Systems and methods for model monitoring for channel state information prediction and compression

Abstract

A wireless transmit receive unit (WTRU) can receive configuration information for channel state information (CSI) prediction and compression. The WTRU can measure a CSI prediction error and / or a CSI compression error, e.g., to obtain a total combined CSI error. The total combined CSI error can include a sum of the CSI prediction error and the CSI compression error. The WTRU can determine a CSI error condition type, e.g., based on the total combined CSI error. The CSI error condition type can include one or more of a CSI prediction limited condition type, a CSI compression limited condition type, and / or a combined CSI prediction and compression limited condition type. The WTRU can send a message to, e.g., a network and / or a network entity. The message can include an indication of the determined CSI error condition type.

Description

Cross-references to related applications

[0001] This application claims the benefit of U.S. Patent Application 18 / 623,185, filed April 1, 2024, the entire contents of which are incorporated herein by reference. Background Technology

[0002] Channel state information (CSI) enhancement based on artificial intelligence / machine learning (AI / ML) can include reducing CSI feedback reporting overhead. AI / ML-based CSI prediction can mitigate the effects of channel aging and / or reduce CSI feedback overhead.

[0003] Compared to non-AI / ML CSI feedback, AI / ML-based CSI compression can reduce overhead; however, AI / ML-based CSI compression needs further improvement. Summary of the Invention

[0004] A Wireless Transmit / Receive Unit (WTRU) can receive configuration information for Channel State Information (CSI) prediction and compression. The WTRU can measure CSI prediction error and / or CSI compression error, for example, to obtain a total combined CSI error. The total combined CSI error can include the sum of the CSI prediction error and the CSI compression error. The WTRU can determine the CSI error condition type, for example, based on the total combined CSI error. The CSI error condition type can include one or more of a CSI prediction-limited condition type, a CSI compression-limited condition type, and / or a combination of both. The WTRU can send a message to, for example, a network (NW) and / or a network entity (NE). This message can include an indication of the determined CSI error condition type. Although described in the context of CSI error conditions, multiple CSI error conditions may be used in some examples.

[0005] The WTRU can determine the relative CSI error, for example, based on one or more of the total combined CSI error, and the CSI prediction error and CSI compression error. The relative CSI error can include one or more of the relative CSI prediction error and / or the relative CSI compression error. For example, when the relative CSI prediction error exceeds a threshold associated with the total combined CSI error, the WTRU can determine the CSI error type as a CSI prediction-limited condition. The WTRU can determine the time window associated with the CSI prediction. For example, when the relative CSI compression error exceeds a threshold associated with the total combined CSI error, the WTRU can determine the CSI error condition type as a CSI compression-limited condition.

[0006] For example, when the difference between the CSI prediction error and the CSI compression error is less than a threshold, the WTRU can determine that the CSI error type is a combined CSI prediction and compression-limited condition type. The WTRU can determine a CSI compression configuration that includes a CSI instance threshold. The message may additionally or alternatively include one or more of the determined CSI compression configuration and / or associated indexes. Additionally or alternatively, the message may include one or more indications of the measured CSI prediction error and / or CSI compression error. The WTRU can determine parameters associated with CSI prediction, for example, based on one or more of the measured CSI prediction error and / or CSI compression error and / or based on the CSI prediction configuration. Attached Figure Description

[0007] Figure 1A This is a system diagram illustrating an exemplary communication system that can implement one or more of the disclosed embodiments.

[0008] Figure 1B This illustrates a method according to one embodiment. Figure 1A The system diagram shows an exemplary wireless transmit / receive unit (WTRU) used in the communication system.

[0009] Figure 1C This illustrates a method according to one embodiment. Figure 1A The system diagram shows an exemplary radio access network (RAN) and an exemplary core network (CN) used in the communication system shown.

[0010] Figure 1D This illustrates a method according to one embodiment. Figure 1A The system diagram shows another exemplary RAN and another exemplary CN used in the communication system shown.

[0011] Figure 2 This is a system diagram showing an example of WTRU and NW.

[0012] Figure 3 This is an example of a CSI procedure.

[0013] Figure 4 This is another example of the CSI process.

[0014] Figure 5A An example of prediction error relative to the index is shown.

[0015] Figure 5B An example table is shown.

[0016] Figure 6 Another example of the CSI process is shown. Detailed Implementation

[0017] Figure 1AThis diagram illustrates an example communication system 100 in which one or more of the disclosed embodiments may be implemented. The communication system 100 may be a multiple access system that provides content such as voice, data, video, messaging, and broadcasting to multiple wireless users. The communication system 100 enables multiple wireless users to access such content by sharing system resources, including wireless bandwidth. For example, the communication system 100 may employ one or more channel access methods, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal FDMA (OFDMA), Single Carrier FDMA (SC-FDMA), Zero-Tail Unique Word DFT Extended OFDM (ZT-UW DTS-s OFDM), Unique Word OFDM (UW-OFDM), Resource Block Filtered OFDM, Filter Bank Multicarrier (FBMC), etc.

[0018] like Figure 1A As shown, the communication system 100 may include wireless transmit / receive units (WTRUs) 102a, 102b, 102c, 102d, RAN 104 / 113, CN 106 / 115, Public Switched Telephone Network (PSTN) 108, Internet 110, and other networks 112. However, it should be understood that the disclosed embodiments contemplate any number of WTRUs, base stations, networks, and / or network elements. Any of the WTRUs 102a, 102b, 102c, and 102d may be any type of device configured to operate and / or communicate in a wireless environment. For example, WTRUs 102a, 102b, 102c, and 102d (any of which may be referred to as a “station” and / or “STA”) may be configured to transmit and / or receive wireless signals and may include user equipment (UE), mobile stations, fixed or mobile subscriber units, subscription-based units, pagers, cellular phones, personal digital assistants (PDAs), smartphones, laptops, netbooks, personal computers, wireless sensors, hotspots or MiFi devices, Internet of Things (IoT) devices, watches or other wearable devices, head-mounted displays (HMDs), vehicles, drones, medical devices and applications (e.g., remote surgery), industrial devices and applications (e.g., robots and / or other wireless devices operating in industrial and / or automated processing chain environments), consumer electronics devices, devices operating on commercial and / or industrial wireless networks, etc. Any of WTRUs 102a, 102b, 102c, and 102d may be interchangeably referred to as WTRUs. Furthermore, any descriptions herein relating to UEs may be equally applied to WTRUs (and vice versa). For example, the WTRU can be configured to execute any procedure or process described herein as being performed by the UE (and vice versa).

[0019] The communication system 100 may also include base station 114a and / or base station 114b. Each of base stations 114a and 114b can be any type of device configured to wirelessly connect to at least one of WTRUs 102a, 102b, 102c, and 102d to facilitate access to one or more communication networks, such as CN 106 / 115, Internet 110, and / or other networks 112. As an example, base stations 114a and 114b may be base transceiver stations (BTS), NodeBs, eNodeBs, home NodeBs, home eNodeBs, gNBs, NR NodeBs, site controllers, access points (APs), wireless routers, etc. Although base stations 114a and 114b are each depicted as a single element, it will be understood that base stations 114a and 114b may include any number of interconnected base stations and / or network elements.

[0020] Base station 114a may be part of RAN 104 / 113, and may also include other base stations and / or network elements (not shown), such as base station controllers (BSCs), radio network controllers (RNCs), relay nodes, etc. Base station 114a and / or base station 114b may be configured to transmit and / or receive radio signals on one or more carrier frequencies, which may be referred to as cells (not shown). These frequencies may be in licensed spectrum, unlicensed spectrum, or a combination of licensed and unlicensed spectrum. A cell may provide coverage of a specific geographic area, which may be relatively fixed or may change over time. A cell may be further divided into cell sectors. For example, the cell associated with base station 114a may be divided into three sectors. Thus, in one embodiment, base station 114a may include three transceivers, i.e., one transceiver per sector of the cell. In one embodiment, base station 114a may employ multiple-input multiple-output (MIMO) technology and may use multiple transceivers for each sector of the cell. For example, beamforming may be used to transmit and / or receive signals in a desired spatial direction.

[0021] Base stations 114a and 114b can communicate with one or more of WTRUs 102a, 102b, 102c, and 102d via air interface 116, which can be any suitable wireless communication link (e.g., radio frequency (RF), microwave, centimeter wave, micrometer wave, infrared (IR), ultraviolet (UV), visible light, etc.). Air interface 116 can be established using any suitable radio access technology (RAT).

[0022] More specifically, as described above, the communication system 100 can be a multiple access system and can employ one or more channel access schemes, such as CDMA, TDMA, FDMA, OFDMA, SC-FDMA, etc. For example, base stations 114a and WTRUs 102a, 102b, and 102c in RAN 104 / 113 can implement radio technologies such as Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access (UTRA), which can use Wideband CDMA (WCDMA) to establish air interfaces 115 / 116 / 117. WCDMA can include communication protocols such as High-Speed ​​Packet Access (HSPA) and / or Evolved HSPA (HSPA+). HSPA can include High-Speed ​​Downlink (DL) Packet Access (HSDPA) and / or High-Speed ​​UL Packet Access (HSUPA).

[0023] In one embodiment, base station 114a and WTRUs 102a, 102b, 102c can implement radio technologies such as Evolved UMTS Terrestrial Radio Access (E-UTRA), which can use Long Term Evolution (LTE) and / or LTE-A Advanced (LTE-A) and / or LTE-A Pro Advanced (LTE-A Pro) to establish air interface 116.

[0024] In one embodiment, base station 114a and WTRUs 102a, 102b, 102c can implement radio technologies such as NR radio access, which can use New Radio (NR) to establish air interface 116.

[0025] In one embodiment, base station 114a and WTRUs 102a, 102b, and 102c can implement multiple radio access technologies. For example, base station 114a and WTRUs 102a, 102b, and 102c can, for instance, use a dual connectivity (DC) principle to implement both LTE and NR radio access. Therefore, the air interface used by WTRUs 102a, 102b, and 102c can be characterized by multiple types of radio access technologies and / or transmissions sent to / from multiple types of base stations (e.g., eNBs and gNBs).

[0026] In other embodiments, base station 114a and WTRUs 102a, 102b, 102c may implement wireless technologies such as IEEE 802.11 (i.e., Wi-Fi), IEEE 802.16 (i.e., Global Microwave Access Interoperability (WiMAX)), CDMA2000, CDMA 2000 1X, CDMA 2000 EV-DO, Provisional Standard 2000 (IS-2000), Provisional Standard 95 (IS-95), Provisional Standard 856 (IS-856), Global System for Mobile Communications (GSM), Enhanced Data Rate GSM Evolution (EDGE), GSMEDGE (GERAN), etc.

[0027] Figure 1A Base station 114b can be, for example, a wireless router, a home NodeB, a home eNodeB, or an access point, and can utilize any suitable RAT to facilitate wireless connectivity in a local area, such as a business premises, home, vehicle, campus, industrial facility, air corridor (e.g., for drone use), road, etc. In one embodiment, base station 114b and WTRUs 102c, 102d can implement radio technologies such as IEEE 802.11 to establish a wireless local area network (WLAN). In one embodiment, base station 114b and WTRUs 102c, 102d can implement radio technologies such as IEEE 802.15 to establish a wireless personal area network (WPAN). In yet another embodiment, base station 114b and WTRUs 102c, 102d can utilize cellular-based RATs (e.g., WCDMA, CDMA 2000, GSM, LTE-A, LTE-A Pro, NR, etc.) to establish picocells or femtocells. Figure 1A As shown, base station 114b can have a direct connection to Internet 110. Therefore, base station 114b does not need to access Internet 110 via CN 106 / 115.

[0028] RAN 104 / 113 can communicate with CN 106 / 115, which can be any type of network configured to provide voice, data, application, and / or Voice over Internet Protocol (VoIP) services to one or more of WTRUs 102a, 102b, 102c, and 102d. Data may have varying Quality of Service (QoS) requirements, such as different throughput requirements, latency requirements, fault tolerance requirements, reliability requirements, data throughput requirements, mobility requirements, etc. CN 106 / 115 can provide call control, billing services, location-based services, prepaid calling, internet connectivity, video distribution, etc., and / or perform advanced security functions (such as user authentication). Although in Figure 1AAlthough not shown, it should be understood that RAN 104 / 113 and / or CN 106 / 115 can communicate directly or indirectly with other RANs using the same RAT as or a different RAT than RAN 104 / 113. For example, in addition to being connected to RAN 104 / 113, which can utilize NR radio technology, CN 106 / 115 can also communicate with another RAN (not shown) using GSM, UMTS, CDMA 2000, WiMAX, E-UTRA, or WiFi radio technology.

[0029] CN 106 / 115 can also be used as a gateway for WTRU 102a, 102b, 102c, 102d to access PSTN 108, the Internet 110, and / or other networks 112. PSTN 108 may include a circuit-switched telephone network providing Common Old-Style Telephone Service (POTS). The Internet 110 may include a global system of interconnected computer networks and devices using common communication protocols, such as Transmission Control Protocol (TCP), User Datagram Protocol (UDP), and / or Internet Protocol (IP) from the TCP / IP Internet Protocol suite. Network 112 may include wired and / or wireless communication networks owned and / or operated by other service providers. For example, network 112 may include another CN connected to one or more RANs, which may use the same RAT as RAN 104 / 113 or a different RAT.

[0030] Some or all of the WTRUs 102a, 102b, 102c, and 102d in the communication system 100 may include multi-mode capability (e.g., WTRUs 102a, 102b, 102c, and 102d may include multiple transceivers to communicate with different wireless networks via different wireless links). For example, Figure 1A The WTRU 102c shown can be configured to communicate with base station 114a, which can use cellular-based radio technology, and with base station 114b, which can use IEEE 802 radio technology.

[0031] Figure 1B This is a system diagram illustrating example WTRU 102. (See diagram below.) Figure 1B As shown, WTRU 102 may include a processor 118, a transceiver 120, a transmit / receive element 122, a speaker / microphone 124, a keyboard 126, a display / touchpad 128, non-removable memory 130, removable memory 132, a power supply 134, a Global Positioning System (GPS) chipset 136, and / or other peripheral devices 138, etc. It is understood that WTRU 102 may include any sub-combination of the foregoing elements while remaining consistent with the embodiments.

[0032] Processor 118 may be a general-purpose processor, a special-purpose processor, a conventional processor, a digital signal processor (DSP), multiple microprocessors, one or more microprocessors associated with a DSP core, a controller, a microcontroller, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) circuit, any other type of integrated circuit (IC), a state machine, etc. Processor 118 may perform signal encoding, data processing, power control, input / output processing, and / or any other functions that enable WTRU 102 to operate in a wireless environment. Processor 118 may be coupled to transceiver 120, and transceiver 120 may be coupled to transmitting / receiving element 122. Although Figure 1B While the processor 118 and transceiver 120 are depicted as separate components, it should be understood that the processor 118 and transceiver 120 may be integrated together in an electronic package or chip.

[0033] Transmitting / receiving element 122 can be configured to transmit signals to or receive signals from a base station (e.g., base station 114a) via air interface 116. For example, in one embodiment, transmitting / receiving element 122 can be an antenna configured to transmit and / or receive RF signals. In one embodiment, transmitting / receiving element 122 can be a transmitter / detector configured to transmit and / or receive, for example, IR, UV, or visible light signals. In yet another embodiment, transmitting / receiving element 122 can be configured to transmit and / or receive both RF and optical signals. It should be understood that transmitting / receiving element 122 can be configured to transmit and / or receive any combination of wireless signals.

[0034] Although the transmitting / receiving element 122 is in Figure 1B While described as a single element, WTRU 102 may include any number of transmitting / receiving elements 122. More specifically, WTRU 102 may use MIMO technology. Thus, in one embodiment, WTRU 102 may include two or more transmitting / receiving elements 122 (e.g., multiple antennas) for transmitting and receiving wireless signals via air interface 116.

[0035] Transceiver 120 can be configured to modulate signals transmitted by transmitting / receiving element 122 and demodulate signals received by transmitting / receiving element 122. As described above, WTRU 102 can have multimode capability. Therefore, transceiver 120 can include multiple transceivers to enable WTRU 102 to communicate via various RATs, such as NR and IEEE 802.11.

[0036] The processor 118 of WTRU 102 may be coupled to a speaker / microphone 124, a keyboard 126, and / or a display / touchpad 128 (e.g., a liquid crystal display (LCD) unit or an organic light-emitting diode (OLED) display unit) and may receive user input data therefrom. The processor 118 may also output user data to the speaker / microphone 124, keyboard 126, and / or display / touchpad 128. Additionally, the processor 118 may access information from any type of suitable memory and store data in said memory, such as non-removable memory 130 and / or removable memory 132. 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. Removable memory 132 may include a user identification module (SIM) card, memory stick, secure digital storage (SD) card, etc. In other embodiments, the processor 118 may access information from memory that is not physically located on WTRU 102 (e.g., located on a server or home computer (not shown)) and store data in said memory.

[0037] The processor 118 can receive power from the power supply 134 and can be configured to distribute and / or control power to other components in the WTRU 102. The power supply 134 can be any suitable device for powering the WTRU 102. For example, the power supply 134 may include one or more dry cell batteries (e.g., nickel-cadmium (NiCd), nickel-zinc (NiZn), nickel metal hydride (NiMH), lithium-ion (Li-ion), etc.), solar cells, fuel cells, etc.

[0038] 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) about the current location of the WTRU 102. In addition to, or alternatively to, the information from the GPS chipset 136, the WTRU 102 may receive location information from base stations (e.g., base stations 114a, 114b) via the air interface 116, and / or determine its location based on the timing of signals received from two or more neighboring base stations. It should be understood that the WTRU 102 may acquire location information using any suitable location determination method, while remaining consistent with the embodiments.

[0039] The processor 118 may also be coupled to other peripheral devices 138, which may include one or more software and / or hardware modules providing additional features, functions, and / or wired or wireless connectivity. For example, peripheral devices 138 may include accelerometers, electronic compasses, satellite transceivers, digital cameras (for photos and / or video), Universal Serial Bus (USB) ports, vibration devices, television transceivers, hands-free headsets, Bluetooth® modules, FM radio units, digital music players, media players, video game player modules, internet browsers, virtual reality and / or augmented reality (VR / AR) devices, activity trackers, etc. Peripheral devices 138 may include one or more sensors, which may be one or more of the following: gyroscopes, accelerometers, Hall effect sensors, magnetometers, orientation sensors, proximity sensors, temperature sensors, time sensors; geolocation sensors; altimeters, light sensors, touch sensors, magnetometers, barometers, gesture sensors, biometric sensors, and / or humidity sensors.

[0040] WTRU 102 may include a full-duplex radio, for which the transmission and reception of some or all of the signals (e.g., signals associated with specific subframes for uplink (e.g., for transmission) and downlink (e.g., for reception)) may be concurrent and / or simultaneous. The full-duplex radio may include an interference management unit 139 to reduce and / or substantially eliminate self-interference through hardware (e.g., a choke) or through signal processing by a processor (e.g., a separate processor (not shown) or processor 118). In one embodiment, WTRU 102 may include a half-duplex radio, for which the transmission and reception of some or all of the signals (e.g., signals associated with specific subframes for uplink (e.g., for transmission) or downlink (e.g., for reception)) are separate.

[0041] Figure 1C This is a system diagram illustrating RAN 104 and CN 106 according to an embodiment. As described above, RAN 104 may employ E-UTRA radio technology to communicate with WTRUs 102a, 102b, and 102c via air interface 116. RAN 104 may also communicate with CN 106.

[0042] RAN 104 may include eNode-Bs 160a, 160b, and 160c, but it should be understood that RAN 104 may include any number of eNode-Bs while remaining consistent with the embodiments. eNode-Bs 160a, 160b, and 160c may each include one or more transceivers to communicate with WTRUs 102a, 102b, and 102c via air interface 116. In one embodiment, eNode-Bs 160a, 160b, and 160c may implement MIMO technology. Therefore, for example, eNode-B 160a may use multiple antennas to transmit and / or receive radio signals from WTRU 102a.

[0043] Each of the eNode-B 160a, 160b, and 160c can be associated with a specific cell (not shown) and can be configured to handle radio resource management decisions, handover decisions, user scheduling in UL and / or DL, etc. Figure 1C As shown, eNode-B 160a, 160b, and 160C can communicate with each other via the X2 interface.

[0044] Figure 1C The CN 106 shown may include a Mobility Management Entity (MME) 162, a Serving Gateway (SGW) 164, and a Packet Data Network (PDN) Gateway (or PGW) 166. While each of the foregoing elements is depicted as part of the CN 106, it should be understood that any of these elements may be owned and / or operated by an entity other than the CN operator.

[0045] The MME 162 can connect to each of the eNodes B162a, 162b, and 162c in RAN 104 via the S1 interface and can be used as a control node. For example, the MME 162 can be responsible for authenticating users of WTRUs 102a, 102b, and 102c, bearer activation / deactivation, selecting a specific serving gateway during the initial attachment of WTRUs 102a, 102b, and 102c, etc. The MME 162 can provide control plane functions for handover between RAN 104 and other RANs (not shown) employing other radio technologies (such as GSM and / or WCDMA).

[0046] The SGW 164 can connect to each of the eNode-Bs 160a, 160b, and 160c in RAN 104 via the S1 interface. The SGW 164 can typically route and forward user data packets to / from WTRUs 102a, 102b, and 102c. The SGW 164 can perform other functions, such as anchoring the user plane during handover between eNode Bs, triggering paging when DL data is available to WTRUs 102a, 102B, and 102c, managing and storing the context of WTRUs 102a, 102B, and 102c, etc.

[0047] The SGW 164 can connect to the PGW 166, which can provide WTRU 102a, 102b, and 102c with access to packet-switched networks (such as Internet 110) to facilitate communication between WTRU 102a, 102b, 102c and IP-enabled devices.

[0048] CN 106 can facilitate communication with other networks. For example, CN 106 can provide WTRU 102a, 102b, and 102c with access to a circuit-switched network (e.g., PSTN 108) to facilitate communication between WTRU 102a, 102b, and 102c and traditional landline communication equipment. For example, CN 106 may include an IP gateway (e.g., an IP Multimedia Subsystem (IMS) server) or can communicate with an IP gateway that serves as an interface between CN 106 and PSTN 108. Furthermore, CN 106 can provide WTRU 102a, 102b, and 102c with access to other networks 112, which may include other wired and / or wireless networks owned and / or operated by other service providers.

[0049] Although WTRU is Figures 1A to 1D While described as a wireless terminal, it is conceivable that in some representative embodiments, such a terminal may use (e.g., temporarily or permanently) a wired communication interface with a communication network.

[0050] In a representative embodiment, the other network 112 may be a WLAN.

[0051] A WLAN in Infrastructure Basic Services Set (BSS) mode can have an Access Point (AP) for the BSS and one or more Stations (STAs) associated with the AP. The AP can access or interface with a Distribution System (DS) or another type of wired / wireless network that carries traffic entering and / or leaving the BSS. Traffic originating outside the BSS destined for a STA can reach the STA via the AP and can be delivered to the STA. Traffic originating from a STA destined for a destination outside the BSS can be sent to the AP for delivery to the appropriate destination. Traffic between STAs within the BSS can be sent via the AP, for example, where a source STA can send traffic to the AP, and the AP can deliver the traffic to the destination STA. Traffic between STAs within the BSS can be considered and / or referred to as peer-to-peer traffic. This peer-to-peer traffic can be sent between the source and destination STAs (e.g., directly between the source and destination STAs) using Direct Link Establishment (DLS). In some representative embodiments, the DLS can use 802.11e DLS or 802.11z Tunneled DLS (TDLS). WLANs using the Standalone BSS (IBSS) mode cannot have access points (APs), and STAs within the IBSS or using the IBSS (e.g., all STAs) can communicate directly with each other. The IBSS communication mode may sometimes be referred to as the "ad-hoc" communication mode in this document.

[0052] When using 802.11ac infrastructure operating mode or a similar operating mode, the AP can transmit beacons on a fixed channel, such as the primary channel. The primary channel can be of fixed width (e.g., a bandwidth of 20 MHz) or dynamically set via signaling. The primary channel can be the operating channel of the BSS and can be used by the STA to establish a connection with the AP. In some representative embodiments, such as in an 802.11 system, Carrier Sense Multiple Access (CSMA / CA) with collision avoidance can be implemented. For CSMA / CA, STAs including the AP (e.g., each STA) can listen on the primary channel. If the primary channel is listened to / detected and / or determined to be busy by a particular STA, that particular STA can back off. A single STA (e.g., only one station) can transmit at any given time within a given BSS.

[0053] High-throughput (HT) STAs can communicate using a 40 MHz wide channel, for example, by combining a primary 20 MHz channel with adjacent or non-adjacent 20 MHz channels to form a 40 MHz wide channel.

[0054] Very High Throughput (VHT) STAs can support channels with widths of 20 MHz, 40 MHz, 80 MHz, and / or 160 MHz. 40 MHz and / or 80 MHz channels can be formed by combining consecutive 20 MHz channels. A 160 MHz channel can be formed by combining eight consecutive 20 MHz channels, or by combining two non-consecutive 80 MHz channels; this can be referred to as an 80+80 configuration. For the 80+80 configuration, after channel coding, the data can pass through a segmented parser that divides the data into two streams. Each stream can be processed separately using Inverse Fast Fourier Transform (IFFT) and time-domain processing. The streams can be mapped onto the two 80 MHz channels, and the data can be transmitted by the transmitting STA. At the receiver of the receiving STA, the operation of the 80+80 configuration can be reversed, and the combined data can be sent to the Media Access Control (MAC).

[0055] Operating modes below 1 GHz are supported by 802.11af and 802.11ah. The channel operating bandwidth and carrier in 802.11af and 802.11ah are reduced compared to those used in 802.11n and 802.11ac. 802.11af supports 5 MHz, 10 MHz, and 20 MHz bandwidths in the TV whitespace (TVWS) spectrum, while 802.11ah supports 1 MHz, 2 MHz, 4 MHz, 8 MHz, and 16 MHz bandwidths using non-TVWS spectrum. According to a representative embodiment, 802.11ah may support instrument-type control / machine-type communications, such as MTC devices in macro coverage areas. MTC devices may have certain capabilities, such as limited capabilities including support for certain and / or limited bandwidths (e.g., only support). MTC devices may include batteries with a battery life exceeding a threshold (e.g., to maintain a very long battery life).

[0056] WLAN systems that can support multiple channels and channel bandwidths (e.g., 802.11n, 802.11ac, 802.11af, and 802.11ah) include a channel that can be designated as the primary channel. The primary channel can have a bandwidth equal to the maximum common operating bandwidth supported by all STAs in the BSS. The bandwidth of the primary channel can be set and / or limited by the STA that supports the minimum bandwidth operating mode among all STAs operating in the BSS. In the 802.11ah example, for STAs that support (e.g., only support) the 1MHz mode (e.g., MTC type devices), the primary channel can be 1 MHz wide 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 Sense and / or Network Assignment Vector (NAV) settings can depend on the status of the primary channel. If the primary channel is busy, for example, because an STA (which only supports the 1 MHz operating mode) is transmitting to the AP, the entire available band can be considered busy even if most of the band remains idle and available.

[0057] In the United States, the available frequency bands for 802.11ah are from 902 MHz to 928 MHz. In South 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 802.11ah is 6 MHz to 26 MHz, depending on the country code.

[0058] Figure 1D This is a system diagram illustrating RAN 113 and CN 115 according to an embodiment. As described above, RAN 113 can communicate with WTRUs 102a, 102b, and 102c via air interface 116 using NR wireless technology. RAN 113 can also communicate with CN 115.

[0059] RAN 113 may include gNBs 180a, 180b, and 180c; however, it should be understood that RAN 113 may include any number of gNBs while remaining consistent with the embodiments. Each of gNBs 180a, 180b, and 180c includes one or more transceivers for communicating with WTRUs 102a, 102b, and 102c via air interface 116. In one embodiment, gNBs 180a, 180b, and 180c may implement MIMO technology. For example, gNBs 180a and 180b may utilize beamforming to transmit signals to and / or receive signals from gNBs 180a, 180b, and 180c. Therefore, gNB 180a may, for example, use multiple antennas to transmit radio signals to and / or receive radio signals from WTRU 102a. In one embodiment, gNBs 180a, 180b, and 180c may implement carrier aggregation technology. For example, gNB 180a can transmit multiple component carriers to 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 one embodiment, gNBs 180a, 180b, and 180c can implement Cooperative Multipoint (CoMP) technology. For example, WTRU 102a can receive coordinated transmissions from gNBs 180a and 180b (and / or gNB 180c).

[0060] WTRUs 102a, 102b, and 102c can communicate with gNBs 180a, 180b, and 180c using transmissions associated with a scalable set of parameters. For example, the OFDM symbol spacing and / or OFDM subcarrier spacing can vary for different transmissions, different cells, and / or different portions of the radio transmission spectrum. WTRUs 102a, 102b, and 102c can communicate with gNBs 180a, 180b, and 180c using subframes or transmission time intervals (TTIs) of varying lengths or scalable lengths (e.g., containing different numbers of OFDM symbols and / or continuously varying absolute times).

[0061] gNBs 180a, 180b, and 180c can be configured to communicate with WTRUs 102a, 102b, and 102c in standalone and / or non-standalone configurations. In standalone configuration, WTRUs 102a, 102b, and 102c can communicate with gNBs 180a, 180b, and 180c without needing to access other RANs (e.g., eNode-Bs 160a, 160b, and 160c). In standalone configuration, WTRUs 102a, 102b, and 102c can utilize one or more of gNBs 180a, 180b, and 180c as mobility anchors. In standalone configuration, WTRUs 102a, 102b, and 102c can communicate with gNBs 180a, 180b, and 180c using signals in unlicensed frequency bands. In a non-standalone configuration, WTRUs 102a, 102b, and 102c can communicate / connect with gNBs 180a, 180b, and 180c, and also with another RAN (such as eNode-Bs 160a, 160b, and 160c). For example, WTRUs 102a, 102b, and 102c can implement DC principles to communicate essentially simultaneously with one or more gNBs 180a, 180b, and 180c, as well as one or more eNode-Bs 160a, 160b, and 160c. In a non-standalone configuration, eNode-Bs 160a, 160b, and 160c can serve as mobility anchors for WTRUs 102a, 102b, and 102c, while gNBs 180a, 180b, and 180c can provide additional coverage and / or throughput for serving WTRUs 102a, 102b, and 102c.

[0062] Each of gNBs 180a, 180b, and 180c can be associated with a specific cell (not shown) and can be configured to handle radio resource management decisions, handover decisions, user scheduling in UL and / or DL, network fragmentation support, dual connectivity, interoperability between NR and E-UTRA, routing of user plane data to User Plane Functions (UPF) 184a and 184b, routing of control plane information to Access and Mobility Management Functions (AMF) 182a and 182b, etc. Figure 1D As shown, gNB 180a, 180b, and 180c can communicate with each other via the Xn interface.

[0063] Figure 1DThe CN 115 shown may include at least one AMF 182a, 182b, at least one UPF 184a, 184b, at least one Session Management Function (SMF) 183a, 183b, and may include data networks (DN) 185a, 185b. While each of the foregoing elements is depicted as part of the CN 115, it should be understood that any of these elements may be owned and / or operated by an entity other than the CN operator.

[0064] AMF 182a and 182b can connect to one or more of gNBs 180a, 180b, and 180c in RAN 113 via the N2 interface and can act as control nodes. For example, AMF 182a and 182b can be responsible for authenticating users of WTRU 102a, 102b, and 102c, supporting network slicing (e.g., handling different PDU sessions with different needs), selecting specific SMF 183a and 183b, managing registration areas, terminating NAS signaling, mobility management, and so on. AMF 182a and 182b can use network slicing to customize CN support for WTRU 102a, 102b, and 102c based on the type of service being used by WTRU 102a, 102b, and 102c. For example, different network slices can be established for different use cases, such as services relying on Ultra Reliable Low Latency Communication (URLLC) access, services relying on Enhanced Massive Mobile Broadband (eMBB) access, and services for Machine Type Communication (MTC) access. AMF 162 can provide control plane functions for handover between RAN 113 and other RANs (not shown) employing other radio technologies (such as LTE, LTE-A, LTE-A Pro and / or non-3GPP access technologies (such as WiFi)).

[0065] SMFs 183a and 183b can connect to AMFs 182a and 182b in CN 115 via the N11 interface. SMFs 183a and 183b can also connect to UPFs 184a and 184b in CN 115 via the N4 interface. SMFs 183a and 183b can select and control UPFs 184a and 184b, and configure them to route traffic through UPFs 184a and 184b. SMFs 183a and 183b can perform other functions, such as managing and allocating WTRU IP addresses, managing PDU sessions, controlling policy enforcement and QoS, and providing downlink data notifications. PDU session types can be IP-based, non-IP-based, Ethernet-based, etc.

[0066] UPF 184a and 184b can connect to one or more of gNB 180a, 180b, and 180c in RAN 113 via the N3 interface. This provides WTRU 102a, 102b, and 102c with access to packet-switched networks (such as Internet 110) to facilitate communication between WTRU 102a, 102b, and 102c and IP-enabled devices. UPF 184 and 184b can perform other functions such as routing and forwarding packets, enforcing user plane policies, supporting multi-homed PDU sessions, handling user plane QoS, buffering downlink packets, and providing mobility anchoring.

[0067] CN 115 can facilitate communication with other networks. For example, CN 115 may include, or communicate with, an IP gateway (e.g., an IP Multimedia Subsystem (IMS) server) serving as an interface between CN 115 and PSTN 108. Furthermore, CN 115 may provide WTRUs 102a, 102b, and 102c with access to other networks 112, which may include other wired and / or wireless networks owned and / or operated by other service providers. In one embodiment, WTRUs 102a, 102b, and 102c can be connected to DNs 185a and 185b via UPFs 184a and 184b through their N3 interfaces and the N6 interface between UPFs 184a and 184b and local data networks (DNs) 185a and 185b.

[0068] Given Figures 1A to 1D and Figures 1A to 1D As described herein, one or more of the functions described with respect to one or more of the following can be performed by one or more emulation devices (not shown): WTRU 102a-d, Base Station 114a-b, eNode-B 160a-c, MME 162, SGW 164, PGW 166, gNB 180a-c, AMF 182a-b, UPF184a-b, SMF 183a-b, DN 185a-b, and / or any other device described herein. An emulation device can be one or more devices configured to emulate one or more of the functions described herein. For example, an emulation device can be used to test other devices and / or simulate network and / or WTRU functions.

[0069] Simulation devices can be designed to perform one or more tests on other devices in laboratory and / or carrier network environments. For example, one or more simulation devices may perform one or more functions while being fully or partially implemented and / or deployed as part of a wired and / or wireless communication network to test other devices within the communication network. The one or more simulation devices may perform one or more or all functions while being temporarily implemented / deployed as part of a wired and / or wireless communication network. Simulation devices may be directly coupled to another device for testing purposes and / or may use over-the-air wireless communication to perform tests.

[0070] One or more simulation devices can perform one or more functions (including all functions) without being implemented / deployed as part of a wired and / or wireless communication network. For example, simulation devices can be used in test scenarios within a test laboratory and / or an undeployed (e.g., tested) wired and / or wireless communication network to perform testing of one or more components. One or more simulation devices can be test devices. Simulation devices can transmit and / or receive data using direct RF coupling and / or wireless communication via RF circuitry (e.g., which may include one or more antennas).

[0071] New radio (NR) multiple-input multiple-output (MIMO) can support eType II codebooks for predicting precoded matrix indicators (PMIs), which can, for example, be derived using conventional (e.g., non-AI / ML) methods. However, this can be improved, for example, by supporting model monitoring, AI / ML methods for CSI prediction compression.

[0072] For scenarios where functionality is implemented in two separate and / or sequentially executed ML models (e.g., AI / ML CSI prediction plus AI / MLCSI compression), identifying the sources of performance degradation and / or major contributing factors can be useful for achieving and / or maintaining the desired system performance. CSI compression using both models can be challenging.

[0073] The systems and methods described in this paper can be used for WTRU-side CSI prediction and / or compression with separate prediction and compression models. The systems and methods described in this paper can describe how to identify sources of degradation (e.g., prediction, compression, or combination), and / or how to determine the configuration of CSI prediction and CSI compression to meet target CSI reconstruction performance under given payload constraints.

[0074] AI / ML CSI compression and prediction can improve the performance of AI / ML-based CSI feedback. The process may include using multiple historical CSI instances to predict one or more future CSI instances. The process may additionally or alternatively include compressing the one or more predicted future CSI instances.

[0075] CSI prediction can use a one-sided AI / ML prediction model located at the WTRU (e.g., referred to as a WTRU-side model). The AI / ML prediction model can be based on a recurrent neural network (RNN) and / or long short-term memory (LSTM) model architecture, for example, to predict one or more future CSI instances by utilizing historical CSI instances within an observation window (e.g., channel estimation based on (e.g., an indication of) the received CSI reference signal (RS). AI / ML-based CSI compression can use a two-sided autoencoder (AE) model, where the encoder portion (e.g., a WTRU-side encoder) is located on the WTRU side and performs compression. For example, a decoder portion located on the network side (e.g., an NW / NE-side decoder) can perform reconstruction based on received compressed CSIs (e.g., an indication of said received compressed CSIs).

[0076] Monitoring the performance of AI / ML models can help ensure consistent system performance across different deployments, channel conditions, local conditions, and / or more scenarios. For example, when trained in one set of channel conditions (e.g., outdoors) and deployed in different channel conditions (e.g., indoors), an ML model may not always generalize well. Model performance monitoring can be useful in the lifecycle management (LCM) of AI / ML models. This paper presents a system and method for monitoring model performance in AI / ML CSI prediction compression.

[0077] Figure 2This is a system diagram illustrating Example 200 of WTRU 202 and Network (NW) or Network Entity (NE) 204. WTRU 202 can perform CSI prediction and / or compression. For example, WTRU can perform combined CSI prediction and compression. WTRU 202 can perform prediction at prediction block 206. The input to prediction block 206 can include the current CSI and / or historical (e.g., previous) CSI from, for example, buffer 210. WTRU 202 can perform compression at encoder 208. WTRU 202 can determine and / or report one or more CSI error condition types. WTRU 202 can determine one or more CSI error condition types based on channel measurements and / or (e.g., configured) performance monitoring thresholds. WTRU 202 can report one or more CSI error condition types based on channel measurements and / or (e.g., configured) performance monitoring thresholds. WTRU 202 can determine (e.g., detect) operation under CSI prediction-constrained conditions. WTRU 202 can determine and / or report, for example, parameters of WTRU-side CSI prediction (e.g., preferred parameters). These parameters can be determined and / or reported based on channel measurements and / or configured performance monitoring thresholds. WTRU 202 can determine (e.g., detect) operations under combined CSI prediction and compression error condition types. WTRU 202 can determine and / or report, based on the index of prediction instances in the measurement and / or prediction window, the WTRU-side CSI compression configuration for each CSI instance in the prediction window (e.g., preferred configuration).

[0078] WTRU 202 can perform combined CSI prediction and compression. WTRU 202 can determine the CSI error condition type based on channel measurements and / or configured performance monitoring thresholds. WTRU can report the CSI error condition type to network 204 based on channel measurements and / or configured performance monitoring thresholds. Additionally or alternatively, WTRU 202 can report predicted and compressed CSI to network 204. WTRU 202 may be able to perform combined CSI prediction and compression, and / or may receive configuration information (e.g., indications of the configuration information) for combined CSI prediction and compression reporting. The configuration information may include one or more of CSI-RS configuration, CSI feedback reporting configuration, CSI prediction configuration (e.g., observation window and / or prediction window), CSI compression configuration, and / or performance monitoring thresholds. Network 204 may include, for example, a decoder 212 for reconstructing CSI.

[0079] Figure 3This is an example of CSI procedure 300. At 302, the WTRU may receive configuration information (e.g., an indication of the configuration information). The configuration information may be used to combine CSI prediction and compression. At 304, the WTRU may be triggered to (e.g., determine to) perform one or more measurements, such as those of a WTRU-side model, for monitoring the performance of combined CSI prediction and compression. Additionally or alternatively, at 304, the WTRU may perform one or more measurements. For example, the WTRU may determine the relative prediction and / or compression error relative to the total combined CSI error. Measurements may include one or more of the following: channel measurements and / or statistics, prediction estimation error metrics, compression reconstruction error metrics, total combined CSI error, relative prediction error of one or more CSI instances in the prediction window, and / or relative compression error relative to the total combined CSI error. Additionally or alternatively, one or more channel measurements and / or statistics may include one or more of the following: time-domain channel properties (TDCP), reference signal received power (RSRP), and / or signal-to-interference and noise ratio (SINR) associated with CSI-RS (e.g., in the observation window), Doppler, channel coherence time, first-order / second-order statistics, and / or distribution measures for out-of-distribution detection.

[0080] Prediction error may include, for example, the difference between the actual CSI and the predicted CSI of a CSI prediction model (e.g., squared generalized cosine similarity (SGCS) and / or normalized mean square error (NMSE), etc.). Compression error may include, for example, a measurement of the difference between the input CSI and the reconstructed CSI of an autoencoder-based CSI compression model (e.g., SGCS, NMSE, and / or similar terms). Total combined CSI error may include, for example, a measurement of the difference between the actual CSI and the reconstructed predicted CSI of a combined CSI prediction / compression model (e.g., SGCS, NMSE, and / or similar terms).

[0081] At point 306, the WTRU can determine the CSI error condition type. The WTRU can determine the CSI error condition type based on (for example, based on) one or more of the following: one or more measurements used to combine CSI prediction and / or compression performance monitoring, and / or the current combined CSI prediction and / or compression configuration. The CSI error condition type can be determined as one or more of CSI prediction-limited condition, CSI compression-limited condition, and / or a combination of CSI prediction and compression-limited conditions.

[0082] CSI prediction-limited conditions may include (for example, being determined to be) the following: the relative prediction error equals the relative compression error or exceeds a configured threshold for the relative compression error. CSI compression-limited conditions may include (for example, being determined to be) the following: the relative compression error equals the relative prediction error or exceeds a configured threshold for the relative prediction error. Combined CSI prediction and compression-limited conditions may include (for example, being determined to be) the following: the difference between the relative prediction error and the relative compression error is equal to or less than a configured threshold. The WTRU may report predicted and / or compressed CSI, and / or the determined CSI error condition type. At 308, the WTRU may report a message to, for example, the network. This message may include one or more of the following: predicted CSI, compressed CSI, predicted and compressed CSI, and / or the determined CSI error condition type.

[0083] Identifying sources of degradation (e.g., prediction, compression, or combination) can enable the adaptation of parameters to the current prediction and / or compression blocks, for example, by reducing overhead by decreasing the need for one or more of the following: model fine-tuning, model switching, and / or model transfer / delivery. CSI prediction-constrained conditions can include (e.g., identified as) a situation where the WTRU operates under conditions where CSI prediction error dominates the total error of the combined CSI prediction and compression blocks. For example, the CSI prediction error may exceed a threshold of the total error. The terms "CSI prediction-constrained," "prediction-constrained," and / or "CSI prediction-constrained condition" and "prediction-constrained condition" are used interchangeably herein. CSI compression-constrained conditions can include (e.g., identified as) a situation where the WTRU operates under conditions where CSI compression reconstruction error dominates the error of the combined CSI prediction and compression blocks. For example, the CSI compression reconstruction error may exceed a threshold of the total error. The terms "CSI compression-constrained," "compression-constrained," "CSI compression-constrained condition," and / or "compression-constrained condition" are used interchangeably herein.

[0084] The combined CSI prediction and compression error condition can indicate (e.g., be determined) that the individual contribution of CSI prediction to the total error of combined CSI prediction and compression is comparable to the individual contribution of CSI compression to the total error of combined CSI prediction and compression. The CSI error condition type can indicate (e.g., be determined) whether the WTRU operates under a CSI prediction-limited condition, a CSI compression-limited condition, and / or a combined CSI prediction and compression error condition. The term "combined CSI prediction and compression" can include (e.g., be determined) the following situation: when CSI is processed sequentially using at least two separate blocks / modules. The first block can perform CSI prediction. The second block can perform compression of the predicted CSI. In this document, "combined CSI prediction and compression" is used interchangeably with "CSI prediction plus compression". Compression ratio operations can include, for example, autoencoder-based data compression operations. The compression ratio can be determined based on the ratio between the input (e.g., uncompressed) size and the output (e.g., compressed) size. A high compression ratio can be associated with more compression of the input data (e.g., meaning more compression of the input data). A low compression ratio can be associated with less compression of the input data (e.g., meaning less compression of the input data). Additionally or alternatively, such as with regard to channel response, the terms “raw channel,” “full channel,” and / or “full channel response matrix” are used interchangeably herein.

[0085] There may be WTRU configurations and / or measurements for combined CSI prediction and compression model monitoring. The WTRU may receive configuration information for measurement (e.g., an indication of that configuration information) and / or determine the configuration information for measurement. Additionally or alternatively, the WTRU may receive configuration information for reporting (e.g., an indication of that configuration information) and / or determine the configuration information for reporting. For example, the WTRU may perform CSI-RS measurements based on configuration (e.g., information). The WTRU may report based on configuration (e.g., information). The WTRU may be configured to perform combined CSI prediction and compression. The WTRU may receive CSI-RS configuration (e.g., information) (e.g., an indication of that CSI-RS configuration (e.g., information)) for example, to perform CSI measurements. CSI-RS configuration information may be periodic, semi-persistent, and / or aperiodic. The WTRU may receive configuration information for CSI reporting (e.g., an indication of that configuration information), which may, for example, be periodic, semi-persistent, and / or aperiodic. In some examples, the WTRU may receive configuration information for performing measurements (e.g., CSI measurements) and configuration information for reporting these measurements (e.g., separate configuration information). In other examples, the WTRU may receive configuration information for configuring both CSI measurements and reporting for the WTRU.

[0086] The WTRU can receive and / or determine the configuration of the AI / ML CSI prediction model (e.g., an indication of that configuration). The configuration of the AI / ML CSI prediction model may include (e.g., for) one or more of the following: a prediction window configuration, an observation window configuration, metrics for performance measurement and / or error detection, a threshold for error detection, a statistical measurement for out-of-distribution (OOD) detection, a threshold for OOD event detection, and / or CSI prediction model information.

[0087] The configuration of the prediction window can include the number N of predicted CSI instances in the prediction window and / or the period (e.g., time) d between consecutive predicted CSI instances. The configuration of the observation window can include the number K of historical CSI samples in the observation window and / or the period (e.g., time) m between consecutive CSI instances in the observation window. The number K of historical CSI samples in the observation window can be equal to or less than the configured ML prediction model capability.

[0088] Metrics used for performance measurement and / or error detection may include one or more of NMSE, mean squared error (MSE), SGCS, and / or similar terms. Thresholds used for error detection may include one or more of a time-domain channel attribute (TDCP) threshold and / or a prediction error threshold (e.g., corresponding to the configured performance metrics such as SCGS, NMSE, and / or similar terms). The WTRU may use metrics for statistical measurement of OOD detection, such as evaluating the distribution of the input data and / or comparing it with the data distribution in the training dataset. For example, when the WTRU measures a (e.g., significant) difference between the actual data distribution (e.g., at inference time) and the data distribution in the training dataset, the WTRU may determine that an OOD event has occurred. Thresholds used for OOD event detection may include energy thresholds. For example, when a model is transferred / delivered to the WTRU, CSI prediction model information may include model capability information. Model capability information may include one or more of the (e.g., maximum) number of instances in the prediction window and / or the (e.g., maximum) buffer size of the observation window. This configuration may additionally or alternatively include information about the dataset used for model training, such as one or more of the following: minimum WTRU velocity, maximum WTRU velocity, and / or statistical characteristics of the channel (e.g., the channel's coherent bandwidth (BW) and / or line-of-sight (LOS) / non-line-of-sight (NLOS) characteristics). Additionally or alternatively, this configuration may include the domain of historical CSI samples at the input of the predictive model (e.g., the original channel and / or feature vectors). Velocities (e.g., maximum WTRU velocity and / or minimum WTRU velocity) in this document may include absolute or relative velocities. For example, a velocity may be relative to another device and / or entity (e.g., WTRU or network / network entity).

[0089] The WTRU can receive and / or determine the configuration of an AI / ML CSI compression model (e.g., an indication of such configuration). The configuration of the AI / ML CSI compression model can include one or more of the following: compression type, compression configuration, CSI compression encoder model information, and / or surrogate decoder information. The compression type can include an indication of whether predicted CSI instances should be compressed individually or jointly (e.g., compressing more than one CSI prediction instance). In the case of compressing CSI prediction instances individually, the configuration can include a lookup table. The lookup table can include one or more rows with a prediction error threshold and / or a corresponding compression ratio. The compression configuration can include, for example, the maximum number of CSI instances to be compressed in the case of jointly compressing multiple CSI instances. Although lookup tables are described herein, it is contemplated that information can be provided to the WTRU in formats other than lookup tables (e.g., lists).

[0090] For example, when transmitting / delivering an encoder model to a WTRU, CSI compressed encoder model information may include one or more of the following: model backbone information, domain information (e.g., whether compression is performed in the raw channel or feature vector domain), preprocessing information, quantization parameters, and / or supported compression rates. This configuration may additionally or alternatively include information about the dataset used for model training, such as the statistical characteristics of the channel (e.g., the channel's coherent BW and / or LOS / NLOS characteristics). Proxy decoder model information may include an indication of the proxy decoder model. The WTRU can use the proxy decoder model to measure CSI reconstruction accuracy. For example, if a proxy decoder model is transmitted / delivered to a WTRU and / or received by a WTRU, the proxy decoder model information may include the backbone type and / or information about the training dataset.

[0091] The WTRU can receive configuration (e.g., an indication of such configuration) for CSI error event detection and / or CSI error event type determination and / or determine the configuration for CSI error event detection and / or CSI error event type determination. The configuration for detecting CSI error events and / or determining CSI error types may include one or more of the following: a threshold for comparing relative compression error and relative prediction error; a threshold for the measured prediction error relative to total (e.g., combined) CSI error (e.g., a metric corresponding to the configuration); a threshold for the prediction estimation error for each predicted CSI instance; a threshold for the absolute difference between relative compression error and relative prediction error; and / or a threshold for CSI reconstruction error.

[0092] The threshold for comparing relative compression error and relative prediction error can use a configured metric (e.g., a metric based on that configuration). This threshold can be used to determine whether the WTRU is operating under CSI prediction-limited conditions and / or CSI compression-limited conditions. The WTRU can use a threshold (e.g., corresponding to a configured metric) for the measured prediction error relative to the total (e.g., combined) CSI error to determine whether it is operating under CSI prediction-limited conditions. The WTRU can use a threshold for the measured compression error relative to the total (e.g., combined) CSI error to determine whether it is operating under CSI compression-limited conditions. The WTRU can use a threshold of the absolute difference between the relative compression error and the relative prediction error, for example, to determine whether the WTRU is operating under combined CSI prediction and / or compression error conditions.

[0093] The WTRU can perform measurements for combined CSI prediction and compressed model monitoring. Measurements may include the WTRU's measurement of CSI prediction error. The WTRU can perform measurements of the error from CSI prediction AI / ML blocks. Measurements may be based on, for example, the input and / or output statistical distributions at the prediction block. Statistical calculations / measures may include one or more of the cumulative distribution function (CDF) and / or probability density function (PDF) of the input and / or output samples, first-order statistics, second-order statistics, and / or statistical distances. Statistical calculations / measures may include the statistical distribution of data at the input and / or output of the AI / ML CSI prediction block, for example, by comparing it to the distribution of training data at the input and / or output of the CSI prediction. For example, if a significant difference is detected by comparing the mean and / or standard deviation of the actual data with the mean and / or standard deviation of the training data (e.g., at the AI / ML CSI prediction input), an error indication (e.g., a possible prediction error) may exist. First-order statistics may include one or more of the mean, standard deviation, and / or variance. Second-order statistics may include one or more of the following: autocorrelation of one or more (e.g., each) historical CSI samples (e.g., inputs) and / or the output of the CSI-predicted AI / ML model, and / or cross-correlation between two or more CSI samples in the observation window. Statistical distance may include one or more of the following: the distance between the input / output distribution and the training dataset distribution (e.g., squared distance, Mahalanobis distance, Bhattacharyya coefficient, Kullback-Leibler (KL) divergence, Z-score, and / or energy-based score).

[0094] The WTRU can additionally or alternatively perform measurements for OOD detection on samples, for example, input to a CSI prediction AI / ML model. The WTRU can use distribution distances and / or computed statistics to determine whether the input samples are in-distribution or out-of-distribution relative to the training dataset of the CSI prediction AI / ML model. The WTRU can generate and / or send OOD reports. OOD reports may include measured statistical measures and / or OOD determinations. The WTRU may include (e.g., configured with) (e.g., a separate) AI / ML model, for example, for detecting anomalous and / or OOD input samples in the data.

[0095] The WTRU can collect a set of measurements (e.g., measurements related to the channel and / or one or more other applicable conditions). This set of measurements enables the WTRU to identify whether one or more other applicable conditions are consistent with conditions assumed (e.g., determined) during model training (e.g., a model trained for a specific site / cell or area, or a model trained for a specific range of WTRU velocity, etc.). The WTRU can perform measurements of one or more of the following: velocity, channel temporal variability (e.g., obtained via channel-related amplitudes used for TDCP reporting), and / or more channel measurements (e.g., Doppler, channel coherence time, and / or RSRP and SINR, etc., associated with CSI-RS in the observation window).

[0096] The WTRU can compute one or more metrics (e.g., NMSE and / or SGCS, for different predicted CSI instances) for prediction error estimation (e.g., for CSI prediction performance estimation). The WTRU can receive CSI-RS (e.g., an indication of the CSI-RS) associated with (e.g., different) predicted CSI instances within the prediction window. The WTRU can compute (e.g., corresponding) prediction error metrics between channel estimates and predicted CSI samples.

[0097] The WTRU can perform measurements of CSI compression reconstruction error. The WTRU can perform measurements related to the distribution of predicted CSI instances at the encoder input (e.g., equivalently at the output of the CSI prediction AI / ML block). The WTRU can compute the distribution of predicted CSI instances at the encoder input (e.g., predicted CSI instances within the prediction window size). The WTRU can perform measurements of statistical metrics used for OOD reporting (e.g., one or more of statistical distance, first-order statistics, second-order statistics, and / or energy-based scores). The WTRU can perform computations (e.g., the same computations as the input samples of the CSI prediction model). The WTRU can compute (e.g., determine) measurements based on (e.g., specific to) the configuration of the CSI compression AI / ML block (e.g., the encoder).

[0098] The WTRU can use one or more metrics (e.g., NMSE and / or SGCS) for CSI compression reconstruction error. The WTRU can determine the reconstruction error of the AI / ML models on both sides of CSI compression. The WTRU can apply a WTRU-side proxy decoder, which, for example, can allow for reducing the overhead required for performance monitoring based on ground fact labels that may be shared between the WTRU and the network (NW) / network element (NE). The WTRU can perform CSI compression on one or more predicted CSI instances based on a prediction window. For example, after quantization, the WTRU can apply a proxy decoder to reconstruct the predicted CSI samples. The WTRU can alternatively or additionally generate synthetic channel noise perturbations and apply them to the compressed samples before applying the proxy decoder. For example, if the WTRU-side proxy decoder is aligned with the NW / NE-side decoder, generating synthetic channel noise perturbations and applying them to the compressed samples can further approximate the true reconstruction performance of the NW / NE-side decoder.

[0099] WTRU can determine reconstruction error performance by using NW / NE side decoders. WTRU can use one or more proxy decoders, and / or can compute an estimate of the NW / NE side reconstruction error performance (e.g., by averaging the individual reconstruction errors provided by each proxy decoder).

[0100] The WTRU can perform relative error measurements. The WTRU can, for example, use different metrics to perform measurements of total / overall combined error (e.g., prediction and compression errors). The WTRU can use error-based metrics (e.g., NMSE and / or SGCS). Error-based metrics can be computed from reconstructed samples and / or ground reality (e.g., channel estimation applied as input to a CSI prediction AI / ML model). The WTRU can determine overall combined performance, for example, based on accuracy metrics and / or throughput (e.g., system throughput). The WTRU can combine the options presented herein and / or determine combined performance based on multiple metrics (e.g., options).

[0101] The WTRU can perform lifetime lifecycle management (LCM) measurements for combined prediction and compression models. The WTRU can use a proxy decoder to, for example, reconstruct CSI samples and / or estimate the combined error performance of the combined CSI prediction and compression system. The WTRU can request and / or receive assistance from the NW / NE (e.g., indications of that assistance, for example, to determine an estimate of the combined error). For example, when the WTRU does not apply the proxy decoder, the WTRU can request and / or receive assistance (e.g., indications of that assistance).

[0102] WTRU can determine the relative error of prediction and / or reconstruction, for example, based on calculated estimates of prediction and compression errors and / or the overall (e.g., total) combined error. WTRU can determine the relative prediction and compression errors, for example, by calculating error ratios. Error ratios can provide an indication of the error impact / contribution with respect to one or more (e.g., each) blocks (e.g., prediction and / or compression). If the error from the corresponding block is more dominant relative to the overall combined performance, the relative error associated with prediction and / or compression (e.g., for prediction and / or compression) is likely to be close to 1, and vice versa. WTRU can determine the relative error, for example, by calculating the difference between the corresponding error (e.g., prediction and / or compression error) and the overall combined error. WTRU can determine the relative performance and / or error dominance, for example, as a percentage of accuracy. For example, WTRU can (e.g., firstly) determine prediction accuracy, compression accuracy, and / or (e.g., then) overall combined accuracy. For example, by calculating the corresponding ratios and / or differences, the relative accuracy of each AI / ML block relative to the overall combined accuracy can be determined by WTRU.

[0103] WTRU can, for example, perform measurements for monitoring model performance of jointly trained CSI prediction and CSI compression models. WTRU can perform different measurements compared to the case where models are trained individually, such as when the AI / ML models used for CSI prediction and CSI compression are independent and operate sequentially. Additionally or alternatively, WTRU can perform different measurements compared to the case where models are trained individually when they are jointly trained (e.g., with the same training dataset and / or the same objective function).

[0104] In some examples, WTRU may determine OOD reporting without considering the input and / or output distribution of the CSI prediction model. If two AI / ML models are jointly trained using the same training dataset, statistical measures specific to the CSI prediction input can be used to determine OOD reporting. WTRU can determine the statistical distance between the input CSI samples (e.g., data points) within the observation window and the distribution of the training dataset. Similar to the case of separate training, WTRU can use different metrics. WTRU can use specific / specialized classifiers (e.g., non-AI / ML based or AI / ML based classifiers). The classifier can perform detection / classification of OOD input samples (e.g., K-channel estimates). For example, WTRU can (e.g., within the observation window) detect a minority of samples as OOD while others are within the distribution, depending on the observation window size.

[0105] There may be one or more triggers for WTRU measurements, such as for model performance monitoring. A WTRU may include at least two separate AI / ML models for CSI prediction compression, and / or may be triggered to perform measurements for performance monitoring. There may be WTRU-based triggers. For example, a WTRU may collect measurements and / or perform model monitoring based on its triggers. A WTRU may be triggered by an NW / NE. For example, a WTRU may receive an instruction to collect measurements from an NW / NE and / or may (e.g., based on that instruction) initiate performance monitoring of a system supporting CSI prediction compression.

[0106] WTRU may be subject to autonomous triggering. WTRU can perform measurements for performance monitoring based on one or more triggers. One or more triggers may include one or more of the following: time-based triggers, periodic time expiration, changes in WTRU-side performance metrics, inference latency, and / or changes in one or more applicable conditions.

[0107] WTRUs can be configured with time-based triggering. Time-based triggering can include periodic time instances used to collect performance monitoring measurements. Periodic time instances can be configured with a period and an offset. WTRUs can be triggered based on a configurable time period, such as since the last measurement used for performance monitoring.

[0108] WTRUs can be configured with periodic timer expirations. For example, WTRUs can be configured with timers (e.g., time values) either periodically or non-periodically. When a timer expires (e.g., the associated time period ends), the WTRU can be triggered to collect measurements for performance monitoring.

[0109] The WTRU can be configured with changes in WTRU-side performance metrics. For example, the WTRU can be triggered when it detects a performance degradation in one or more of the following: a CSI predictive AI / ML model, a CSI compressed AI / ML model, and / or a combined CSI predictive plus compressed model. Configured WTRU-side metrics can be used to evaluate performance. For example, monitoring the performance of one or more predictive and / or compressed instances can be configured (e.g., individually) to trigger measurements for WTRU-side model performance monitoring. For instance, the WTRU can be triggered based on the NACK frequency when it detects that the NACK frequency is at or above a received or pre-configured threshold.

[0110] WTRUs can be configured with inference latency. For example, a WTRU can be triggered by the inference latency of a combined CSI prediction and compression system. For instance, if the inference latency of a CSI instance exceeds a pre-configured threshold, a WTRU can be triggered to collect measurements for performance monitoring.

[0111] A WTRU can be configured with one or more applicable conditions. For example, a WTRU can be triggered based on speed. For instance, a WTRU can be triggered when its speed increases (e.g., rapidly increases and / or exceeds a threshold). Additionally or alternatively, speed or speed changes (e.g., decreases or increases) can trigger a WTRU. The speed or speed change can be relative to another device and / or NW / NE. For example, a WTRU can be triggered based on its speed or speed change relative to another device and / or NW / NE. The speed and / or speed change can be compared to a threshold.

[0112] For example, the WTRU can be triggered based on the time variability of the channel based on TDCP reports (when the channel correlation amplitude is below a pre-configured threshold). The WTRU can be triggered when one or more channel condition measurements (e.g., RSRP, Received Signal Strength Indicator (RSSI), Reference Signal Received Quality (RSRQ), Rank Indicator (RI), PMI, Channel Quality Indicator (CQI), SINR, Doppler Spread, Doppler Shift, Angle of Arrival (AoA), Angle of Departure (AoD), Delay Spread, Average Delay, and / or Location Coordinates) are above or below a threshold. The WTRU can calculate the difference between the current channel condition and the previous channel condition measurement, and / or be triggered if the difference is above a threshold.

[0113] The WTRU can be triggered by the NW / NE. NW / NE-based triggering can be based on one or more of the following: RRC configuration / reconfiguration, receipt of measurement requests, periodic / aperiodic triggering, environmental / scenario changes, configuration changes, mobility and / or location, and / or updates to AI / ML models. The WTRU can receive periodic and / or aperiodic RRC configuration / reconfiguration (e.g., indications of such RRC configuration / reconfiguration) for example, to collect measurements for performance monitoring. The WTRU can receive indications from the NW / NE to perform measurements and / or activate performance monitoring. Indications can be received via one or more of Downlink Control Information (DCI), Media Access Control (MAC) Control Element (CE), and / or Radio Resource Control (RRC) signaling. The WTRU can receive periodic and / or aperiodic indications (e.g., indications of such activation) for activating performance monitoring and / or collecting measurements.

[0114] The WTRU can receive indications when environmental changes (e.g., changes from LOS to NLOS, changes from NLOS to LOS, coverage changes, and / or environmental changes equal to or greater than a threshold) occur (e.g., rapid changes). The WTRU can be triggered by changes in configuration (e.g., the number of transmit (Tx) / receive (Rx) antennas, panels, TRP, and / or frequency bands, etc.) based on the NW / NE. The WTRU can be triggered by the NW / NE when it reports changes in velocity (e.g., absolute or relative velocity) and / or location (e.g., either can be absolute or relative to a reference location). The WTRU can be triggered by the NW / NE when one or more of the AI / ML models used for CSI prediction and CSI compression change (e.g., after model fine-tuning, after model transmission, after model switching, and / or after model activation).

[0115] There may be a WTRU process for combined CSI prediction and compression model monitoring. The WTRU can determine the type of CSI error condition. The WTRU can evaluate the condition of CSI error event detection, for example, while performing combined CSI prediction and compression. For example, the WTRU can begin (e.g., determine) evaluating the condition of CSI error event detection upon receiving a request from the NW / NE. This request may include a model performance monitoring request. CSI error events may indicate that the combined CSI prediction and compression blocks do not meet the target performance.

[0116] The WTRU can determine that a CSI error event has occurred when one or more of the following conditions are met: the CSI reconstruction error of the combined CSI prediction and compressed block is equal to or exceeds a threshold; an OOD event is detected for historical CSI samples at the input of the prediction block; the TDCP measured at the input of the prediction block is equal to or less than the configured TDCP threshold; the velocity estimated by the WTRU is outside the velocity range used for the training dataset; the measured channel properties (e.g., the channel's coherent BW and / or LOS / NLOS properties) are outside the range used for model training; and / or the WTRU receives an indication of a CSI error event detected by the NW / NE, etc.

[0117] The combined CSI reconstruction error of the CSI prediction and compression blocks can be referred to as the total CSI error. WTRU can determine that a CSI error event has occurred when the combined CSI reconstruction error of the CSI prediction and compression blocks equals or exceeds a threshold. For example, when the prediction window length is N=1 CSI samples, a CSI error event may occur when the CSI reconstruction error measured at the WTRU-side proxy decoder output equals or exceeds a configured threshold. Similarly, when the prediction window length is N>1 CSI samples and / or the predicted samples are compressed individually, a CSI error event may occur when the minimum CSI reconstruction error of the CSI instance in the prediction window (e.g., as measured at the WTRU-side proxy decoder output) exceeds a configured threshold. Finally, when the prediction window length is N>1 CSI samples and / or the predicted samples are compressed individually, a CSI error event may occur when the maximum CSI reconstruction error of the CSI instance in the prediction window (e.g., as measured at the WTRU-side proxy decoder output) equals or exceeds a configured threshold. The metrics used for reconstructing error measurement can be configured as one or more of NMSE, MSE, CS, and / or SGCS.

[0118] When an OOD event is detected, for example, at the input of the prediction block against historical CSI samples, the WTRU can determine that a CSI error event has occurred. The WTRU can perform a statistical measurement of the channel sample distribution at the input of the CSI prediction block. For example, when the actual channel distribution (e.g., at the CSI prediction input) differs significantly from the distribution of the training dataset (e.g., by a threshold), the WTRU can detect an OOD event at the input of the prediction block. For example, when such an event (e.g., an OOD event) is detected, the WTRU can determine that a CSI error event has occurred. For example, errors during the CSI prediction step can affect compression, potentially leading to a degraded performance of combined CSI prediction and / or compression.

[0119] When the TDCP measured at the CSI prediction block input is less than the configured TDCP threshold, the WTRU can determine that a CSI error event has occurred. Samples at the CSI prediction block input may decorrelate, which may degrade CSI prediction performance and / or (e.g., therefore) the performance of combined CSI prediction and / or compression.

[0120] When the speed estimated by WTRU is outside the speed range used for the training dataset (e.g., exceeding a threshold), WTRU can determine that a CSI error event has occurred. For example, the training dataset may contain data based on a low-speed mix of WTRU (e.g., 3 km / h to 30 km / h). For instance, when the model operates in vehicle scenarios at higher speeds (e.g., 60 to 120 km / h or higher), the CSI prediction model, the CSI compression model, or both may not generalize well, thus degrading the performance of combined CSI prediction and compression. WTRU can compare the estimated speed with the WTRU speeds included in the model and dataset configuration information, for example, to determine whether a CSI error event has occurred.

[0121] When measured channel properties (e.g., the channel's coherent BW and / or LOS / NLOS properties) are outside the range used for model training (e.g., a threshold), WTRU can determine that a CSI error event has occurred. The CSI prediction model and / or CSI compression model may not generalize well, thus (e.g., therefore) degrading the performance of combined CSI prediction and compression.

[0122] For example, when the WTRU determines that a CSI error event has occurred, the WTRU can assess the CSI error condition type. The CSI error condition type can identify whether a block (e.g., the ML model and / or processing step) has a dominant contribution to the combined CSI prediction and compression error. Dominant contribution identification can be based on a threshold. For example, when the CSI prediction step has a dominant contribution to the total combined CSI error, the CSI error condition type can be assessed as a CSI prediction-limited condition. When the CSI compression step has a dominant contribution to the total combined CSI error, the CSI error condition type can be assessed as a CSI compression-limited condition. When the CSI prediction and CSI compression steps have similar contributions to the total combined CSI error, the CSI error condition type can be assessed as a combined CSI prediction and compression error condition.

[0123] WTRU can assess a CSI error condition as a CSI prediction-constrained condition based on one or more of the following: the measured prediction error relative to the total combined CSI error (e.g., relative CSI prediction error) is equal to or exceeds a configured or received threshold; the relative prediction error is equal to or exceeds a relative compression error reaching a configured or received threshold; the minimum relative prediction error of a CSI instance in the prediction window is equal to or exceeds a relative compression error reaching a configured or received threshold; and the relative prediction error of the last CSI instance in the prediction window is equal to or exceeds a relative compression error reaching a configured or received threshold. The minimum relative prediction error of CSI instances in the prediction window is equal to or exceeds the maximum relative compression error of CSI instances in the prediction window, or the maximum relative compression error exceeds the configured or received threshold. The average prediction error of CSI instances in the prediction window is equal to or exceeds the total combined CSI error, or the prediction window length is equal to or greater than the measured channel coherence time. An OOD event is detected at the input of the prediction model for historical CSI samples, and / or the measured WTRU velocity is outside the velocity range of the training dataset used for the CSI prediction model.

[0124] For example, for a prediction window length of N=1 CSI instances, WTRU can assess the CSI error condition type as a CSI prediction-limited condition based on the measured prediction error relative to the total combined CSI error (e.g., relative CSI prediction error) being equal to or exceeding a configured or received threshold.

[0125] For example, for a prediction window length of N=1 CSI instances, WTRU can assess the CSI error condition type as a CSI prediction-limited condition based on the relative prediction error being equal to or exceeding a configured or received threshold.

[0126] For example, in the case of a prediction window length N > 1 CSI instances and individual compression, WTRU can assess the CSI error condition type as a CSI prediction-limited condition based on the minimum relative prediction error of the CSI instances in the prediction window being equal to the relative compression error (e.g., the average CSI compression reconstruction error of the instances in the prediction window) or exceeding the relative compression error reaching a configured or received threshold.

[0127] For example, for a prediction window length N > 1 CSI instances using individual compression, the WTRU can assess the CSI error condition type as a CSI prediction-limited condition based on the relative prediction error of the last CSI instance in the prediction window being equal to or exceeding the relative compression error (e.g., the average CSI compression reconstruction error across the instances in the prediction window) or a configured or received threshold. Similarly, for a prediction window length N > 1 CSI instances using individual compression, the WTRU can assess the CSI error condition type as a CSI prediction-limited condition based on the minimum relative prediction error of the CSI instances in the prediction window being equal to or exceeding the maximum relative compression error across the instances in the prediction window or a configured or received threshold. Finally, for a prediction window length N > 1 CSI instances using joint compression, the WTRU can assess the CSI error condition type as a CSI prediction-limited condition based on the average prediction error of the CSI instances in the prediction window being equal to or exceeding the total combined CSI error across the instances in the prediction window or a configured or received threshold.

[0128] WTRU can assess the CSI error condition type as a CSI prediction-limited condition based on a prediction window length equal to or greater than the measured channel coherence time. For example, for a period (e.g., time) d between configured consecutive prediction instances, WTRU can calculate the number n instances falling within the channel coherence time. fit for: For example, if the number N of prediction instances in the prediction window exceeds (or significantly exceeds) the number n of instances falling within the channel coherence time. fit Then WTRU can determine the CSI error condition type as CSI prediction limited.

[0129] WTRU can assess a CSI compression-limited condition based on one or more of the following: the measured compression error relative to the total combined CSI error (e.g., relative CSI compression error) is equal to or exceeds a configured or received threshold; the relative compression error is equal to or exceeds the relative prediction error reaching a configured or received threshold; the relative compression error of the CSI instance in the prediction window is equal to or exceeds the relative prediction error reaching a configured or received threshold; an OOD event is detected for the CSI sample at the input of the compressed block; and / or the measured WTRU velocity is outside the velocity range of the training dataset used for the CSI compression model.

[0130] For example, for a prediction window length of N=1 CSI instances, WTRU can assess the CSI error condition type as a CSI compression-limited condition based on the measured compression error (e.g., relative CSI compression error) relative to the total combined CSI error being equal to or exceeding a configured or received threshold. Similarly, for a prediction window length of N=1 CSI instances, WTRU can assess the CSI error condition type as a CSI compression-limited condition based on the relative compression error being equal to or exceeding a configured or received threshold for the relative prediction error.

[0131] For example, in the case of a prediction window length N > 1 CSI instances and individual compression, WTRU can assess the CSI error condition as a CSI compression-constrained condition based on whether the relative compression error of the CSI instances in the prediction window equals the relative prediction error or exceeds a configured or received threshold. The relative compression error can be calculated as the minimum and / or average relative compression of the CSI instances in the prediction window. The relative prediction error can be calculated for the last CSI instance in the prediction window.

[0132] WTRU can evaluate a CSI error condition type as a combined CSI prediction and compression error condition based on one or more of the following: the absolute difference between the relative compression error and the relative prediction error is equal to or less than a configured or received threshold, and / or the absolute difference between the relative compression error and the relative prediction error for some or every CSI instance in the prediction window is equal to or less than a configured threshold. For example, for a prediction window length of N=1 CSI instances, WTRU can evaluate a CSI error condition type as a combined CSI prediction and compression error condition based on the absolute difference between the relative compression error and the relative prediction error being less than a configured threshold. For example, for a prediction window length of N>1 CSI instances and using individual compression, WTRU can evaluate a CSI error condition type as a combined CSI prediction and compression error condition based on the absolute difference between the relative compression error and the relative prediction error for some or every CSI instance in the prediction window being equal to or less than a configured or received threshold.

[0133] The WTRU can operate under prediction-constrained conditions. For example, a WTRU performing combined CSI prediction and compression can determine that a CSI error event has occurred. The WTRU can determine (e.g., evaluate) the type of CSI error condition as CSI prediction-constrained. For example, if the CSI error condition is CSI prediction-constrained, the WTRU can perform actions. These actions can include, for example, determining (e.g., preferred) parameters of the prediction block based on measured channel conditions and / or (e.g., current) CSI prediction configuration. For example, the WTRU can determine to perform one or more of the following: reduce the number of prediction instances, reduce the distance between prediction instances (e.g., time), increase the number of samples in the observation window, and / or reduce the distance between historical instances in the observation window (e.g., time).

[0134] WTRU can determine the number of prediction instances to reduce. For example, when the measured channel coherence time is equal to or less than the prediction window length, WTRU can determine the number (e.g., the preferred number) of CSI instances in the prediction window. d can represent the period (e.g., time) between consecutive predicted CSI instances. The number of predicted CSI instances (e.g., the preferred number) n fit It can be equal to or less than the configured value N; for example, this can actually reduce the forecast window. For example, for the current CSI reporting timing, WTRU can alternatively or additionally report the forecast window (e.g., the preferred forecast window) length n. fit WTRU can include information in its feedback report. This information may include or indicate an index of n. fit The prediction instance is repeated, for example, until the end of the prediction window. WTRU can indicate this repetition to NW / NE.

[0135] For example, when NW / NE configures WTRU to use n for future (e.g., the next) reporting timing. fit When predicting (e.g., preferred prediction) windows for n instances, WTRU can perform preprocessing at the input of the CSI compressed block (e.g., from n instances). fit (Zero padding for N CSI instances). For example, when NW / NE is configured to use WTRU for future (e.g., the next) reporting timing, n... fit When the preferred prediction window for an instance is reached, the WTRU can perform model switching for the CSI compressed block and / or can send a model switching instruction to the NW / NE.

[0136] WTRU can determine how to reduce the distance (e.g., time) between predicted instances. For example, when the measured channel coherence time is equal to or less than the prediction window length, WTRU can reduce the (e.g., preferred) distance d between consecutive predicted CSI instances. pref Determined as: N can represent the number of prediction instances configured in the prediction window.

[0137] WTRU can determine the number of samples to increase within the observation window. For example, when the observation window is shorter than the channel coherence time (e.g., much shorter), WTRU can determine the number of CSI instances to increase within the observation window. WTRU can then determine the number (e.g., the preferred number) of CSI instances within the observation window, K. pref Determined as: ). m can represent the time period between consecutive CSI instances in the observation window (e.g., the time period corresponds to a CSI-RS period). For example, WTRU can switch CSI prediction models (e.g., to support a larger number of samples in the observation window).

[0138] WTRU can reduce the distance (e.g., time) between historical instances within an observation window. For example, when the distance is equal to or longer than the channel coherence time (e.g., much longer / exceeding a threshold), WTRU can determine to reduce the distance *m* between CSI instances within the observation window. WTRU can then reduce the distance (e.g., preferred distance) *m* between CSI instances within the observation window. pref Determined as: m pref =K, where K is, for example, the number of CSI instances in the observation window.

[0139] The WTRU can determine to fine-tune the CSI prediction model. The WTRU can determine to perform a model switch. For example, if the WTRU detects an OOD event, it can determine to perform a model switch (e.g., a CSI prediction model switch). The WTRU can determine to fall back to a non-AI / ML-based prediction. For example, when no AI / ML CSI prediction model suitable for the current channel conditions is available (e.g., locally available on the WTRU side and / or transmit / deliverable from the NW / NE), the WTRU can determine to fall back to a non-AI / ML-based prediction (e.g., Rel-18 eType-II for predicting PMI). The WTRU can determine to fall back to reporting the current CSI. For example, in situations with very high speeds (e.g., very short channel coherence times), the WTRU can determine to fall back to reporting the current CSI measurement.

[0140] WTRU behavior can operate under combined CSI prediction and compression error conditions. For example, a WTRU configured with individual CSI prediction and CSI compression can calculate the CSI prediction error for one or more predicted CSIs within a prediction window. WTRU can calculate the CSI prediction error for each prediction index based on the historical CSI prediction error for each prediction index. For example, WTRU can calculate the historical CSI prediction error for each prediction index as a measure (e.g., SGCS, NMSE, etc.) of the difference between the actual and predicted CSIs for one or more (e.g., each) of the predicted CSI indices within the prediction window.

[0141] WTRU can, for example, determine the compression ratio of one or more (e.g., each) of the CSI prediction indices based on the calculated CSI prediction error for each CSI prediction index in the prediction window. WTRU can also determine the compression ratio of each prediction index based, for example, on one or more of the analytical model, lookup table, and / or models with lower compression ratios.

[0142] WTRU can determine the compression ratio based on an analytical model. WTRU can be configured with an analytical model to calculate the compression ratio R for a given CSI prediction error E, for example, R = f(E). WTRU can use the analytical model to calculate the compression ratio Ri for a CSI prediction index i with error Ei.

[0143] WTRU can determine the compression ratio based on a lookup table. WTRU can be configured with a lookup table to determine the compression ratio for each CSI prediction instance. For example, the lookup table can include one or more rows with a threshold for CSI prediction error and / or the corresponding compression ratio.

[0144] WTRU can select / choose models with lower compression ratios. For higher prediction indices, WTRU can determine the compression ratio based on models with lower compression ratios (e.g., less compression). WTRU can be configured with CSI compression models with different compression ratios for one or more (e.g., each) of the CSI prediction indices. For example, the compression ratio of one or more (e.g., each) models used to compress CSI prediction instances can decrease as the prediction index increases (e.g., less compression). For example, the compression ratio R of the CSI prediction instance at time i... i The compression ratio R of the CSI prediction instance at time i+1 can be higher than that of the previous time. i+1 .

[0145] WTRU can compress one or more (e.g., each) predicted CSIs based on a compression ratio determined for the corresponding predicted index. For example, WTRU can use different AI / ML-based CSI compression models to compress one or more (e.g., each) of the predicted CSI instances.

[0146] For example, a WTRU configured with multiple CSI prediction instances for joint compression can determine the block size (e.g., subset, group, etc.) used for joint compression of one or more CSI prediction instances. For instance, the WTRU can determine to compress a first subset of CSI prediction instances at a time using a first CSI compression model, and / or to compress a second subset of CSI prediction instances at a time using a second CSI compression model. The WTRU can determine one or more subsets of CSI prediction instances based, for example, on one or more of the following: a measurement of CSI prediction error, payload size, and / or overall compression performance.

[0147] WTRU can determine one or more subsets of CSI prediction instances based on a metric of CSI prediction error. WTRU can group CSI prediction instances (e.g., divide them into subsets) based on the average prediction error of the CSI prediction instances. WTRU can determine to compress a first group of CSI predictions (e.g., indices with an average CSI prediction error below a threshold) using a first compression ratio, and / or compress a second group of CSI predictions (e.g., indices with an average CSI prediction error equal to or above a threshold) using a second compression ratio.

[0148] The WTRU can determine one or more subsets of CSI prediction instances based on payload size. The WTRU can be configured with (e.g., payload size constraints to be satisfied). The WTRU can determine to compress a first set of predicted CSI instances using a first compression ratio and / or compress a second set of predicted CSI instances using a second compression ratio. The total payload size of the two jointly compressed CSI prediction instances can satisfy the total payload size constraint. The WTRU can determine the first compression ratio and / or the second compression ratio based on the payload size constraint. The first set of predicted CSI instances can correspond to a first portion of the CSI prediction instances, where, for example, the CSI prediction error is lower compared to a second portion of the CSI prediction instances.

[0149] The WTRU can determine one or more subsets of CSI prediction instances based on overall compression performance. The WTRU can be configured with overall compression performance, for example, measured by SGCS and / or NMSE. For example, the WTRU can determine a compression ratio to compress all CSI prediction instances at once, thereby satisfying compression performance. For example, the WTRU can determine a compression ratio based on the average CSI prediction error and / or the maximum CSI prediction error of the prediction window. The WTRU can compress predicted CSI. The WTRU can compress predicted CSI based on one or more determined groups (e.g., subsets), for example, according to one or more compression ratios determined using one or more CSI compression models.

[0150] The WTRU can operate under compression-constrained conditions. For example, during performance monitoring of a combined CSI predictive compression system, the WTRU can determine (e.g., identify) that its operating mode is under compression-constrained conditions. For example, when the WTRU is configured or pre-configured with a counter / frequency threshold, the WTRU can increment a counter and / or frequency specific to the detected compression-constrained condition. For example, if the updated counter / frequency exceeds the threshold, the WTRU can perform a model switch to another CSI compressed AI / ML model.

[0151] For example, after one or more of the following: CSI compressed model activation, model download, and / or model fine-tuning, the WTRU can detect that its operating mode is compression-limited. The WTRU can, for example, request / recommend a model switch to the NW / NE without incrementing the compression-limited counter / frequency. The WTRU can (e.g., previously) operate with (e.g., a previous) AI / ML model used for CSI compression. The WTRU can determine (e.g., begin) to increment the counter until a limited (e.g., lower) second threshold is reached. For example, if the frequency of the new model is higher than the frequency of the previously operated CSI compressed AI / ML model, the WTRU can compare the compression-limited frequencies of the two AI / ML models and / or can request / receive / recommend / determine a model switch to the previous model (e.g., the model switch indication). Additionally or alternatively, if the model was recently activated (e.g., based on time constraints and / or based on the number of inference instances) and / or if the WTRU is using a previous AI / ML model operated for CSI compression, the WTRU can determine (e.g., begin) to increment the counter until a limited (e.g., lower) second threshold is reached.

[0152] The WTRU can detect when the frequency of compression-limited conditions occurring in the current AI / ML model and / or (e.g., all) supported AI / ML models for CSI compression (e.g., AI / ML models previously activated in a combined CSI prediction and compression system) is equal to or exceeds a configured threshold. The WTRU can request / recommend a fallback to the default CSI reporting configuration, which, for example, may include CSIs corresponding to the most recently received CSI-RS (e.g., an indication of the most recent CSI-RS).

[0153] WTRU can indicate combined CSI prediction and compression model monitoring metrics and / or parameters. WTRU can report, for example, to a network one or more of the following: CSI prediction and compression model monitoring, metrics, and / or parameters. WTRU can be configured to report one or more of the following indicators periodically, aperiodically, and / or semi-persistently: CSI, CSI prediction and compression model monitoring, metrics, and / or parameters.

[0154] Report instances may include one or more of the following: predicted CSI, compressed CSI, predicted and compressed CSI, (e.g., determined) CSI error condition type, and / or an identifier and / or set of feedback reports applicable to the determined CSI error condition type. Report instances may include or indicate predicted CSI, such as the output of a CSI prediction module (e.g., CSI prediction AI / ML model output). The input to the CSI prediction module may be one or more channel measurements. Report instances may include or indicate compressed CSI, such as the output of a CSI compression module (e.g., CSI compression AI / ML model output). The input to the CSI compression module may be one or more channel measurements.

[0155] Report instances may include or indicate predicted and compressed CSI. The WTRU may input channel measurements into the CSI prediction module, and / or may use the output of the CSI prediction module as input to the CSI compression module. The WTRU may report the output of the compression module. For example, the WTRU may input channel measurements into a combined CSI prediction / compression module, and / or the WTRU may report the output of the combined CSI prediction / compression module. The WTRU may be configured with the order (e.g., compression first / predict later, and / or prediction first / compression later) and / or type (e.g., separate prediction and compression modules or combined prediction and compression modules) of the joint prediction and compression CSI reporting.

[0156] Report instances may include or indicate the determined CSI error condition type. For example, a WTRU may indicate that the primary source of error in combined prediction and compression is prediction error. A WTRU may indicate that the primary source of error in combined prediction and compression is compression error. A WTRU may indicate that there is no primary source of error in combined prediction and compression (e.g., prediction error and compression error contribute similarly to the total error and / or are within a threshold). For example, a WTRU may determine that one or more of the prediction error, compression error, and / or combined prediction and compression error are above or below a threshold. The threshold may be received, configurable, or pre-configured. Report instances may include, for example, identifiers and / or sets of feedback reports applicable to the determined CSI error condition type. The threshold herein may refer to an indication of a threshold. For example, a WTRU may receive an indication of a threshold (e.g., an index).

[0157] WTRU may, for example, determine whether to report feedback on one or more of the following based on one or more of the determined CSI error condition type, prediction error value, compression error value, combined prediction and compression error value, and / or comparison of one or more error values ​​with a threshold: prediction only, compression only, combined prediction and compression of CSI, and / or no compression and no prediction of CSI.

[0158] WTRU can determine and / or update one or more of the following: CSI prediction module parameters, CSI compression module parameters, and / or combined CSI prediction / compression module parameters. The determination and / or updating of parameters can be based on one or more of the following: the determined CSI error condition type, the prediction error value, the compression error value, the combined prediction and compression error value, and / or a comparison of one or more error values ​​with a threshold.

[0159] The WTRU may report the determination and / or updating and / or requesting updates of one or more parameters (e.g., parameters for one or more of the CSI prediction module, CSI compression module, and / or combined CSI prediction / compression module) in, for example, a CSI report. The WTRU may indicate the updated set of parameters and / or the updated parameter values ​​in the report. The WTRU may indicate the set of parameters, the request to update, and / or the request for updated parameter values ​​in the report. For example, parameters may include one or more of the following: prediction window size, number of prediction outputs, RS parameters, compression ratio, compression type, and / or input and / or output domains of the module. The prediction window size may include the window size on which measurements are performed to achieve the prediction; for example, it may include the window size on which prediction measurements can be determined and / or reported. RS parameters may include one or more of the following: density, period, type, and / or resource allocation. The compression ratio may include the maximum and / or average payload of the compression module's output. The compression type may include compression for a single time instance and / or joint compression for multiple time instances. The inputs or outputs of at least one of the CSI prediction module and / or CSI compression module may include one or more of the following forms and / or one or more of the following forms: full channel matrix domain, eigenvector domain, CQI / PMI / RI format and / or AI / ML latent space domain.

[0160] WTRU can identify and / or report the source or cause of error for at least one module. For example, based on one or more measurements or performance metrics, WTRU can identify and / or report the source of error for at least one module, including one or more of the following: an undertrained AI / ML model, inappropriate AI / ML model selection and / or configuration, and / or incompatibility between the combination of a CSI predictive AI / ML model and a CSI compressed AI / ML model.

[0161] WTRU can identify and / or report the source of error for at least one module, including an undertrained AI / ML model. An undertrained AI / ML model may include an AI / ML model for one or more of the following that may not have been trained for the current scene / configuration: a CSI prediction module, a CSI compression module, and / or a combined prediction / compression module.

[0162] WTRU can identify and / or report the source of error for at least one module, including and / or indicating inappropriate AI / ML model selection and / or configuration. For example, inappropriate AI / ML model selection and / or configuration may include AI / ML models used for one or more of the following that may not be suitable for the current scenario / configuration: CSI prediction module, CSI compression module, and / or combined prediction / compression module.

[0163] WTRU can identify and / or report sources of error for at least one module, including and / or indicate that a combination of CSI predictive AI / ML models and CSI compressed AI / ML models may be incompatible. For example, one or more models (each on its own) may have acceptable performance, but the combined performance may be unacceptable. Performance acceptability can be determined by identifying the type of CSI contribution error and / or at least one error value (e.g., prediction, compression, and / or combination). For example, error values ​​can be compared to thresholds (e.g., to determine performance acceptability).

[0164] For AI / ML model management, such as when triggered by a determined CSI contribution error type or error value, the WTRU may report the results of AI / ML model training or retraining. Additionally or alternatively, for AI / ML model management, such as when triggered by a determined CSI contribution error type and / or error value, the WTRU may report requests for AI / ML model training or retraining. For AI / ML model management, such as when triggered by a determined CSI contribution error type and / or error value, the WTRU may report the results of AI / ML model validation. Additionally or alternatively, for AI / ML model management, such as when triggered by or based on a determined CSI contribution error type and / or error value, the WTRU may report requests for AI / ML model validation. For AI / ML model management, such as when triggered by or based on a determined CSI contribution error type and / or error value, the WTRU may report the results of AI / ML model selection. Additionally or alternatively, for AI / ML model management, such as due to or based on the determined CSI contribution error type and / or error value, WTRU may report a request for selection of AI / ML models.

[0165] WTRU can determine and / or report CSI reporting rules. WTRU can determine the type of CSI to be reported based on one or more of the following: the determined CSI error condition type, the predicted error value, the compressed error value, the combined predicted and compressed error values, a comparison of one or more error values ​​with a threshold, whether the feedback report is periodic / non-periodic / semi-persistent, the timing of the feedback report, the configuration and / or indications in the CSI feedback request, etc.

[0166] The CSI type of the report may be associated with one or more of the domains described herein, such as the full channel matrix domain, EV domain, CQI / PMI / RI format, and / or AI / ML latent space domain. The CSI type may include one or more of the following: regular CSI, CSI report of a single set of CSI values, CSI report of multiple sets of CSI values, and / or CSI report of one or more sets of CSI values.

[0167] CSI types can include regular CSI. For example, regular CSI can be unpredictable and uncompressed. CSI types can also include CSI reports that are a single set of CSI values ​​associated with a single time instance reference. For example, a CSI report can include, for instance, a single set of predicted and / or compressed CSI values ​​associated with a single time instance reference.

[0168] A CSI report can include multiple sets of CSI values. Each set can be associated with a single (e.g., unique) time instance reference. For example, a CSI report can include multiple sets of predicted and / or compressed CSI values. Each set of predicted and / or compressed CSI values ​​can be (e.g., uniquely) associated with a single time instance reference. Each set of predicted CSI values ​​associated with a single time instance reference can be compressed individually.

[0169] A CSI type may include a CSI report containing one or more sets of CSI values. Each set of CSI values ​​may be associated with multiple time instance references. For example, a CSI report may include one or more sets of predicted and / or compressed CSI values, each of which may be associated with multiple time instance references, for example, disjointly. Multiple sets of predicted CSI values ​​(e.g., each associated with a single (possibly unique) time instance) may be jointly compressed.

[0170] The WTRU may determine one or more time instance references for CSI feedback reports based on one or more of the following: the timing of the feedback report, whether the feedback report is periodic / non-periodic / semi-persistent, the type of feedback report, the determined type of CSI error condition, the predicted error value, the compressed error value, the combined predicted and compressed error value, a comparison of one or more error values ​​with a threshold, indications and / or configurations in the CSI feedback request, etc. For example, for periodic reports, the WTRU may include and / or indicate CSI feedback associated with a time instance reference associated with a specific periodic feedback report instance. For example, for non-periodic reports, the WTRU may include CSI feedback associated with all available and / or unreported time instances.

[0171] The WTRU can be configured to report CSI periodically, non-periodically, and / or semi-persistently. The WTRU can determine the timing of at least one report and / or the period of multiple reports based on one or more of the following: the type of CSI report, the determined type of CSI error condition, the predicted error value, the compressed error value, the combined predicted and compressed error values, and / or a comparison of one or more error values ​​with a threshold.

[0172] The WTRU can determine and / or report CSI error condition types. The WTRU can perform combined CSI prediction and compression. The WTRU can determine and / or report CSI error condition types, for example, based on channel measurements and / or configured performance monitoring thresholds.

[0173] Figure 4 This is another example of CSI process 400. At 402, a WTRU, for example configured to perform combined CSI prediction and compression, may receive configuration information (e.g., an indication of that configuration information). The configuration information can be used to combine CSI prediction and compression reports. The configuration information may include and / or indicate one or more of the following: CSI-RS configuration, CSI feedback reporting configuration, CSI prediction configuration (e.g., observation window and / or prediction window), CSI compression configuration, and / or performance monitoring thresholds.

[0174] At 404, the WTRU can be triggered to perform measurements and / or measurements for (e.g., WTRU-side model) combined CSI prediction and compression performance monitoring. Measurements may include one or more of the following: channel measurements and / or statistics, prediction estimation error metrics, compression reconstruction error metrics, total combined CSI error and / or relative prediction error of one or more CSI instances in the prediction window and / or relative compression error relative to the total combined CSI error. Channel measurements and / or statistics may include one or more of, for example, TDCP, RSRP, and / or SINR associated with CSI-RS in the observation window. Additionally or alternatively, channel measurements and / or statistics may include (e.g., for OOD detection) one or more of Doppler, channel coherence time, first / second-order statistics, and distribution metrics.

[0175] At point 406, the WTRU can determine the CSI error condition type based on measurements and / or the current combined CSI prediction and compression configuration (e.g., for combined CSI prediction and compression performance monitoring). The CSI error condition type can be determined as one or more of a CSI prediction-limited condition, a CSI compression-limited condition, and / or a combined CSI prediction and compression-limited condition. A CSI prediction-limited condition can be determined when the relative prediction error equals the relative compression error or exceeds a configured or received threshold for the relative compression error. A CSI compression-limited condition can be determined when the relative compression error equals the relative compression error or exceeds a configured or received threshold for the relative prediction error. A combined CSI prediction and compression-limited condition (e.g., combined CSI prediction and compression-limited condition) can be determined when the difference between the relative prediction error and the relative compression error is equal to or less than a configured threshold. The WTRU can report the predicted and compressed CSI and / or the determined CSI error condition type to, for example, the network / NE.

[0176] At 408, when operating under CSI prediction-constrained conditions, the WTRU can determine (e.g., preferably) CSI prediction parameters. The WTRU can detect WTRU operation under CSI prediction-constrained conditions. The WTRU can determine and / or report (e.g., preferably) parameters of the WTRU-side CSI prediction (e.g., prediction blocks), for example, based on channel measurements and / or configured performance monitoring thresholds. For example, a WTRU configured to perform combined CSI prediction and compression can receive configuration information (e.g., an indication of the configuration information). The configuration information can be used for combined CSI prediction and compression reporting. The configuration information may include and / or indicate one or more of the following: CSI-RS configuration, CSI feedback reporting configuration, CSI prediction configuration (e.g., observation window and / or prediction window), CSI compression configuration, and / or performance monitoring thresholds. At 410, the WTRU can report a message to, for example, the network / NE. This message may include one or more of the following: predicted CSI, compressed CSI, and / or determined CSI error condition type.

[0177] The WTRU can be triggered to perform measurements for combined CSI prediction and compression performance monitoring (e.g., for WTRU-side models). Measurements may include and / or indicate one or more of the following: channel measurements and / or statistics, prediction estimation error metrics, compression reconstruction error metrics, total combined CSI error, and / or relative prediction error of one or more CSI instances within the prediction window, and / or relative compression error relative to the total combined CSI error. Channel measurements and / or statistics may include and / or indicate one or more of, for example, TDCP, RSRP, and / or SINR associated with CSI-RS within the observation window. Additionally or alternatively, channel measurements and / or statistics may include one or more of, for example, Doppler, channel coherence time, first / second-order statistics, or distribution metrics (e.g., for OOD detection).

[0178] The WTRU can determine the CSI error condition type based on the measurements used for combined CSI prediction and compression performance monitoring and / or the current combined CSI prediction and compression configuration. The CSI error condition type can be determined as one or more of a CSI prediction-limited condition, a CSI compression-limited condition, and / or a combined CSI prediction and compression-limited condition. For example, the WTRU can determine the CSI error condition as a CSI prediction-limited condition when the relative prediction error is equal to or exceeds a configured or received threshold for the relative compression error. Similarly, the WTRU can determine the CSI error condition as a CSI compression-limited condition when the relative compression error is equal to or exceeds a configured or received threshold. Finally, the WTRU can determine the CSI error condition as a combined condition (e.g., a combined CSI prediction and compression-limited condition) when the difference between the relative prediction error and the relative compression error is equal to or less than a configured or received threshold.

[0179] The WTRU can determine preferred parameters for the prediction block, for example, based on measurements used for combining CSI prediction and compression performance monitoring and / or the current CSI prediction configuration. For instance, when the WTRU determines (e.g., detects) that it is operating under CSI prediction-constrained conditions, it can determine preferred parameters for the prediction block. For example, when the prediction window length is equal to or exceeds the channel coherence time, the WTRU can determine to reduce the prediction window (e.g., the number of prediction instances or the distance between prediction instances). For example, when the observation window length is equal to or shorter than the channel coherence time, the WTRU can determine to increase the observation window (e.g., the number of instances within the observation window).

[0180] WTRU may report one or more of the following parameters to, for example, the network / NE: predicted and / or compressed CSI, the determined CSI error condition type (e.g., CSI prediction limited), and / or the predicted block (e.g., preferred).

[0181] Figure 5A Example 500 shows the prediction error relative to the index. Figure 5BExample table 505 is shown. The WTRU can, for example, determine and / or report (e.g., preferably) the WTRU-side CSI compression configuration based on the index of the prediction instances in the measurement and / or prediction window. For example, a WTRU configured to perform combined CSI prediction and compression can receive a compression configuration (e.g., an indication of the compression configuration) for each CSI prediction instance in the prediction window. The WTRU can determine the CSI error condition type based on channel conditions and / or the prediction and compression configuration. The WTRU can determine recommended changes to the compression configuration (e.g., model type, model ID, model parameters, or compression rate) for one or more (e.g., each) CSI prediction instances in the prediction window based on one or more of the index of one or more prediction instances in the prediction window, the measured CSI prediction error of the instance, and / or the configured performance threshold.

[0182] Tables (e.g.) Figure 5B The table (in the table) may include indexes, prediction losses, and / or compression ratios. Each index may be associated with a prediction and / or compression ratio. Each index may be an index within a prediction window. Additionally or alternatively, each prediction loss may be relative to the start of the window (e.g., the start of the prediction window). One or more of the prediction losses may be thresholds (e.g., as described herein). The WTRU may determine and / or receive indications of one or more of the indices, such as those in the table.

[0183] The WTRU can determine the compression configuration, for example, based on an index within the prediction window. The WTRU can determine the compression configuration when operating under a combined CSI error condition. (For example, if it detects that it is operating under a combined CSI prediction and compression error condition type) The WTRU can determine and / or report (e.g., preferably) the WTRU-side CSI compression configuration, for example, based on the index of the prediction instance in the measurement and / or prediction window.

[0184] Figure 6 Another example of CSI procedure 600 is shown. At 602, the WTRU may receive configuration information (e.g., an indication of the configuration information) (e.g., a compression rate for one or more (e.g., each) CSI prediction instances in the prediction window). For example, a WTRU configured to perform combined CSI prediction and compression may receive a compression configuration for each CSI prediction instance in the prediction window (e.g., an indication of the compression configuration). At 604, the WTRU may determine the CSI error condition type based on channel conditions, measurements, and / or (e.g., the current) prediction and compression configurations.

[0185] At point 606, for example, if the WTRU detects that it is operating under a combined CSI prediction and compression error condition type, the WTRU may determine a recommended change to the compression configuration (e.g., model type, model ID, model parameters, or compression ratio) for one or more (e.g., each) CSI prediction instances in the prediction window. Additionally or alternatively, the WTRU may determine the recommended change to the compression configuration (e.g., model type, model ID, model parameters, or compression ratio) for one or more (e.g., each) CSI prediction instances in the prediction window based on one or more of the index of one or more prediction instances in the prediction window, the measured CSI prediction error of the instance, and / or the configured performance threshold.

[0186] At 608, the WTRU may send an indication of one or more of the following: the determined CSI error condition type (e.g., combination), a recommended change to the compression configuration, and / or the corresponding index of one or more predicted instances. At 610, the WTRU may report predicted and / or compressed CSI (e.g., an indication of predicted and / or compressed CSI).

[0187] Examples of abbreviations and / or acronyms in this article are shown below.

[0188] LCM Lifecycle Management OOD distribution outside TDCP Time Domain Channel Attributes CE control elements CQI Channel Quality Indicator CSI Channel State Information DCI Downlink Control Information DL downlink LTE Long Term Evolution (e.g., from 3GPP LTE R8 and later versions) NACK (Negative Acknowledgment) MIMO (Multiple Input Multiple Output) NR New Radio OFDM (Orthogonal Frequency Division Multiplexing) RRC (Radio Resource Control) RS reference signal RSRP reference signal received power RSSI Received Signal Strength Indicator TRP Transmit / Receive Point UL uplink URLLC offers highly reliable and low-latency communication. WLAN (Wireless Local Area Network) and related technologies (IEEE 802.xx domain).

Claims

1. A method implemented by a wireless transmit / receive unit (WTRU), the method comprising: Receive configuration information for Channel State Information (CSI) prediction and compression; Measure the CSI prediction error and CSI compression error to obtain the total combined CSI error; The CSI error condition type is determined based on the total combined CSI error, wherein the CSI error condition type includes at least one of the CSI prediction-limited condition type, the CSI compression-limited condition type, and the combined CSI prediction and compression-limited condition type. as well as A message is sent to the network, the message including an indication of the determined CSI error condition type.

2. The method of claim 1, further comprising determining a relative CSI error based on at least one of the total combined CSI error, the CSI prediction error, and the CSI compression error.

3. The method of claim 2, wherein the relative CSI error includes the relative CSI prediction error, and wherein determining the CSI error condition type includes: When the relative CSI prediction error is equal to or exceeds a threshold associated with the total combined CSI error, the CSI error type is determined to be the CSI prediction-limited condition type.

4. The method of claim 3, wherein the method further comprises determining a time window associated with the CSI prediction.

5. The method of claim 2, wherein the relative CSI error includes a relative CSI compression error, and wherein determining the CSI error condition type includes: When the relative CSI compression error exceeds a threshold associated with the total combined CSI error, the CSI error type is determined to be the CSI compression-limited condition type.

6. The method of claim 1, wherein determining the CSI error condition type comprises: When the difference between the CSI prediction error and the CSI compression error is equal to or less than a threshold, the CSI error type is determined to be the combined CSI prediction and compression restricted condition type.

7. The method of claim 6, wherein the method further comprises determining a CSI compression configuration that includes or indicates a CSI instance threshold.

8. The method of claim 1, wherein the message further includes the determined CSI compression configuration and associated index.

9. The method of claim 1, wherein the message further includes an indication of the measured CSI prediction error and CSI compression error.

10. The method of claim 1, further comprising determining parameters associated with CSI prediction based on the measured CSI prediction error and CSI compression error, and the CSI prediction configuration.

11. A wireless transmit / receive unit (WTRU), comprising: The processor is configured as follows: Receive configuration information for Channel State Information (CSI) prediction and compression; Measure the CSI prediction error and CSI compression error to obtain the total combined CSI error; The CSI error condition type is determined based on the total combined CSI error, wherein the CSI error condition type includes at least one of the CSI prediction-limited condition type, the CSI compression-limited condition type, and the combined CSI prediction and compression-limited condition type. as well as A message is sent to the network, the message including an indication of the determined CSI error condition type.

12. The WTRU of claim 11, wherein the processor is further configured to determine a relative CSI error based on at least one of the total combined CSI error, the CSI prediction error, and the CSI compression error.

13. The WTRU of claim 12, wherein the relative CSI error includes a relative CSI prediction error, and wherein the processor is configured to determine the CSI error condition type by: The processor is configured to determine the CSI error type as the CSI prediction-limited condition type when the relative CSI prediction error is equal to or exceeds a threshold associated with the total combined CSI error.

14. The WTRU of claim 13, wherein the processor is further configured to determine a time window associated with the CSI prediction.

15. The WTRU of claim 12, wherein the relative CSI error includes a relative CSI compression error, and wherein the processor is configured to determine the CSI error condition type by: The processor is configured to determine the CSI error type as the CSI compression-limited condition type when the relative CSI compression error exceeds a threshold associated with the total combined CSI error.

16. The WTRU of claim 11, wherein the processor is configured to determine the CSI error condition type by: The processor is configured to determine the CSI error type as the combined CSI prediction and compression restricted condition type when the difference between the CSI prediction error and the CSI compression error is equal to or less than a threshold.

17. The WTRU of claim 16, wherein the processor is further configured to determine a CSI compression configuration that includes or indicates a CSI instance threshold.

18. The WTRU of claim 11, wherein the message further includes the determined CSI compression configuration and associated index.

19. The WTRU of claim 11, wherein the message further includes indications of measured CSI prediction error and CSI compression error.

20. The WTRU of claim 11, wherein the processor is further configured to determine parameters associated with CSI prediction based on measured CSI prediction error and CSI compression error, and CSI prediction configuration.

Images