0297] The WTRU may receive a CSI-RS transmission. The WTRU may estimate the channel and (e.g., optionally) stores the measured CSI (e.g., full channel) .

[0298] The WTRU may generate the TSF-compressed CSI parts (e.g., common and specific), for example, based on a configured first TSF type and associated parameters. The WTRU may (e.g., optionally) store one or more parts in the corresponding buffer(s) (e.g., update the long-term buffer or shortterm-buffer or both).

[0299] The WTRU may select / recommend a second (e.g., preferred) TSF type, and associated parameters (e.g., part size, quantization associated with each part) based on preconfigured conditions and measurements. For example, the WTRU may determine the TSF type (e.g., preferred TSF type), based on one or more of the following. The WTRU may determine the TSF type (e.g., preferred TSF type) based on measured channel parameters (e.g., Doppler, WTRU speed). For example the WTRU may determine the type based on the estimated Doppler and comparing it with one or more threshold(s). A (e.g., each) Doppler range may map to a type (e.g., specific type). For example, TSF Type 1 may be used in a stable environment (e.g., no change in the WTRU speed). TSF Type 4 may be used in a dynamic (e.g., highly dynamic) environment (e.g., wide range of speed). The WTRU may determine the preferred TSF type based on CSI measurements. For example, the WTRU may compare the measured long / short-term SGCS with the configured long / short-term SGCS threshold to determine the identity of the long / short term part (e.g., fixed or variable).

[0300] The WTRU may determine the parameter(s) associated with the selected type (e.g., part size, part-specific quantization information). The part size associated with the selected type may be transmitted in one or more of the following formats: percent with respect to the full size of the common part, incremental with respect to the latest transmitted overlap from zv(e.g., a number of latents / elements that may be fully attached to the common part, for example, to increase its size to zc+ zv„ or full attached to the specific part to increase its size to zs+ zv)). The WTRU may determine the parameters associated with the selected type based on one or more of the following: WTRU speed (e.g., reduce common part size and increase specific part size for high speeds and vice versa); CSI measurements (e.g., comparing the measured short-term SGCS with the configured long-term SGCS threshold); long-term latentmeasurements (e.g., decrease the common part size if there is a discrepancy in the long-term latent generated from multiple consecutive CSI samples).

[0301] The WTRU may report the TSF-compressed CSI (e.g., compressed using the configured first TSF Type) and the selected second TSF type and associated parameters. The WTRU may report the TSF type index and associated parameters (e.g., size, number of parts, overlap information).

[0302] Although features and elements described above are described in particular combinations, each feature or element may be used alone without the other features and elements of the preferred embodiments, or in various combinations with or without other features and elements.

[0303] Although the implementations described herein may consider 3GPP specific protocols, it is understood that the implementations described herein are not restricted to this scenario and may be applicable to other wireless systems. For example, although the solutions described herein consider LTE, LTE-A, New Radio (NR) or 5G specific protocols, it is understood that the solutions described herein are not restricted to this scenario and are applicable to other wireless systems as well.

[0304] The processes described above may be implemented in a computer program, software, and / or firmware incorporated in a computer-readable medium for execution by a computer and / or processor. Examples of computer-readable media include, but are not limited to, electronic signals (transmitted over wired and / or wireless connections) and / or computer-readable storage media. Examples of computer- readable storage media include, but are not limited to, a read only memory (ROM), a random access memory (RAM), a register, cache memory, semiconductor memory devices, magnetic media such as, but not limited to, internal hard disks and removable disks, magneto-optical media, and / or optical media such as compact disc (CD)-ROM disks, and / or digital versatile disks (DVDs). A processor in association with software may be used to implement a radio frequency transceiver for use in a WTRU, terminal, base station, RNC, and / or any host computer.

Claims

CLAIMSWhat Is Claimed Is:1 . A wireless transmit / receive unit (WTRU) comprising, a processor configured to: receive configuration information associated with temporal-spatial-frequency (TSF) compression; determine a first TSF parameter based on at least the configuration information and a measurement associated with a CSI-RS; determine a second TSF parameter based on at least the configuration information and the measurement associated with the CSI-RS; determine a first TSF-compressed CSI part based on the first TSF parameter; determine a second TSF-compressed CSI part based on the second TSF parameter; and send, to a network, at least one of the first TSF-compressed CSI part, the first TSF parameter, the second TSF-compressed CSI part, or the second TSF parameter.

2. The WTRU of claim 1 , wherein the configuration information associated with TSF compression comprises a configuration parameter associated with TSF compression, wherein the configuration parameter associated with TSF compression comprises at least one of a number of TSF-compressed CSI parts, a minimum CSI part size, a maximum CSI part size, or a part allocation determination and compression indication.

3. The WTRU of claim 1 , wherein the configuration information associated with TSF compression comprises reporting configuration information.

4. The WTRU of claim 1 , wherein the at least one of the first TSF-compressed CSI part, the first TSF parameter, the second TSF-compressed CSI part, or the second TSF parameter is sent based on the reporting configuration information.

5. The WTRU of claim 1 , wherein the first TSF parameter is at least one of a first size, a validity time, or first quantization parameter, and wherein the second TSF parameter is at least one of a second size or second quantization parameter.

6. The WTRU of claim 1 , wherein the determination of the first TSF parameter is further based on a first measured channel parameter or a first look up table, and wherein the determination of the second TSF parameter is further based on a second measured channel parameter or a second look up table.

7. The WTRU of claim 1 , wherein the first TSF-compressed CSI part is a long-term CSI part and the first TSF parameter is associated with the long-term CSI part, and wherein the second TSF-compressed CSI part is a short-term CSI part and the second TSF parameter is associated with the short-term CSI part.

8. The WTRU of claim 1 , wherein the configuration information associated with TSF compression comprises a channel condition associated with TSF part determination, wherein the channel condition is a metric threshold associated with TSF part determination, and wherein the first TSF-compressed CSI part is further determined based on the metric threshold, and wherein the second TSF-compressed CSI part is further determined based on the metric threshold.

9. The WTRU of claim 8, wherein the metric threshold is one of a short-term performance metric threshold, a long-term performance metric threshold, or a combined performance metric threshold.

10. A method comprising: receiving configuration information associated with temporal-spatial-frequency (TSF) compression; determining a first TSF parameter based on at least the configuration information and a measurement associated with a CSI-RS; determining a second TSF parameter based on at least the configuration information and the measurement associated with the CSI-RS; determining a first TSF-compressed CSI part based on the first TSF parameter; determining a second TSF-compressed CSI part based on the second TSF parameter; and sending, to a network, at least one of the first TSF-compressed CSI part, the first TSF parameter, the second TSF-compressed CSI part, or the second TSF parameter.11 . The method of claim 10, wherein the configuration information associated with TSF compression comprises a configuration parameter associated with TSF compression, wherein the configuration parameter associated with TSF compression comprises at least one of a number of TSF-compressed CSIparts, a minimum CSI part size, a maximum CSI part size, or a part allocation determination and compression indication.

12. The method of claim 10, wherein the configuration information associated with TSF compression comprises reporting configuration information.

13. The method of claim 12, wherein the at least one of the first TSF-compressed CSI part, the first TSF parameter, the second TSF-compressed CSI part, or the second TSF parameter is sent based on the reporting configuration information.

14. The method of claim 10, wherein the first TSF parameter is at least one of a first size, a validity time, or first quantization parameter, and wherein the second TSF parameter is at least one of a second size or second quantization parameter.

15. The method of claim 10, wherein the determination of the first TSF parameter is further based on a first measured channel parameter or a first look up table, and wherein the determination of the second TSF parameter is further based on a second measured channel parameter or a second look up table.

16. The method of claim 10, wherein the first TSF-compressed CSI part is a long-term CSI part and the first TSF parameter is associated with the long-term CSI part, and wherein the second TSF- compressed CSI part is a short-term CSI part and the second TSF parameter is associated with the shortterm CSI part.

17. The method of claim 10, wherein the configuration information associated with TSF compression comprises a channel condition associated with TSF part determination, wherein the channel condition is a metric threshold associated with TSF part determination, and wherein the first TSF-compressed CSI part is further determined based on the metric threshold, and wherein the second TSF-compressed CSI part is further determined based on the metric threshold.

18. The method of claim 17, wherein the metric threshold is one of a short-term performance metric threshold, a long-term performance metric threshold, or a combined performance metric threshold.

Images